Bernard Bachrach - Merovingian Military Organization

Merovingian
Military Organization
481-751
This page intentionally left blank
Merovingian
481-751
BERNARD S. BACHRACH
UNIVERSITY OF MINNESOTA PRESS

Minneapolis
© Copyright 1972 by the University of Minnesota.
All rights reserved.
Printed in the United States of America
at the University of Minnesota Printing Department, Minneapolis.
Published in the United Kingdom and India by the Oxford University
Press, London and Delhi, and in Canada
by the Copp Clark Publishing Co. Limited, Toronto
Library of Congress Catalog Card Number: 70-187164

ISBN 0-8166-0621-8
to
MY PARENTS
Preface
T his monograph is a study of the military institutions which

flourished in Gaul from the accession of Clovis in 481 to the
deposition of Childeric III in 751. Previous work on this subject
has been limited to specialized articles and sections of general
works. These efforts have been based upon a small selection of
the available evidence and have resulted in a distorted under-
standing of Merovingian military organization. In the present
study the use of all the significant sources has made it possible
to ascertain that there was great diversity as well as substantial
change in the military institutions of Merovingian Gaul.
A secondary aim of this book has been to investigate the rela-
tion of the organization of the armed forces to the exercise of
political power during this period. Although it would be a dis-
tortion of the situation to assert that military institutions were
the only forces which impinged upon political life in early me-
dieval Gaul, it would be equally false to argue that such insti-
tutions were not of great importance in the political sphere. It
is worth recalling here the words of the Roman Emperor Sep-
timius Severus that the army should be showered with gold and
all else should be regarded as of secondary consequence. For it
cannot be ignored that Merovingian society and its military or-
ganization grew up among the ruins of the Roman Empire, that
they reflect Romania more than Germania,
vii
PREFACE
Through their encouragement and advice many people have
made my work on this book both easier and more pleasant. I
would like, however, to make special mention of Professors F. L.
Ganshof, Walter Goffart, R. S. Hoyt, and Bryce Lyon who read
the entire manuscript at one or another stage in its history and
provided valuable criticism. It would be less than fair, in addi-
tion, if I did not share with my wife Debby the credit for bring-
ing this work to completion. Despite her obligations as wife,
mother, and scholar she found the time to read proofs and help
with the compilation of the index. Lastly, it gives me pleasure to
note that this study was awarded the McKnight Foundation Prize
in European History in 1968.
BERNABD S. BACHRACH
November 3,1971
St. Paul, Minn.
viii
Contents
Chapter I
Clovis: 481–511 3
Chapter II
The Sons of Clovis: 511–561 18
Chapter III
Clovis's Grandsons: 561-593 36
Chapter IV
The Last of the Ruling Merovingians: 593–638 74
Chapter V
The Rois Fainéants and the Mayors of the Palace: 638–751 92
Chapter VI
Conclusion 113
Appendix
The Byzantine's View of the Franks 131
Bibliography 141
Index 151
Abbreviations
AHR American Historical Review
BEC Bibliothèque de l'École des Chartes
CRAI Comptes Rendus de I'Académie des Inscriptions
et Belles-Lettres
HZ Historische Zeitschrift
JRS Journal of Roman Studies
MA Le Moyen Âge
MGH Monumenta Germaniae Historica
AA Auctores Antiquissimi
Ep Epistolae
LL Leges
SS Scriptores
SSRL Scriptores Rerum Langobardicarum
SSRM Scriptores Rerum Merovingicarum
MIÖG Mitteilungen des Instituts für Österreichische Ge-
schichtsforschung
RE A Revue des Études Anciennes
RH Revue Historique
RPRH Revue Beige de Philologie et d'Histoire
ZRG Zeitschrift der Savigny-Stiftung fur Rechtsgeschichte
Merovingian
481-751
CHAPTER I
Clovis: 481-511
A FTER the death of Aetius in 454, imperial power in Gaul

rapidly disintegrated and the wealth of this Roman province
was controlled by those who were able to muster the armed
force necessary to keep it. By 481 the two peoples competing for
predominance in this territory were the Visigoths in southwestern
Gaul and the Burgundians in the southeast. Among the lesser
groups contending for power were the Armorid (a loose confed-
eration of Gallo-Romans, Britons, Alans, and erstwhile imperial
soldiers with their families), who lived in the area between the
Seine and the Loire. To the north, between the Seine and the
Somme, was Syagrius's Roman kingdom of Soissons and to the
east along the upper Rhine were settlements of Alamans. North
of these was a small band of Thuringians. The remainder of the
Rhineland and the area to the west were ruled by Prankish reguli
or chieftains, who, with their warbands, were settled around
Tournai, Cambrai, Cologne, and the other cities of the region.1
It was one of these reguli, Clovis, the ruler of Tournai, who,
starting with a small warband, was to gain the support of an
armed force large enough to take control of most of Gaul. The
1
Procopius, H.W., V, xii, 8ff; Jordanes, Getica, LV. On this see Bernard
S. Bachrach, "Procopius and the Chronology of Clevis's Reign," Viator, I
(1970), 21-25, and cf. J. M. Wallace-Hadrill, The Long-Haired Kings
(London, 1962), p. 161.
3
MEROVINGIAN MILITARY ORGANIZATION
Merovingian military began with Clevis's warband and grew with

his and his descendants' successes.
In 481, at about the age of sixteen, Clovis succeeded his
father, Childerie, as ruler of Tournai. He thereby came into
possession of the imperial installations including the fisc, and
took command of his father's warband, which probably amount-
ed to no more than 400 or 500 warriors. Among Childeric's
followers a prominent role seems to have been played by a Hun
named Wiomad, which suggests that other than Prankish ele-
ments may have served Clovis.2
In 486 Clovis began his efforts to expand the kingdom when
he allied with his relative Ragnachar, the ruler of Cambrai, to
attack Syagrius of Soissons. As the battle progressed, Syagrius,
seeing his troops defeated, fled south to Toulouse and sought
refuge with the Visigothic king, Alaric II. The latter, however,
handed Syagrius over to Clovis who liquidated him.3 This vic-
tory over Syagrius gave Clovis control of the civitas of Soissons
with its fortifications and arms factory, the existence of which
may well have been partly responsible for the Roman ruler's
choosing it as a headquarters initially.* Other cities within the
area between the Somme and the Seine did not immediately
accept Clovis's rule. He laid siege to Verdun and it finally sur-
rendered. Paris was also besieged, but its conquest presented
Clovis with a knotty tactical problem. Though the extant sources
concerning the siege of Paris are admittedly of limited value,
they suggest that it took Clovis several years, perhaps as many
as five, to gain control of the city, or more precisely of the
fortress. The He de la Cite seems to have been supplied by
boat, and Clovis, at that point in his career, did not have at his
disposal naval forces of sufficient strength to halt such activity.5
2
Though much has been written since its appearance, G. Kurth's Clovis,
2nd ed., 2 vols. (Paris, 1901), remains fundamental. Wallace-Hadrill's
Long-Haired Kings, pp. 159-185, is of special worth. On the size of Clovis's
forces see Bachrach, "Clovis's Reign," p. 28, and the literature cited there.
8
Gregory, Hist., II, 27.
* Wallaee-Hadrill, Long-Hatred Kings, p. 159.
B
V. Genov., ch. 35: "Factum est autem, ut Genovefa in Arciacinse opido
navali effectione ad conparandam annonam profiscisceretur"; and eh. 40:
4
CLOVIS
It is not known precisely at what date Clovis finally secured

control of the entire Roman kingdom of Soissons. But probably
by 491, when he directed his attention elsewhere, not only had
the Roman units which had served Syagrius been integrated into
his following, but the military colonists (laeti) settled in that
area under the empire had also come under his control.6 A rel-
atively large amount of evidence has survived concerning the
laeti. In Belgica II between Rheims and Amiens there was a
Sarmatian military colony. From place names which are still
extant, indicating a long-lasting habitation, it can be ascertained
that there were settlements within this colony at Sermiers
(Marne), Sermaize les Bains (Marne), Sermoise (Aisne), and
Sermaise (Oise).7 In the colony established between Paris and
Saint-More (Yonne), two other settlements can be identified —
Sermaise (Seine-et-Oise) and Sermaise (Seine-et-Marne).8 There
were Alaman military colonies at Rheims and Senlis.9 All these
military colonies were corpora publica under the empire and it
seems reasonable to assume that Clovis gained control of them
in the same manner as he took over the remains of the imperial
fisc.10
When Clovis turned his attention from the area between the
Somme and the Seine in about 491, he moved against a small
group of Thuringians settled in eastern Gaul north of the Bur-
gundians. Nothing is known about this campaign except that
Clovis was victorious. It can be assumed, however, that as a re-
sult of this campaign his interests were more likely to come into
"Regressa itaque Parissius, unicuique, prout opus fuit, frugem dispersit."
V. Maximini, p. 393. On these see Kurth, Clovis, I, 245-248.
6
Bachrach, "Clevis's Reign," pp. 23-24. Cf. Wallace-Hadrill, Long-Haired
Kings, pp. 164-165.
7
Notitia Dignitatum, oc., XLII, 67: "Praefectus Sarmatarum gentilium,
inter Renos et Tambianos provinciae Belgicae secundae." A. Longnon, Les
noms de lieu de la France (Paris, 1920), no. 528, for Sermiers, and no. 532,
for Sarmaize les Bains, Sermoise, and Sermaise.
8
Notitia Dignitatum, oc., XLII, 66: "Praefectus Sarmatarum gentilium,
a Chora Parisios usque." Longnon, Les noms, no. 532, for both Sermaises.
9
Notitia Dignitatum, oc., XLII, 42.
10
A. H. M. Jones, The Later Roman Empire (Norman, Okla., 1964), I,
620; A. Grenier, Manuel d'archeologie Gallo-Romain (Paris, 1931), V,
398ff.
5
direct conflict with those of the Alamans along the middle Rhine
and those of the Burgundians to the south.11
In 496 the Alamans moved into Prankish territory, and the
various Salian and Ripuarian reguli seem to have joined their
followings under Clovis's leadership to stop this invasion. At
Tolbiac, some thirty miles south of Cologne, the two armies
met and the Alamans were defeated. Though Tolbiac was a
formidable fortress, it seems that this battle, unlike several of
Clovis's previous actions, did not involve siege warfare. The exact
nature of the battle, however, aside from its outcome, remains
unknown.12
Clovis's victory at Tolbiac was of much greater consequence
than merely the turning back of an enemy invasion; his ad-
herence to Christianity and subsequent baptism were by and
large directly related to this campaign. According to Gregory
of Tours's account, Clovis seems to have undergone a religious
experience: the Alamans were treating Clovis's forces roughly
and he asked the Christian God for help, promising in effect
that if his enemy were defeated he would become a Christian.13
Although it is not impossible that Clovis underwent a reli-
gious experience, it may be suggested that the interpretation of
Clovis's postvictory baptism by Gregory or those he used as his
sources resulted more from sincere wish fulfillment than from
solid evidence. The political and military realities of Clovis's
situation and his personality (even as depicted by Gregory)
would reasonably lead us to conclude that the Prankish chief
found motivation for his actions in other than spiritual demands.
Clovis is known to have believed that if he were baptized he
would lose the support of many of his pagan followers. To have
been baptized in spite of this implies strong motivation indeed.14
"Bachrach, "Clovis's Reign," p. 21. Cf. Kurth, Clovis, I, 266ff.
13
Gregory, Hist., II, 30. The date of the battle of Tolbiac and Clovis's
baptism, which seems to be intimately connected with it, is a major prob-
lem in the chronology of Clovis's reign. On this see Bachrach, "Clovis's
Reign," p. 27, and the bibliography cited there.
13
" Ibid., II, 31: "At ille [Clovis] ait: 'Libenter te, sanctissime pater, audie-
bam; set restat unam, quod populus, qui me sequitur, non patitur reliquere
deus suos; sed vado et loquor eius iuxta verbum tuum.'"
6
CLOVIS
It may be hypothesized that Clovis accepted Christianity and
attempted to impose it upon his pagan followers to ensure the
support of the Christian Gallo-Romans, especially the lay and
ecclesiastical magnates who controlled great wealth and power.
It should be remembered that the Alamans, Clovis's enemies,
were pagans like the Franks and that the Gallo-Romans, while
surely hostile to Arian barbarians like the Visigoths, would
find little difference between one pagan group and another.
Shortly after Clovis had come to power, he had received a
letter from Remigius, bishop of Rheims, indicating to the young
king that he would find it advantageous to have the support of
the Gallo-Roman church. The bishop's letter even implied that
Clovis should show deference to the bishops and ask their advice;
a meeting of minds between the young ruler and the influential
churchmen, wrote Remigius, would assure a prosperous reign.
This letter may be regarded as more a veiled threat than a hum-
ble request to a pagan ruler for religious toleration, for the
Gallo-Roman church was the powerful leader of the Christian
community and the bishops, many of whom were from old and
established families, were not to be slighted.15 Men like Hilarius,
bishop of Aries, who during the previous generation had led his
band of armed retainers contrary to imperial proscription, could
be either useful supporters or very dangerous enemies of a
young and untried barbarian monarch. Bishops Quintianus of
Rodez, Volusianus of Tours, Aprunculus of Langres, and Verus
of Tours had caused so much trouble for the Visigothic and
Burgundian leaders that ultimately they had to be driven from
their sees. The details of their opposition to such barbarians as
the Visigoths and Burgundians are unknown, but it is doubtful
that they went as far as Bishop Galactorius of Beam who was
killed while leading an armed force to support Clovis against
the Visigoths. On the other hand, these bishops may have been
more like Cautinus of Auvergne who used his armed supporters
to pillage his neighbors.16
M
MGH epist. Merov. et Karl, II, 113.
16
N. Vol., 17, 1, for Hilarius and his armed followers. Gregory, Hist., II,
36, for Quintianus; II, 22, and X, 31, for Volusianus; II, 23, for Aprunculus;
7
Whether or not Clovis had made overtures to the bishops as a

result of Remigius's letter is not known. But he apparently did
not enjoy the support of the ecclesiastical magnates in 486 when
he went to battle against Syagrius. Clovis had followed up his
victory in Soissons by pillaging Syagrius's kingdom, including
the churches. The booty had not even been divided when Clovis
received a messenger from Remigius requesting that the items
taken from the church at Rheims, especially a very valuable ewer,
be returned. Clovis did return at least the ewer, and it seems
somewhat likely that this gesture marked an effort by the Prank-
ish king to establish or reestablish cordial relations with the
powerful bishop of Rheims.17
By the time of the battle of Tolbiac (ten years after the pil-
laging of Syagrius's kingdom), Clovis had made some further
concessions to the Gallo-Roman power structure: he had taken
a Christian wife and had permitted his children to be baptized.
These might not have been regarded as sufficient, however, and
he realized the wisdom of Remigius's earlier admonitions that
the support of the churchmen would be advantageous. Clovis
now was in a difficult position: with the help of the Gallo-Roman
magnates he could rule Gaul; without it he was just another
regulus. The Alaman invasion presented a choice to the Gallo-
Romans and they were in a position to demand a large price
for their support; presumably they did. Additional support for
this interpretation is lent by a story preserved in the Liber
Historiae Francorum in which Aurelianus, a Gallo-Roman who
served Clovis in a military capacity, is credited with suggesting
X, 31, for Verus; and V. Galoot., p. 434, for Galactorius. For discussions of
this see Kurth, Clovis, II, 84-85, and Samuel Dill, Roman Society in Gaul
in the Merovingian Age (London, 1926), p. 95. Cf. E. A. Thompson, The
Goths in Spain (Oxford, 1969), pp. 26-27. For Cautinus see Gregory, Hist.,
IV, 12, and the discussion below in ch. 2, n. 23.
"Gregory, Hist., II, 27. The legend of the recalcitrant warrior and the
ewer which is interwoven into this episode may perhaps symbolize a fac-
tional struggle among Clovis's followers concerning the church, a struggle
which seems to have had later repercussions. See G. Kurth, Histoire poeti-
que des Merovingiens (Paris, 1893), pp. 223-224, and Kurth, Clovis, I, 244.
8
CLOVIS
to the Prankish chief at Tolbiac that his military situation would
prosper if he became a Christian.18
In any event, Clovis judged correctly when he concluded that
his baptism would not be supported fully by his pagan followers.
Of the 6000 or so Prankish warriors who formed the warbands
of the reguli, only some 3000 followed Clovis to the baptismal
font. The other half joined with Ragnachar, the regains of Cam-
brai, who remained a pagan.19 Although Clovis had probably lost
about one half of his Prankish followers, he was still able, in the
years following his baptism, to continue his military activities.
In 500 or 501 relations between Godigisel and Gundobad, the
Burgundian kings, became strained and the former asked Clovis
for military aid against the latter. In return for this aid, Godigisel
agreed to give Clovis a yearly payment as well as a portion of the
Burgundian kingdom. When Clevis's army entered Burgundy,
Gundobad told his brother of the invasion and asked him to take
the field against the Prankish ruler. To Gundobad's surprise,
when the three armies arrived outside of Dijon, he was attacked
by his brother as well as by Clovis. Though his army was de-
feated, Gundobad managed to escape to the fortified city of
Avignon.20
After this battle, Godigisel retired to Vienne, but Clovis led
his followers against Gundobad at Avignon and laid siege to the
fortress, presumably with the siege engines (instrumenta belli]
with which he had begun the campaign. Clovis's men devastated
the area around the city, pillaging and burning everything within
reach. But a certain Aridius, a Gallo-Roman magnate in the
service of Gundobad, gained Clovis's confidence and convinced
18
L.H.F., ch. 15. For further discussion of Aurelianus see n. 41 below.
Cf. Wallace-Hadrill, Long-Hatred Kings, p. 168.
19
V. Remigii, ch. 15: "Multi denique de Francorum exercitu, necdum ad
fidem conversi, sum regis parente Ragnario ultra Sumnam fluvium aliquandiu
deguerunt, donee, Christi gratia cooperante, gloriosis potitus victoriis, eundem
Ragnarium, flagitiis turpitudinum inservientem, vinctum a Francis sibi
traditum, rex Hludowicus occidit et omnem Francorum populum per beatum
Remigium ad fidem converti et baptizari obtinuit." Gregory, Hist., II, 31:
"De exercito vero eius baptizati sunt amplius tria milia." Dill, Merovingian
Age, pp. 77, 80; Bachrach, "Clovis's Reign," pp. 27-28; and Wallace-Had-
rill, Long-Haired Kings, p. 170.
20
9
him that Avignon could not be taken. Peace was then negotiated;
Gundobad agreed to pay Clovis a yearly tribute and the latter
returned home with his followers.21
A year or two later Clovis led his forces into Armorica, the
area between the Seine and the Loire. Unlike his previous raids
into this area (which may have taken place during the late 480s),
this campaign seems to have been aimed at subjugating the Ar-
morici. Clevis's forces, however, were defeated. Yet the Armorici,
who, like Clovis, were Christians, desired cordial relations with
him to counterbalance the hostile Arian Visigoths who coveted
the land north of the Loire. Therefore, an accord was arranged
by which Clovis came to rule the various peoples of Armorica
and the military strength of the area was integrated into the
Merovingian military.22
Among the inhabitants of the Tractus Armoricani who joined
Clovis's military following were the Alans, who dominated the
Orleanais and the area to the north, Gallo-Roman magnates and
their adherents, former Roman soldiers and their descendants
who had maintained their military organization, Bretons, im-
perial military colonists (laeti) from Chartres, Bayeux, Coutance,
Le Mans, and Rennes, and Rome's Saxon allies of Bayeux.23 Not
long after acquiring the support of the inhabitants of Armorica,
Clovis again went to war against the Alamans, who had regained
some of their previous strength during the decade of disunity
among the Prankish reguli following Clovis's baptism. No details
of the campaign survive other than the fact that Clovis was
victorious. After this victory some of the defeated Alamans fled
to the lands of Theodoric the Great, while others remained
21
Ibid.
^Procopius H.W., V, xii, 13ff. On this text see Bachrach, "Clovis's
Reign," pp. 28-29.
""Procopius, H.W., V, xii, 16ff. On the Alans see Bernard S. Bachrach,
"The Origin of Armorican Chivalry," Technology and Culture, X (1969),
166-171, and "The Alans in Gaul," Traditio, XXIII (1967), 476-489. For
the laeti see Notitia Dignitatum, oc., XLII, 33, 34, 35, 36. On the survival
of Roman units in Gaul see Bernard S. Bachrach, "A Note on Alites,"
Byzantinische Zeitschrift, LXI (1968), 35.
10
CLOVIS
under Merovingian domination and ultimately served in the
armed forces of Clovis's successors.24
In 507, after having obtained the consent of his magnates,
Clovis launched an invasion of the Visigothic kingdom. The
Merovingian king made a concerted effort to keep the support
of the church by forbidding his troops to plunder its possessions.
One of Clovis's orders restricted his men to the use of only grass
and water in provisioning their mounts on the church lands they
traversed. Clovis's horsemen generally seem to have abided by
this order, although present knowledge of it may well have sur-
vived because of the drastic manner in which a violator was
punished. Much to the joy of such clerical observers as Gregory
of Tours, Clovis enforced this order by killing the man who vio-
lated it with his own hand.25
Though Clovis's advance south was slowed somewhat by the
rain-swollen Vienne River, his forces reached the Poitiers area
and engaged the Visigoths at Vouille, a plain to the south of
the city on the Roman road to Nantes. Clovis's archers and
spear throwers, apparently deployed at the rear of his formation,
showered the Goths with missiles from a distance. While this
barrage battered the enemy, other elements of Clovis's force
advanced and engaged the Goths in hand-to-hand combat. Clovis
is alleged to have killed Alaric, the Visigothic king, in single
combat, and only just escaped death himself when the speed of
his horse and the strength of his cuirass combined to thwart the
efforts of two Gothic warriors. The Visigoths turned and fled
and Clovis's forces won the day.26
Clovis followed up his triumph over the Visigoths by sending
his son Theuderic to seize the cities held by the Goths in the
south of Gaul. Albi, Rodez, and Clermont all fell under Mero-
24
F. Lot, "La victoire sur les Alamans et la conversion de Clovis," RBPH,
XVII (1938), 63-69, and Bachrach, "Clovis's Reign," p. 30. Cf. A. Van de
Vyver, "L'unique victoire centre les Alamans et la conversion de Clovis en
506," RBPH, XVII (1938) 793-813.
25
Gregory, Hist., II, 37: ". . . ut nullus de regione ilia aliud, quam her-
barum alimenta aquamque praesumeret."
28
Ibid.: "Chlodovechus rex cum Alarico rege Gothorum in compo Vogla-
dense decimo ab urbe Pictava milario convenit, et confligentibus his eminus,
resistunt comminus illi."
11
vingian control, and Gregory of Tours somewhat optimistically

remarks that the whole country from the Gothic to the Bur-
gundian border was brought under Clevis's domination.27 The
Merovingian advance, however, did not go unchallenged. A
force of Merovingians and Burgundians, which besieged Aries
for several months, was driven off by an Ostrogothic army and
suffered great losses.28 Clovis took Bordeaux in 507 and win-
tered in the city. During the next year he captured Toulouse
and the Visigothic treasure stored there. While besieging Car-
cassonne, however, Clevis's army was driven off by Ostrogothic
troops. Clovis then moved north and took Angouleme, forcing
out the Gothic garrison. To secure his conquests, Clovis estab-
lished garrisons throughout the newly won territories; the sources
make specific mention of garrisons at Toulouse, Rodez, Saintes,
and Bordeaux.29
Although Clevis's control of the Visigothic kingdom in Gaul
was challenged and he lost Septimania, he was able to secure
the more northerly areas. The Taifal and Sarmatian laetl in the
Poitiers area, as well as the Sarmatians in the Rodez-Velay re-
gion, joined the ranks of the Merovingian military. The Alaman
laeti who garrisoned fortresses in the Auvergne, and some of the
Alans living in the area between the Mediterranean and Tou-
louse may probably also be counted as additions to Clovis's
forces.30
"Ibid.
28
V. Caes., ch. 28, and Jordanes, Getica, LVIII.
29
Gregory, Hist., II, 37; Procopius, H.W., V, xii, 4 Iff; Isid., Hist. Goth.,
281-282: "Alaricus . . . tandem provoeatus a Francis in regione Pictaven-
sis urbis proelio initio extinguitur eoque interfecto regnum Tolosanum
occupantibus Francis destruitur"; L.H.F., ch. 17: "In Sanctonico vel Burdiga-
linse Francos precepit manere ad gothonim gentem delendam"; Gesta Fran-
corum, col. 414-415: "His igitur ita patratis, cum ad solum proprium re-
dire deliberaret, electos milites atque fortissimos cum parvulis atque mulieri-
bus ad pervasas civitates custodiendas et ad reprimendam Gothorum saevi-
tiam dereliquit, et immensis muneribus ampliavit." For a discussion of the
garrisons and a defense of the Gesta on this point see Wallace-Hadrill,
Long-Haired Kings, p. 31, n. 2. To the above references add V. Dalmat.,
ch. 6: "Cumque ad ilium devotissimus ardue festinaret, in Ultralegeretanis
partibus quodam loco, ubi aliqua, ut dicam prope legio Bretonum manet.
. . ." See also n. 23 above and n. 38 below.
^Notitia Digntiatum, oc., XLII, 65: "Praefectus Sarmatarum et Taifal-
orum gentilium, Pictavis [in Galia]." The area around Poitiers was so thor-
12
CLOVIS
In the decade after his baptism, Clovis vastly increased the

size of his kingdom, added huge quantities of wealth to his
treasury, and greatly augmented his military establishment. He
then turned his attention, in what were to be the last years of his
life, to the Prankish warbands which had deserted him in 496
to follow Ragnachar. Largely through bribery, assassination, and
murder, Clovis was able to eliminate the reguli and take over
their warbands. Among the many with whom he dealt (all
seemingly his relatives) was Sigibert of Cologne who was mur-
dered by his own son at Clevis's instigation. Clovis then had the
son murdered. Chararic and his son were captured and executed
soon after. Ragnachar and his brothers Riccar and Rignomer
seem to have been the last of the reguli to fall. Clovis bribed
their leudes, who consequently refused to fight, then captured
their former lords; the brothers were handed over to Clovis who
killed them. With Ragnachar's death the remaining warbands
joined Clovis and received baptism.31
Concerning Clovis's following and the warriors of the other
reguli little is known. Whether the members of these forces may
be identified with the antrustiones is a matter of conjecture,
though surely at least some of them were bound by the special
oath of those in the trustis.32 The leudes of a regulus were also
part of his following. The important role played by Ragnachar's
leudes in his fall from power suggests that they were not simple
warriors, but perhaps minor leaders with small warbands of their
own.
Though the nature of these warbands, of whatever size, still
generally remains obscure, some light may be thrown on the
oughly influenced by the Taifals that it was called Thifilia during the sixth
century. Gregory, V.P., XV, 1: "Igitur beatus Senoch, gente Theiphalus,
Pictavi pagus, quem Theiphaliam vocant oriundus fuit." A. Longnon, Geo-
graphie de la Gaule au V7" siecle (Paris, 1878), p. 176, and Grenier, Man-
uel d'archeologie, V, 398. For Alaman laeti see n. 37 below, and for the
Alans see n. 22 above.
31
Gregory, Hist,, II, 40, 41, 42. See Bachrach, "Clovis's Reign," p. 30,
and cf. Kurth, Clovis, I, 266-268.
32
M. Deloche, La trustis et I'antrustion royal sous les deux premieres
races (Paris, 1873), though old and in need of revision, is still the most
complete work on the subject.
13
equestrian nature of Prankish arms. The Roman empire had

found the Franks to be able horsemen and therefore, under its
pragmatic policy of using barbarians for the kind of service to
which they were best suited, recruited them for cavalry regi-
ments. Of the four manifestly Prankish units listed in the Notitia
Dignitatum, which was current during the first third of the fifth
century, if indeed not later, all were mounted.33 Childeric him-
self was so equestrian-oriented that he had the head of his war-
horse entombed with him at Tournai.34 And Clevis's own order
regarding the taking of food and water for mounts immediately
before the Visigothic campaign in 507 and the execution of an
unimportant Frank for its violation illlustrate the significance
of horsemen in his army. The importance of horses to the Franks
is clearly indicated by the frequent legislation regarding them,
even in the earliest compilation of the Lex Salica, about 510.35
It may even be speculated that at the time of Clevis's baptism
Bishop Remigius referred to Clovis as a "Sicamber" because the
Frankish leader and his followers brought to mind their putative
ancestors, the Sicambri, whose 2000 horsemen had gained some
fame from Caesar's writings. Had Remigius been impressed with
the pedestrian, rather than the equestrian, nature of the Franks'
military mien, he probably would have called Clovis a Chatua-
rian, evoking memories of another putative ancestral tribe whose
forte was infantry.36
Among the laeti added to Clovis's fighting forces horsemen
were also prominent. Sarmatians, Taifals, and Alamans had a
long history as mounted warriors and the empire had responded
predictably by recruiting them for cavalry regiments; in fact, all
Taifal and Sarmatian units in the Roman army were cavalry.
Clovis acquired laeti as his followers and even admitted them
to his personal following as antrustiones?'*
33
Notitia Dignitatum, or., XXI, 51; XXXII, 35; XXXVI, 33; XXXI, 67.
34
Cf. Wallace-Hadrill, Long-Haired Kings, p. 162.
38
See Appendix, n. 27.
•""Gregory, Hist., II, 31: "'. . . Mitis depone colla, Sigamber; adora quod
incendisti, incendi quod adorasti.' " Caesar, E.G., VI, 35: "Cogunt equitum
duo milia sugambir. . . ." and Tacitus, Germ., 30: ". . . omne robur in
pedite. . . ." concerning the Chatti.
37
Notitia Dignitatum, oc., VI, 59; XL, 54; or., V, 31; XXVIII, 26; XXXII,
14
CLOVIS
The remains of the formal units of the Roman military also

swelled the ranks of Clevis's army. Many of these infantry and
cavalry regiments apparently maintained their military organi-
zation and training after the collapse of the empire, even hand-
ing down their customs to their descendants.38 Surely only a
small percentage of the descendants of the approximately 70,000
troops listed as being stationed in Gaul about 408 continued the
military heritage of their ancestors to the end of the fifth century
and beyond, but nevertheless some did. Of these the horsemen,
of which there were about 6000, were probably the most useful
to Clovis since they were elite troops. At least some of them
possibly continued to function as they seem to have done in
Britain.39
Among Rome's former allies in Gaul who came under Mero-
vingian control were the Saxons of Bayeux, the Alans of Armor-
ica, and some of those Alans established in the area between
Toulouse and the Mediterranean. The Saxons may well have
been primarily seamen, but the Alans were renowned as horse-
men. The latter's equestrian tactics, common to the steppes,
continued to influence military development in Armorica well
into the Middle Ages.40
Another element in the Merovingian military was the armed
forces of the Gallo-Roman magnates. As early as 493, Clovis
could count on the support of a Gallo-Roman magnate like Aureli-
anus, who led his band in Clovis's service and who was given
Melun as a reward for his loyalty.41 Although the sources are
36; XXXIII, 31. Ley. Sal., XLII, 1: " . . . si in truste dominica fuerit iuratus
ille qui occisus, LXXIIM denarios . . . culpabilis. . . ." Also 4: "De Ro-
manis vero occisis vel letis [et pueris] haec lex [superius conpraehensa ex]
medietate solvatur."
88
Procopius, H.W., V, xii, 16ff. Bachrach, "Alites," p. 35.
88
J. B. Bury, "The Notitia Dignitatum," JRS, X (1920), 146; Jones,
Later Roman Empire, I, 619; II, 1449-1450. R. G. Collingwood and J. N. L.
Myres, Roman Britain, 2nd ed. (Oxford, 1937), pp. 320ff.
40
Bernard S. Bachrach, "The Feigned Retreat at Hastings," Mediaeval
Studies, XXXIII (1971), 264-267, and Bachrach, "Armorican Chivalry,"
pp. 166-171.
41
L.H.F., chs. 11, 12,14,15.1 cannot agree with Kurth, Clovis, II, 156, that
Aurelianus is a legendary figure. Support for his historical existence may be
found in K. F. Stroheker's Der senatorische Adel in spatantiken GalUen
15
extremely limited, it does seem reasonable to conclude that

many such lay magnates joined Clovis during his successful
career. It would be difficult to explain their failure to do so
in light of the support Clovis received from the ecclesiastical
magnates, the Roman lawyers who helped in drawing up Lex
Salica, and the bureaucrats who operated his chancery.42
The magnates and their armed retainers had long played a
role throughout the empire, and Rome had an ambivalent atti-
tude toward the keeping of armed followers by these powerful
men. Rome recognized that such groups served her own interests
when used for the suppression of banditry and for the thwart-
ing of barbarian raids, but she was equally aware that the mag-
nates tended to utilize their armed retainers to help them gain
control of large areas and to make the populations subject to
private jurisdictions.43 In Gaul there were Gallo-Roman magnates
like Ecdicius, the brother-in-law of Sidonius Apollinaris, who
raised a band of horsemen supported by his own funds and led
it against the Visigoths.44 Another such magnate was Titus; his
activities as the leader of a band of buccellarii in Gaul gained
the notice of the Emperor Leo, who was so impressed that he
invited him to come to the East and gave him the title of
comes.45 Sidonius, in encouraging Eutropius to take a more
active part in political affairs, listed his friend's qualifications
for such action: an abundance of horses, armor, clothing, money,
(Tubingen, 1946), which lists four Aureliani (numbers 46-49), any of
whom might be identified with Clevis's follower.
43
Wallace-Hadrill, Long-Haired Kings, pp. 177ff.
43
R. MacMullen, Soldiers and Civilians in the Later Roman Empire
(Cambridge, Mass., 1963), ch. VI. C. Th., 7.18.14; 9.14.2; N. Vol., 91.
While many edicts in the Theodosian Code are specifically issued for a
particular area, the conditions with which they deal may well have been
more widespread at any given time. See Bernard S. Bachrach, "Was There
Feudalism in Byzantine Egypt?" Journal of the American Research Center in
Egypt, VI (1967), 163-166.
44
Sidonius Apollinaris, ep. Ill, iii, 3, 5, 7: ". . . cum interiectis aequoribus
in adversum perambulatis et vix duodeviginti equitum sodalitate comitatus
aliquot milia Gothorum. . . ." Also, "taceo deinceps collegisse te privatis
viribus publici exercitus speciem parvis extrinsecus maiorum opibus adiutum
et infreniores hostium ante discursus castigatis cohercuisse populatibus."
45
V. Danielis, ch. 60. See on this text Jones, Later Roman Empire, I, 666.
16
CLOVIS
and retainers.46 With such assets, and protected in his fortified
villa, the Gallo-Roman magnate was a formidable figure in local
affairs as well as on the battlefield.47 With the support of this
class of men and their followers, Clovis could provide profes-
sional fighting men for local defense and for the preservation of
law and order as the empire had done before him. For they were
an important element in the development of royal power.
The heterogeneous nature of the Merovingian military — war-
bands of the reguli, armed followings of the magnates, descen-
dants of Roman soldiers who preserved their military customs,
laeti, and Rome's barbarian allies — gave Clovis an experienced
professional fighting force which could campaign throughout
the year and was not restricted by the demands of agricultural
labor as a part-time army of warrior-farmers would be. With a
substantial mounted force of Alans, Sarmatians, Taifals, and some
Franks and Alamans, Clevis's troops could successfully fight hand
to hand against the Visigothic cavalry. His archers and spear
throwers, probably drawn from erstwhile Roman units and from
his own Franks, provided firepower on the field.48 In siege war-
fare, undertaken by Clevis's forces on numerous occasions — at
Verdun, Paris, Avignon, Aries, Carcassonne, and Angouleme —
siege engines and other apparatus were probably employed.
Not only did Clevis's troops have the offensive capability to
fight effectively on the open field as well as against fortified posi-
tions, but they also had the organizational flexibility both to fight
year-round campaigns and to serve as garrisons in the many sur-
viving fortified cities and castra in Gaul.49 It can therefore be
concluded that Clovis resuscitated the remains of the imperial
military in Gaul and created the Merovingian military.
"Sidonius Apollinaris, ep. I, 6, 2: ". . . dein quod equis armis, veste
sumptu famulicio instructus solum. . . ."
47
MacMullen, Soldiers and Civilians, pp. 143, 147, and C. Jullian, His-
toire de la Gaule Romaine (Paris, 1928), VIII, 139ff especially 141. Mac-
Mullen, p. 147, n. 101, feels that some of the fortified villas identified by
Jullian are dubious.
48
Natitia Dignitatum, oc., VII, 75, and Gregory, Hist., II, 9: "... sagittas
turmentorum ritu effundere inlitas herbarum venensis. . . ."
49
R. M. Butler, "Late Roman Town Walls in Gaul," The Archaeological
Journal, CXVI (1959), 48-50.
17
CHAPTER II
The Sons
of Clovis: 511-561
W HEN Clovis died in 511, his kingdom was divided among his
four sons, Theuderic, Childebert, Chlodomer, and Chlotar.
The first military encounter in which the Merovingian kings en-
gaged after their father's death (at least the first of which some
record has survived) was against a force of Danish raiders who
invaded northern Gaul about 515. The Danes had come by sea
and had carried out a successful raid in which they acquired a
good deal of loot. The main force had returned to the ships and
only their king, Chlochiliach, remained on land, presumably
with a small group of his armed followers. Gregory of Tours
notes that Theudebert, having been sent by his father, Theuderic,
to repel the invaders, met them in a sea battle, defeated them
decisively, killed Chlochiliach, and recovered the booty from
the enemy ships.1
Several points are worth noting about this encounter. King
Theuderic, or at least his son Theudebert, had at his disposal
ships capable of engaging in sea warfare and men with the ability
to defeat a seagoing people at their own game. Also, Theudebert's
troops are described by Gregory as being a validus exercitus or
1
Gregory, Hist., Ill, 3: "Quod cum Theudorico nuntiatum fuisset, quod
scilicet regio eius fuerit ab extraneis devastata, Theudobertum, filium suum,
in illis partibus cum valido exercitu ac magno armorum apparatu direxit.
Qui, interfectu rege, hostibus navali proelio superatis oppraemit omneque
rapinam terrae restituit."
18
THE SONS OF CLOVIS
strong force "cum . . . magno armorum apparatu" — with exten-
sive military equipment. There is no record, however, of the
Franks as an ethnic group taking part in any noteworthy naval
activity either before or after this event. It should be recalled
that when Clovis besieged Paris twenty-five years earlier, he
did not have sufficient naval forces to keep the city from being
supplied by water. All this suggests that at least a part of Theu-
debert's validus exercitus was not composed of Franks. The
abundance of sophisticated military equipment noted by Gregory
also implies that there was probably a non-Frankish element in
Theudebert's army on this occasion.2
The sources provide little information concerning the Mero-
vingian military between the defeat of the Danes in 515 and
Theuderic's Thuringian campaign in 531. Shortly after Theude-
bert's encounter with the Danes, Theuderic aided one of the
Thuringian kings in a civil war; in return for this support he
had been promised a share in whatever land was conquered.
Although territory was taken, Theuderic, much to his dismay,
did not receive his share.3 Some fifteen years later, he felt ready
to take not only what he regarded was his, but more. He enlisted
the help of his brother Chlotar I and Theudebert, assembled his
forces, and moved into Thuringia. The Thuringians, learning of
the Franks' approach, dug ditches in the field where they knew
a battle would be joined and covered them with sod so that the
approaching Prankish horsemen could not see the trap. When
the Franks charged across the field to attack the Thuringians,
who were waiting for them on the far side, the horses were
tripped up by the ditches. This stratagem broke the momentum
of the Frankish cavalry charge, but Theuderic's forces managed
to recover and finally defeated the Thuringians.4
3
Cf. E. Zollner, Geschichte der Franken (Munich, 1970), p. 153.
3
Gregory, Hist., Ill, 4.
4
Ibid., Ill, 7: "Theudoricus autem, Chlothacharium fratrem et Theudo-
bertum filium in solatio suo adsumptos, cum exercito abiit. Thoringi vero
venientibus Francis dolos praeparant. In campum enim, quo certamen agi
debebant, fossas effodiunt, quarum ora operta denso cispete planum adsimi-
lant campum. In his ergo foveis, cum pugnare coepissent, multi Francorum
equites conruerunt, et fuit eis vald© inpedimentum; sed cognitum hunc
dolum, observare coeperunt. Denique cum se Thoringi caedi vehementer
19
Although troops other than Franks probably took part in

Theudebert's naval campaign against the Danes, Gregory indi-
cates specifically that only Franks (Franci) formed the army
which invaded Thuringia, and that the primary striking force,
if not indeed the entire force, was mounted. Hermanfrid, the
Thuringian king, had fought side by side with Theuderic fifteen
years earlier, but his knowledge of the Merovingians' tactics
proved to be of no avail. For, even though he had expected the
enemy to fight as cavalry and had oriented his defense along
those lines, the Franks were still victorious in their mounted
attack.
Despite the adeptness of Theuderic's Prankish followers in
mounted combat, they seem to have been less than adequate
in siege warfare. This weakness was demonstrated in the
Auvergne campaign in 532 and the Munderic affair.
While Theuderic was in Thuringia, a rumor spread in Au-
vergne that he had been killed in battle, and in consequence
some Arvernians invited King Childebert to take possession of
their district. This attempted change in leadership came to
naught, however, when it was learned that Theuderic was, in
fact, alive. In the following year, Theuderic planned to avenge
himself against the Arvernians for their treachery, but the Franks
who looked to him as their king wanted him to join in the attack
on Burgundy planned by his brothers Childebert and Chlotar.
The Franks threatened to desert him and follow his brothers if
he did not attack Burgundy, and it was only after he promised
them they would obtain much booty and could keep all they
took, even the inhabitants as slaves, that they followed him to
Auvergne. Theuderic and his Franks devastated the territory. A
raid was made on the Church of St. Julien at Brioude; the fort
at Vollore (Lovolautrum castrum) was besieged and stormed,
but the garrison of laeti was taken only by treachery; the fort
at Chastel-Marlhac (Meroliacense castrum} also withstood a
viderent, fugato Hermanefredo rege ipsorum, terga vertunt et ad One-
strudem fluvium usque perveniunt." On the Thuringian campaigns see R.
Leibmann, Der Untergang des thuringischen Konigsreichs in den Jahren
531-5 (Meiningen, 1911).
20
THE SONS OF CLOVIS
siege. At Chastel-Marlhac part of the garrison became overcon-
fident, and in attempting to sack the enemy camp, it was cap-
tured. The defenders had to ransom the captives for a triens
each.5 When Theuderic was satisfied that he had taught the
Arvernians a lesson, he returned to the north. He granted lands
in Auvergne to his kinsman Sigivald, who in turn was to garrison
the area with his followers.6
It is interesting to note here that the fortifications of Chastel-
Marlhac and Vollore, with their garrisons of laeti, mark direct
continuity with the Roman past, both in the physical structures
being defended and in their defenders, who were a corpus pub-
licum under the empire and had passed with the fortifications to
the Merovingian kings. These laeti may also be identified as
descendants of Sueve military colonists stationed in that area and
listed in the Notitia Dignitatum.7
Theuderic's stratagems in the Auvergne campaign had proved
to be superior to his strategy and made up for his forces' in-
ability to reduce fortified positions by siege. Such also was the
case in the Munderic affair, Munderic was a Frank who owned
large estates probably in the region of Vitry-le-Brule near Cha-
lons-sur-Marne; he claimed to be of royal blood and tried to
obtain a part of Theuderic's kingdom. He gathered around him-
self a band of armed men who swore to be his faithful follow-
ers. Theuderic attempted to lure him to his court, but failed
and therefore ordered a force to be sent against the pretender.
Munderic withdrew with his followers behind the walls of Vitry
(Victoriacus castrum). Theuderic's army laid siege to the fortress,
but because they lacked siege engines and even the magnus
aparatus armorum which Theudebert had employed against the
Danes, they could only hurl their spears at the defenders on the
walls. After seven days of such futile activity, Theuderic was in-
5
Gregory, Hist., Ill, 11; HI, 12, 13; Gregory, M, ch. 13.
"Gregory, JuL, ch. 14: "Tune Sigivaldus cum rege praepotens cum omni
familia sua in Arverna regione ex regis jussu migravit. . . ." See Bernard S.
Bachrach, "Charles Martel, Mounted Shock Combat, the Stirrup, and Feu-
dalism," Studies in Medieval and Renaissance History, VII (1970), 68-72.
7
Notitia Dignitatum, oc., LXII, 44: "Praefectus laetorum gentilium Sue-
vorum, Arumbernos Aquitanicae primae."
21
formed by his commander that it was impossible to take the
fortress.8 Theuderic then sent a group led by Arigisel, one of
his personal retainers, to entice Munderic out of Vitry. After
taking many oaths of peace and friendship, Arigisel managed to
lure Munderic out of the fortress; once outside he and his fol-
lowers were cut down in a bloody hand-to-hand combat.9
Theuderic's appointment of Arigisel to lead this expedition indi-
cates that the king had a structure of command within his mili-
tary organization. Arigisel's mission represents the second time,
so far as can be ascertained, that Theuderic delegated command
to someone other than his son. The first to have been so dis-
tinguished was Sigivald, who had been given a command in
Auvergne.
In contrast to the weakness of Theuderic's Prankish followers
in siege warfare, the forces from the south of Gaul led by his
son Theudebert and his brothers Chlotar and Childebert were
considerably more effective in siege operations. This may be at-
tributed to the fact that they recruited their troops from the
more Romanized parts of Gaul. Theudebert and his army con-
quered the Gothic fortress city of Beziers in southern Gaul, and
then seized and sacked the fortress of Dio to the north of Be-
ziers. Cabriers capitulated to Theudebert after learning the fate
of Dio.10
The most important result of successful siege warfare was the
conquest of the Burgundian kingdom by Chlotar and Childebert
in 534. By besieging Autun, the headquarters of Godomar, the
Burgundian king, they were able to take the city, causing the
monarch to flee. The ultimate consequence was the subjugation
of the entire Burgundian realm to Merovingian control.11 With
this conquest, the Merovingian kings added the lands, wealth,
population, and military resources of the Burgundian kingdom
to their own holdings. A more important effect of this victory,
however, was the further penetration of Roman influences into
8
Gregory, Hist, III, 14.
• Ibid.; Fred., Ill, 36.
10
Gregory, Hist,, III, 21. See Longnon, Geographic, pp. 611-612.
"Gregory, Hist., Ill, 11.
22
THE SONS OF CLOVIS
the Merovingian military and political structure. The Burgun-
dian military had for some time seen the syncretism of bar-
barian and Roman institutions. From the second half of the fifth
century, Gallo-Romans served in the Burgundian army. In addi-
tion there was at least one noteworthy unit in the Burgundian
military which was composed completely of Gallo-Romans.12
The important men of the kingdom, Gallo-Roman and Bur-
gundian magnates alike, had for a long time supported private
bands of armed retainers. Sidonius Apollinaris, whose writings
provide a great deal of information on late Roman society in
the Burgundian kingdom, praises his son-in-law Ecdicius for
using his private army against the Visigoths as a public service.
On another occasion, however, Sidonius complains bitterly to
his friend Thaumastus about certain important men whose
access to the Burgundian king endangers the favored position
usually enjoyed by the Gallo-Roman senatorials. Among Sidoni-
us's criticisms of these men is their lack of noblesse oblige. In
this vein he contends that they not only failed to provide their
military retainers with the stipend which was due them, but
also deprived their bodyguards of the sustenance they required.
He writes: "hi sunt, qui invident . . . stipendia paludatis . . .
praetorianis sportulas. . . ,"13 Of course, Sidonius is exaggerat-
ing the degeneracy of his enemies. These men surely would be
the last to weaken their positions by neglecting their military
retainers and bodyguards. The sportulae which were due the
bodyguards (whom Sidonius pedantically calls praetoriani) may
be compared to the buccellata from which the buccettarii got
their name. A more prosaic writer might have used a term like
sportularii to describe these guards because they were provided
with sportulae. The military retainers who received the stipen-
12
L. Musset, Les invasiones, 2nd ed. (Paris, 1969), pp. 114ff, 247ff;
E. A. Thompson, "The Barbarian Kingdoms in Gaul and Spain," Notting-
ham Mediaeval Studies, VII (1963), 9. L.B. (prim, const.}, no. 5; Lex Rom.
Burg., XLV, 3; and V. Eptadii, ch. 12, for the Roman force in the Bur-
gundian army. Procopius, H.W., V, xiii, 3, for the use of Burgundian mili-
tary forces by the Merovingian kings.
13
Sidonius Apollinaris, ep. V, 7, 3. On Sidonius see C. E. Stevens, Sidonius
Apottinaris and His Age (Oxford, 1933).
23
dia paludatis were probably former Roman soldiers who con-

tinued to wear at least part of their uniforms — namely, the
cloak (paludamentum). Procopius, writing almost a century later,
noted that even in his day the descendants of erstwhile Roman
soldiers in Gaul still wore their uniforms. During the later Ro-
man empire imperial military officers often became local mag-
nates and used their troops in essentially private capacities.
These magnates were neither senatorials nor barbarians; never-
theless they had considerable influence and power.14
The magnates of the Burgundians drew their armed followers
from diverse sources. Slaves were used as well as non-Burgun-
dian barbarians. Sigismir, a Burgundian prince, supported a
large armed force which included Franks among his horsemen.
Whereas some of Sigismir's socii were Franks of no noteworthy
rank, others of his socii were reguli or kinglets with socii of their
own. This points to a command structure among the armed
retainers of at least one important magnate.15
The Burgundian monarchs, in addition to their personal armed
followers, used mercenaries to garrison the forts and walled
cities of their kingdoms. Such a garrison served at Vienne in 502
during Clovis's invasion of the Burgundian kingdom. A part of
this garrison included Frankish mercenaries while another, com-
posed of archers, may have been what remained of an imperial
regiment.16 Among the more important fortified places in the
Burgundian kingdom were Lyons, Avenches, Geneva, Grenoble,
Vienne, Langres, Dijon, Autun, and Besan9on, which also had
two regiments of Roman cavalry (milites). Milites as well as
laeti were stationed at Langres, and there were also probably
laeti at Autun. In addition to the laeti and milites, who as
descendants of military colonists and regular soldiers garrisoned
the fortified places of the Burgundian lands, there were a num-
ber of Burgundian magnates who received land from the king
" See MacMullen, Soldier and Citizen, ch. 6, and ch. I, n. 43, above.
"Sidonius Apollinaris, ep. IV, 20, 1, 2, 3; Gregory, Hist., Ill, 5; L.B.,
II, 3, 4; V, 7; X, 1.
M
Gregory, Hist., II, 32, 33; V. Epiphanii, 171; Procopius, H.W., V, xii,
29; xiii, 3.
24
THE SONS OF CLOVIS
in return for their services as garrison commanders similar to
centenarii.17
If the simple Burgundian freeman, because of his legal status,
had ever owed military service in a general levy of some kind,
such an obligation probably had disappeared before the end of
the fifth century when the Burgundian laws were written down,
for they make no mention of this military responsibility. It may
be hypothesized that all able-bodied men in the kingdom could
be called upon to take up whatever arms might be at hand for
the needs of local defense, and it seems equally likely that when
a campaign in which plunder might be obtained was to be un-
dertaken, at least some hardy souls would heed the call of
adventure, though they were not required to do so by either
law or custom. By and large, however, the backbone of the
Burgundian military, as it passed under Merovingian domination
in 534, was not made up of these temporary soldiers, but of
laeti and milites, erstwhile imperial fighting men, and the
armed followings of the royal family and magnates.
After the conquest of the Burgundian kingdom, the Mero-
vingian monarchs might have concentrated their efforts on the
invasion of new lands. Instead, they insisted upon fighting
among themselves. In the same year that the Burgundians yield-
ed to the military might of the Merovingians, Gregory reports,
Theudebert and Childebert agreed to attack Chlotar. Fortui-
tously for the new allies, they learned that Chlotar was traveling
southward through his kingdom; on his journey he had to cross
Childebert's lands. The allies caught Chlotar and his small fol-
lowing off guard on the south bank of the Seine, just across
from the village of Caudebec, about twenty-two miles west-
northwest of Rouen.18 Realizing that he could neither escape
nor defeat his enemies' superior force in an open battle, Chlotar
retreated into the Forest of Arelaunum where he ordered trees
17
Butler, "Late Roman Town Walls in Gaul," pp. 48-50; Gregory, V.P.,
VII, 4; "Cumque milites cum aequitibus praecedentes, cum post terga
traherent vinctum, ad locum ubi confessoris artus quieverant pervenerunl."
Notitia Dignitatem, oc., XLII, 69, 70; XXXVI, 5. Grenier, Manuel d'
archeologte, V, 400, n. 2. L.B., I, especially no. 4.
"Gregory, Hist,, III, 28; L.H.F., ch. 25. Longnon, Geographic, p. 137.
25
to be cut down and barricades to be built; this made his position

less accessible and provided him with additional cover. Childe-
bert and Theudebert surrounded Chlotar's position, pitched
camp, tethered their mounts, which were not of much use in
the heavy forest (explaining in part Chlotar's defensive tactics),
and prepared to attack Chlotar's stronghold in the morning.
However, in the morning a hailstorm accompanied by lightning
and thunder struck the attackers' camp, blowing away their
tents, and scattering their mounts. The disaster demoralized them
to the extent that they broke off the engagement and made
peace.19 Small campaigns like this one seem to interest Gregory
because he is able to credit St. Martin with bringing the storm
against the brothers and saving Chlotar. On the other hand,
more important engagements — the invasions of Burgundy in
523, 524, and 534, Childebert's victory over Amalaric, the Visi-
gothic king, in 531, and the lengthy though unsuccessful siege
of Saragossa undertaken by Chlotar and Childebert in 542 —
receive comparatively little notice and are described with scarce-
ly any military details.20
The conquest of Burgundy in 534 had brought the Meroving-
ian monarchs into closer contact with Italy. During the reigns
of the Austrasian monarchs Theudebert and his son Theudebald,
campaigns were undertaken in Italy in 539 and 554; these re-
ceived little attention in the Merovingian sources, but were
much publicized by the Byzantine authors Procopius and Aga-
thias. In 539, Theudebert, aware that the Ostrogothic and By-
zantine armies were hard pressed in their war, decided to lead
an expedition into Italy. Both the Gothic and imperial com-
manders believed Theudebert to be on their side and no effort
was made to stop the advance of his army, extravagantly esti-
mated at 100,000 fighting men by Procopius. Both the Goths
and the Byzantines welcomed Theudebert's army and were cut
down in turn. After these intial successes and the acquisition of
" Gregory, Hist., Ill, 28.
w
lbid., Ill, 6, 10, 11, 29; Isid., Hist. Goth., DLVIII; Chron. Caesaraug.,
s.a., 541; Jordanes, Get., Iviii, 302. Thompson, The Goths in Spain, pp.
12-15.
26
THE SONS OF CLOVIS
a considerable amount of booty, lack of supplies and disease
weakened the invaders and they retreated north into Gaul. The
lack of an organized system of supply defeated Theudebert's
army where the armies of the empire and the Ostrogoths could
not.21
The campaign of 554 differed in many ways from that of
539. Theudebald was a weak and sickly monarch, nothing like
his father. Where the latter himself had led the earlier Italian
invasion, Theudebald entrusted the campaign of 554 to a pair
of Alaman magnates, Buccelin and Leutharis, who Agathias
exaggeratedly states led a force of some 75,000 Franks and Ala-
mans into Italy. After taking Parma and Pavia, the invaders
moved toward Rome, where they divided into two groups. One,
under the command of Buccelin, bypassed Rome and went
south to Messina; the other force, under Leutharis, went
through Apulia and Calabria to Hydruntum. Both chiefs are
said to have collected a huge amount of booty, some of which
was sent back to Gaul. On the return to Gaul, Leutharis's troops,
forced to fight on foot, suffered severe setbacks from the Roman
garrison at Pisaurum and lost much booty. Continuing north,
Leutharis's forces camped at Ceneta where a plague broke out
killing many of the invaders, including their commander. Buc-
celin returned to the north by way of Campania. While camping
at Casilinum, he engaged in a battle against an imperial army
under Narses. Buccelin's forces, fighting on foot, were trapped
in a crossfire by Narses's mounted archers and were destroyed.22
In this campaign there was another noteworthy battle in
which Narses destroyed a small group of invaders. This band,
composed about equally of infantry and cavalry, deployed in a
phalanx in a forest and faced Narses's cavalry. Narses ordered a
charge, and his horsemen, after some contact, feigned a dis-
orderly retreat. The Merovingian troops broke ranks and pur-
21
Procopius, H.W., VI, xxv, 11; Gregory, Hist., Ill, 32. See Bernard S.
Bachrach, "Procopius, Agathias, and the Frankish Military," Speculum, XLV
(1970), 435-441, and the appendix below, pp. 131-138.
22
Agathias, Hist., I, 6; II, 2, 8, 9; Gregory, Hist., Ill, 32. J. B. Bury, His-
tory of the Later Roman Empire (New York, 1958), II, 277ff; Bachrach,
"Procopius, Agathias," pp. 435-441, and the appendix below, pp. 131-138.
27
sued them, but the Byzantines turned their horses and slaugh-
tered the scattered infantry, winning the day. The Merovingian
horsemen, however, fled to safety.23
Apart from these invasions of Italy, only two other campaigns
of this period — Chlotar's campaigns against the Saxons and
against his rebellious son, Chramn — are described in the sources
in sufficient detail to be useful. After Theudebald's death in
555, Chlotar had inherited his lands. While making a progress
through his newly acquired Austrasian kingdom in the follow-
ing year, Chlotar learned that the Saxons had rebelled again
and so he moved against them. When he approached their bor-
ders, the Saxons sent envoys seeking peace. Chlotar was willing
to reestablish peace, but the Austrasian Franks in his entourage
would not hear of it and attacked Chlotar. Gregory suggests
that they would perhaps even have killed him had he refused
to fight. The Franks' bellicosity, however, seems to have been
more manifest in counsel than on the battlefield, for when they
met the Saxons on the field they were decisively defeated.2*
Chlotar's most pressing problem, however, seems to have been
his son Chramn, who led a warband of mounted followers and
was served by two officers, Scapthar and Imnachar, whom Greg-
ory describes as the leaders of his bodyguard. Chramn made his
headquarters at Clermont, where he had driven out the Count
Firminus. He harassed the area by seizing women and terror-
izing Bishop Cautinus, who kept a saddled horse at his side to
escape being ridden down by Chramn's followers,28 Chramn's
activities led him to abandon Auvergne and to move to Poitiers,
where he made his new headquarters. Shortly thereafter, he
allied with his uncle Childebert with the aim of going against
his father in force. Chlotar finally acted and sent his sons Chari-
bert and Guntram against Chramn, who was at that time en-
camped at Black Mountain in the Limousin. Negotiations proved
futile and the opponents prepared for the battle with "magno
Âgathias, Hist., I, 21. Bernard S. Bachrach, "The Feigned Retreat at
Hastings," pp. 265-266.
* Gregory, Hist., IV, 14.
35
Ibid., IV, 13.
28
THE SONS OF CLOVIS
armorum apparatu," but when both forces were about to en-
gage, a storm broke with thunder and lightning and each side
returned to camp. Chramn then forged a message to his brothers
saying that their father had been killed in battle against the
Saxons. Charibert and Guntram mounted and left with all
speed, returning to Burgundy. Chramn followed with his force
and moved against Chalons-sur-Marne, which he took after a
siege. He continued on to Dijon, but did not enter, and then
proceeded to Paris where he met with Childebert.26
Chramn continued his operations against his father by moving
against the governor of Theiphalia, while Childebert seized
Rheims. Shortly thereafter, however, Childebert died and
Chramn, unable to come to terms with his father, fled to Brit-
tany where he found an ally in Count Chanao. Chlotar moved
his forces into Brittany and a battle was fought against Chramn
and the count. Chanao fled the field soon after the battle had
begun, and Chramn, as he retreated to the ships which he had
prepared for his escape in case of disaster, was captured and
killed.27
Chramn's activity in 556 illustrates not only the mobility of
his forces, but their ability to take a fortified position. This may
have been due to the military equipment which his troops, like
those of his brothers, carried with them. All three princes made
their home in, and presumably drew their forces from, the
more Romanized parts of Gaul — suggesting the influence of
imperial military traditions on Merovingian warfare.
Chramn's activities and those of the pretender, Munderic, are
described in considerable detail by Gregory, who in doing so
indicates the role which personal armed followings played in
the struggle for power. These personal armed followings differed
from each other just as did the men who led them. Chramn, a
prince of royal blood, controlled a comparatively large, well-
organized personal following, commanded by officers, mounted,
and armed with an abundance of sophisticated military equip-
ment. This force constituted nothing less than a private army.
86
Ibid., IV, 16,17.
"Ibid., TV, 18,20.
29
Munderic's group, on the other hand, was not organized under

a command structure, but was a more amorphous unit of sworn
men who nevertheless fought with epic spirit. Arigisel also had
armed supporters (although surely not of the magnitude of
Chramn's band), as did another of Theuderic's retainers, a bar-
barian (Frank?) who directed certain royal estates for the breed-
ing and raising of horses near Treves. Sigivald, whom Theuderic
provided with land in return for his garrisoning the Auvergne
after the revolt of 531, had an armed following with which he
plundered the area he was required to protect. Gallo-Romans,
as noted above, also had personal armed retainers, which they
employed to gain various ends. For example, Syagrius, a member
of the famous noble family which flourished in Gaul during the
late empire and early Middle Ages, used his retainers to help
him liquidate his enemy Syrivald, the son of Sigivald. Men like
Sigivald, Syagrius, Munderic, and Bishop Cautinus were land-
owners, but possessed neither the stature nor the prestige of one
such as Chramn. Munderic's claim to royal blood, Sigivald's kin-
ship with Theuderic, and Syagrius's family connections did aug-
ment their importance, but never did they attain an equal foot-
ing with a prince. These men shared a relatively similar position
in society not only because they all controlled armed forces, but
also because of their comparable relations with the humble
peasant, small landholder, and economic dependent whom they
exploited and harassed.28
Loyalty and support from these magnates (if such a term may
be used to describe these landholders with armed followings)
were vital to any monarch who hoped to rule successfully. The
tumult caused by Munderic's attempted usurpation and by
Chramn's activities underscores this point well. Equally illumi-
nating is an event which took place shortly after Theuderic's
death in 533. When Theudebert tried to succeed his father, he
found that his uncles Chlotar and Childebert were planning to
bar his way. Theudebert, however, gained the support of his
leudes with gifts, and with their aid was able to ascend the
28
Ibid., Ill, 15, 16, 35; IV, 12. Bachrach, "Charles Martel," pp. 71-72.
30
THE SONS OF CLOVIS
throne.29 The significance of the leudes has already been sug-
gested in the description of Clovis's victory over Ragnachar in
509. When Clovis went against Ragnachar, he bribed his enemy's
leudes and thus deprived him of his military followers.30 Theu-
debert, unlike Ragnachar, was able to gain the support of his
leudes, and thus prevailed against his uncles. Another example
of the significance of such followers to a ruler concerns Chlo-
tar's son Chilperic, who when his father died in 561, used a part
of the royal treasure which fell into his hands to ensure the sup-
port of those men whom he thought would be of value to his
cause. To strengthen his position still further, Chilperic per-
mitted his supporters (ftdeles) to plunder and seize estates with-
in his kingdom, thus buying their loyalty with wealth that was
not his to give.81 It is probable that the kind of men whose sup-
port ensured Theudebert's succession and whose loyalty Chil-
peric tried to purchase were of that group who controlled bands
of armed followers — men very much like Sigivald and Syagrius,
who when serving in the royal interest could help to keep a king
in power, but when alienated could weaken his position greatly.
Among the Austrasian Franks in particular, important men
of this type asserted their influence in military matters most
conspicuously. When Theuderic wanted to invade Thuringia in
531, he had to go to great lengths to convince these men and
their retainers to follow him. Ultimately they consented and
distinguished themselves in a cavalry charge on the Thuringian
plains. When, in the following year, Theuderic wanted to punish
the Arvernians for their revolt, these Franks with their armed
supporters threatened to desert him and to follow his brothers
29
Gregory, Hist., Ill, 23: "Cumque abissit, Theudoricus non post multos
dies obiit vicinsimo tertio regni sui anno. Consurgentes autem Childeber-
thus et Chlothacharius contra Theudobertum, regnum eius auferre voluerunt,
sed ille muneribus placatis a leodibus suis defensatus est et in regnum
stabilitus."
30
Ibid., II, 42.
21
Ibid., IV, 22: "Chilpericus vero post patris funera thesaurus, qui in
villa Brannacum erant congregati, accepit et ad Francos utiliores petiit
ipsusque muneribus mollitus sibi subdidit." VII, 7: "Guntchramnus vero rex
omnia quae fidelis regis Chilperici non recte diversis abstulerant.., restituit.
. . ." Bachrach, "Charles Martel," pp. 71-72.
31
who were planning an invasion of Burgundy, As has already
been mentioned, it was not until Theuderic promised them an
abundance of booty in gold, silver, slaves, and raiment, as well
as the right to keep everything they seized including the inhabi-
tants, that they followed him into Auvergne. The enormous pow-
er of the Austrasian magnates was also exemplified when Chlo-
tar's Prankish followers threatened to kill him if he accepted a
treaty with the Saxons who were harassing his border in 556.82
Yet another aspect of the personal armed following deserves
notice, that relating to the role of the king's guards. While on
campaign in Thuringia in 531, Chlotar's life was saved by the
armed men (armati) who surrounded his person. Certain guards
(pueri) who normally served Queen Clotild, Clovis's widow,
were assigned by her to protect her nephews, the sons of the
deceased Chlodomer. Arigisel may be considered part of Theu-
deric's guard.33 The identification of such guards with the an-
trustiones or members of the royal trust suggests a force of con-
siderable size and scope. Men of the royal trust not only served
at the king's side, but were also stationed in centenae estab-
lished throughout Gaul. These centenae were used by the Mero-
vingian kings for both colonizing and military purposes. Among
the inhabitants of the centenae were the antrustiones who prob-
ably performed military and paramilitary services required
of them under the command of the centencuius, a royal official
of imperial origin. Since the antrustiones were composed of
laeti and Romans as well as Franks and other barbarians, it is
likely that the military functions of these settlements had some
connection with the military colonies of laeti and other garri-
sons which had been established earlier by the empire.34
32
Gregory, Hist., IV, 14.
53
Ibid., Ill, 7, 18.
34
For a convincing definition of the centena see Wallace-Hadrill, Long-
Haired Kings, p. 193, n. 1, and cf. H. Dannenbauer, "Die Freien im karo-
lingische Heer," Aus Verfassungs- und Landesgeschichte: Festschrift fur T.
Mayer (Lindau, 1954), I, 49-64. MGH Cap., I, 3, 16: "Pro tenere pads
iubemus, ut in truste electi centenarii ponantur, per quorum fidem adque
sollicitudinem pax praedicta observetur. Et quia propiciante Domino inter
nos germanitatis caritas, indisruptum vinculum, custoditur, centenarii inter
communes provincias licentiani habeant latrones persequere vel vestigia
32
THE SONS OF CLOVIS
Any effort to estimate the size of the Merovingian military
organization founders upon the predilection of chroniclers for
exaggeration. In only one text — a copy of a letter from Theude-
bert to Justinian in which he apologizes for not having sent the
emperor the 3000 troops he had promised him during the pre-
ceding year — is a reliable figure probably given.35 This epistle
implies that Theudebert had at his disposal during the late
530s a force large enough so that he could spare 3000 men and
still maintain his own position. In any event, it seems unlikely
that Theudebert believed he could count on the followers of
the ill-disciplined Austrasian magnates whose caprice and greed
made them unreliable in the extreme. It might be suggested,
however, that Theudebert had considered sending a unit from
the Burgundian kingdom, which was reliable and well disci-
plined.38
This picture of the Merovingian military under Clevis's sons
is necessarily incomplete, since the answers to many questions
simply cannot be found. Nevertheless, it can be discerned that
the milites who garrisoned fortifications, the laeti who pro-
tected fortresses and served as antrustiones in centenae, the per-
sonal armed followings of the magnates, as well as other rem-
nants of the later Roman military establishment were militarily
significant.37 The Taifal laeti living in the area around Poitiers
may have been used as garrison troops, though the sources are
mute on the point; in any event, Theiphalia was organized un-
der a dux.SB This use of erstwhile imperial military formations
adsignata minare et in truste quod defecerit, sicut dictum est causa reman-
eat, ita ut continue capitale ei qui perdiderit reformare festinet, tamen
ut latronem perquirat. Quern si in truste pervenerit, medietatem sibi vin-
dicet, vel dilatura si fuerit, de facultate latronis ei qui damnum pertulit
sarciatur." See also Lex Sal., XLII, 1-4.
35
MGH, ep., Ill, 19. F. Lot, I/Art militaire et les armees au moyen Age
en Europe et dans le proche orient (Paris, 1946), I, 78-79, discusses this
letter in some detail.
88
Procopius, H.W., VI, xii, 38, suggests that Theudebert also agreed to
send aid to the Ostrogoths and in fact sent 10,000 Burgundians. Although
the number is most probably exaggerated, it can be assumed that Theude-
bert had it within his power to use Burgundian troops.
" Notifia Dignitatum, oc., LXII, 44.
33
Gregory, Hist., IV, 18, and ch. 1, n. 30, above.
33
was supplemented by the use of Roman fortifications through-

out Gaul.
The utilization of past imperial forms of organization as well
as the descendants of Roman personnel is illustrated by the
situation at Langres. This fortified city, which served as a mili-
tary base under the empire, continued its original function un-
der Clovis and his sons. The Sarmatian laeti who were settled
there during the later Roman empire gave their names to towns
and villages in the area — for example, Salmaise (Latin, Sar-
matia). One of the suburbs (modern Lassois) was known
throughout the Middle Ages as the laeti quarter, Laticensus.
This section of the city was under the direct control of an
official of the king's household. During the reign of Theude-
bert, a Gallo-Roman of distinguished parentage named Valen-
tinus held the position. In addition to the laeti serving at Lan-
gres there were also mounted milites, according to Gregory.89
This notice of milites at Langres is only one of several such
references in the sources to milites, which at times are depicted
as serving under tribuni.*0 Other Roman influences are sug-
gested by Gregory's cryptic passages concerning military equip-
ment which was used to great advantage by both Theudebert
and Chramn. And it can perhaps be ascertained that Roman
elements lingered to affect Merovingian naval organization
and provided the men and ships which defeated the Danes in
515. In the general area of this encounter, there had been an
W
V. Valent., cli. I: "Beatus Valentinus in Laticensi suburbano Lingo-
nensium oriundus fuit: parentibus nobilibus ortus, trahens ex paterni generis
sanguine originem a Romanis. Cujus pueritia futuram virtutem jam tune
praeclara indole depingebat. Peridem tempus Theodebertus Rex una cum
fratribus Francoram regmim potenter regebat: sub quo Palatinam militiam
praedictus vir in adolescentia pro dignitate parentum administravit. . . ."
Gregory, V.P., VII, 4: "Cumque milites cum aequitibtis " Longnon, Geogra-
phle, p. 101, n. 4, identifies Lassois and defends the reliability of the V. Valent.
as a historical source. See also Edgard Boutaric, Institutions militaires de la
France want les armies permanentes (Paris, 1863), p. 66. Notitia Dignita-
tum, oc., XLII, 69: "Praefectus Sarmatarum gentilium, Lingonas." On
place names see L. Berthoud and L. Matruchot, Etude Historique et £tymo-
logiqve des noms de lieux habites du departement de la Cdte-D'Or (Semur,
1901), I, 2.
40
V. Genoa, eh. 36, and ch. I, n. 5, above.
34
THE SONS OF CLOVIS
imperial naval base during the fifth century.41 Whether Roman
or not, the fleet which served Theudebert in northern Gaul
and the ships prepared by Chramn were not of Prankish origin.
The Prankish role in Merovingian military organization
during the reign of Clovis's sons, although overestimated in
most scholarly accounts, should nevertheless not be underesti-
mated, even if by such underestimation a valuable corrective
to the popular view might be supplied. Prankish units seem
to have had an important function in Austrasia, the least Ro-
manized region of Gaul. Large numbers of Franks played a
significant part in the Italian campaign of 539, and Franks as
well as Alamans served in large numbers in the Italian venture
of 554. The descriptions of the Franks who participated in these
transalpine operations indicate that they were even less Roman-
ized than were the Franks of Austrasia. Many of the Franks
used in the Italian campaigns were probably recruited by the
Austrasian monarchs from beyond the Rhine. In doing this
Theudebert and Theudebald seem to have initiated a policy
which was developed during the next generation by Sigibert,
their successor in Austrasia.42
The diverse development of the Merovingian military organi-
zation, which had been manifest under Clovis, continued along
the same lines under his sons. The heterogeneous ethnic compo-
sition of the military forces, including Franks, Alamans, Gallo-
Romans, descendants of imperial military personnel, Alans,
Saxons, and Taifals, was expanded by Clovis's sons with the
inclusion of the Burgundians and their military establishment.
The complexity of this organization as it passed into the hands
of Clovis's grandsons defies simple characterization, although its
Roman elements are worth emphasizing in light of previous
scholarly neglect.
a
Notttia Dignitatum, oc., XXXVIII, 8: "Praefectus classis Sambricae, in
loco Quartensi sivi Hornensi." Grenier, Manuel d'archSologie, V, 390ff.
"Gregory, Hist., IV, 49, 50. Cf. A. Cameron, "Agathias on the Early
Merovingians," Annali detta Scuola Normale Superiore di Pisa, XXXVII
(1968), 129ff, 136-140. Appendix, n. 7, for the trans-Rhenish origin and
the primitive nature of the Merovingian armies in the Italian campaigns.
35
CHAPTER III
Clovis's Grandsons:
561-593
W HEN Chlotar died in 561, Merovingian military activity
continued under the leadership of his sons Chilperic, Sigi-
bert, Guntram, and Charibert, each of whom received a share
of Chlotar's kingdom. Although the period of their reigns was
plagued by internal fighting, the military organization begun
by Clovis increased in power and complexity with the intro-
duction of a new fighting force — the local levy.
561-575
The year after Chlotar's death, when Sigibert was called upon
to defend his eastern border against the Avars, Chilperic in-
vaded the western part of his kingdom, taking possession of
Rheims, Soissons, and several other cities which Gregory does
not name. Upon successfully driving out the Avars, Sigibert
moved west and retook these cities; at Soissons he captured
Chilperic's son Theudebert, who had been left there to com-
mand a garrison. This incident touched off the first spark that
eventually flared up into the bella dvilia which were to char-
acterize the reigns of Clevis's grandsons.1
Approximately four years after Sigibert defeated the Avars,
these barbarians once again attacked the Merovingian kingdom.
Sigibert went to meet them in battle, but in this encounter he
1
Gregory, Hist., TV, 23. Wallace-Hadrill, Long-Haired Kings, p. 195.
36
CLOVIS'S GRANDSONS
was not as fortunate as he had been in 562. According to

Gregory, his forces became so frightened by fantastic shapes
which the Avars made to appear before them through the use
of magic that they fled the field. Sigibert, abandoned by his
army, arranged a treaty with the Avar leader after giving him
rich presents. Because of this treaty, the Avars made no more
attacks against Sigibert's lands during his lifetime.2 Gregory,
desiring to protect the reputation of Sigibert, to whom he was
partial, contends that the treaty may be counted to his credit.
It is probable that Sigibert's treaty called for a yearly tribute
to be paid to the Avars, and that his forces were not capable
of thwarting a concerted Avar effort. In short, Sigibert bought
peace when he had little or no hope of winning it.
In 568 Sigibert expanded the bella civilia by attacking his
brother Guntram. According to a provision in Chlotar's will,
the city of Aries was divided between Sigibert and Guntram,
and now the former wanted all of it. He ordered Count Fir-
minus of Clermont to lead the Arvemians against Aries and
to take control of the city. (This order is the earliest extant
reference in the Merovingian sources to a local levy.) Sigibert
also commanded a certain Audovar to lead a force from the
other direction which seems to have been raised or stationed
in Provence (perhaps near Aix) against Aries. Both leaders en-
tered Aries unopposed, and they proceeded to take oaths of
fealty to Sigibert from the inhabitants. As soon as Guntram
learned that Sigibert's forces had seized Aries, he sent the
partician Celsus with a force to retake the city. On the way to
Aries, Celsus took Avignon without a battle. Arriving at Aries,
he surrounded the city and commenced his attack. Sigibert's
officers were convinced that they could not defend the city
against a siege so they sallied forth in the hope of routing Cel-
sus's forces. Sigibert's troops, however, were soundly defeated,
and when they tried to reenter Aries, the gates were locked
and the inhabitants hurled stones on them from the walls. With
all hope of victory lost, they abandoned their mounts and tried
3
37
to swim across the Rhone. A large number of Arvernians were

lost, including many noteworthy citizens. Firminus and Audo-
var were captured but were later released unharmed.3
The year before this animosity between Sigibert and Guntram
developed, their brother Charibert had died, and his kingdom
was divided among his brothers. Although little is known of
Charibert's military campaigns, Gregory does discuss in some
detail one of his paramilitary operations because it involved the
alleged intervention of St. Martin. Greogry relates that Chari-
bert, upon learning that the Church of Tours had illegally
come into possession of a royal estate where horses were bred,
dispatched a group of mounted and armed pueri from his court
to take over the estate and to protect the wranglers and grooms
he needed there. Charibert's adventure proved unsuccessful,
however, because, according to Gregory, St. Martin routed his
men.4
With the division of Charibert's kingdom, Sigibert obtained
cities in both Neustria and Aquitaine — cities Chilperic coveted.
Probably in 569 Chilperic took Tours, which had been part of
Sigibert's share, and left the city under the command of his
son Chlodovech. In Poitiers he induced important men like
Basilius and Sighar to join his cause; they either gave Chilperic
their support with armed forces of their own or perhaps they
collaborated with royal followers (pueri regis).5 Even though
Sigibert and Guntram had contested the control of Aries, they
allied in the following year to stop Chilperic's attacks. This was
accomplished with the help of Mummolus, who at the death
of Amantus, Celsus's successor, had become patrician. In 570
3
Ibid., IV, 30: "Sigyberthus vero rex Arelatensim urbem capere cupiens,
Arvernus commovere praecipit. Erat enim tune Firminus comes urbis illius,
qui cum ipsis in capite abiit. Sed et de alia parte Audovarius cum exercitu
advenit." Longnon, Geographic, p. 147, for Aix belonging to Sigibert. The
words "alia parte" would seem to point to this part of Gaul as contrasted
to Clermont.
* Gregory, Virt. S. Mart., I, 29. It is curious that Charibert was buried at
Blavia castellum, one of the Roman fortifications in the Aquitanian part of
the Tractus Armoricani. The town was of military importance only. See
L.H.F., ch. 31.
6
38
CLOVIS'S GRANDSONS
Mummolus was able to move west, driving Chlodovech from

Tours and suppressing opposition at Poitiers. As Firminus and
Audovar had done in Provence, Mummolus exacted oaths of
fealty to Sigibert from the people of Tours and Poitiers.6
In the years immediately following this episode, the Mero-
vingian kings undertook two extended series of campaigns.
Guntram's forces, led by Mummolus, fought four campaigns
in the course of as many years. In 571, hearing that the Lom-
bards had entered Gaul and had advanced to the Plan de
Fazi near Embrun, he assembled his forces and trapped the
Lombards in the forests by blocking the roads with trees. Once
he had nullified the effectiveness of the Lombard horsemen
in this way, he attacked them in the woods and cut them to
pieces. Those who survived and could not escape were taken
prisoner and sent to various strongholds in the Burgundian
kingdom where they were kept under guard.7
A year after this massacre, a large band of Saxons who had
entered Italy in 569 as Lombard allies raided Guntram's lands
and encamped at the villa of Stable in the territory of Riez.
When Mummolus learned of their invasion, he moved against
them and was able to kill large numbers because they had di-
vided into small groups in order to plunder the countryside.
This slaughter continued until nightfall when under cover of
the ensuing darkness the remaining Saxons slipped back into
their camp. By morning they had regrouped at Stable and
faced Mummolus's forces. A showdown was averted, however,
when a peace treaty was arranged. This treaty allowed the
Saxons and their families to settle in Sigibert's kingdom. The
warriors therefore returned to Italy for their women and chil-
dren. They reentered Gaul in 573 in two groups, one proceed-
ing through Nice and the other through Embrun. They re-
grouped at Avignon where they plundered the crops. However,
when they prepared to cross the Rhone and enter Sigibert's
* Ibid., IV, 42.
''Ibid.
39
kingdom, Mummolus barred their way and forced them to pay

for what they had stolen and destroyed.8
In 574 Gaul was invaded once again, this time by a Lombard
force under the command of Dukes Amo, Zaban, and Rodan.
At Embrun the Lombards deployed into three groups, with
Amo leading one group south through Gap and on towards
Avignon, camping at Mummolus's villa at Macho. Amo does not
seem to have taken Avignon, but he subdued Aries and raided as
far as Marseilles. At Aix he collected a tribute of twenty-two
pounds of silver and withdrew. Zaban also went south through
Gap, but then turned north through Die and on to Valence,
which he besieged. Rodan traveled north and west from Embrun
and camped near Grenoble to which he laid siege. When Mum-
molus learned of the Lombard invasion, he immediately moved
to attack Rodan at Grenoble. After crossing the Isere, Mummo-
lus's troops engaged the Lombards and defeated them handily.
Rodan and his remaining followers fled south to Valence through
the mountains and forests of the Dauphine. Mummolus did not
pursue them but, taking advantage of his interior lines of trans-
portation, rode east along the Roman road from Grenoble to
Embrun in order to cut off the Lombards' line of retreat into
Italy. Along the way Mummolus seems to have gathered a multi-
tude of men from the countryside; with his forces thus enlarged,
he awaited the appearance of Rodan and Zaban. As Mummolus
had expected, the Lombards reached Embrun and there he dealt
them a crushing defeat; only a small number managed to escape
and return to Italy. Amo, who had remained in the south and
probably spent much of his time in Sigibert's lands of Provence,
escaped Guntram's patrician, but was trapped in the Alps by the
snows and almost perished.9
Gregory's accounts of Mummolus's activity against the Lom-
bards and Celsus's successes at Avignon and Aries offer useful
insights into the command structure and military operations of
the Merovingian-ruled Burgundian kingdom at this point in his-
8
Ibid,
9
Ibid., IV, 44. On the roads see Grenier, Manuel d'archeologie, VI, fig.
48 and pi. 1.
40
CLOVIS'S GRANDSONS
tory, as well as further evidence of Roman influence on these

early medieval institutions. Both Mummolus and Celsus had at
their command a standing force capable of rapidly moving into
action, rather than a group of warriors which had to be raised
(collectus or congregatus). Gregory, after learning that the
Lombards had reached Embrun, claimed that in Mummolus's
first campaign he "exercitus movit, et cum Burgundionibus illus
proficiscetur." Of the battle against the Saxons, he writes: "Quod
cum Mummolus conperisset, exercitum movet," and of the last
Lombard campaign: "Qui Mummolo perlata fuissent, exercitum
movit." The same conclusion can be drawn from the actions of
Celsus's troops when Guntram ordered them to retake Aries as
soon as he learned that the city had fallen to Sigibert ("Quod
cum Gunthramnus rex conperisset, Celsum patricium cum exer-
citu illuc dirigit") .10
The military descriptions offered by Gregory also suggest that
Celsus's and Mummolus's forces as well as their structure of
command may have been of Roman origin. The rank of patrician
held by these commanders seems to have been the title given to
the military leaders of the old Burgundian army (many of whom,
like Celsus and Mummolus, were Gallo-Romans).11 As already
noted, the Burgundian monarchs had integrated Roman milites
stationed in their realms into the Burgundian military. And these
same milites and their descendants may well have formed a
standing army (like that of Celsus and of Mummolus) which
functioned in the Burgundian kingdom under Merovingian domi-
nation and was capable of rapid military response.
Shortly after the Lombard invasion of 571, Guntram and Sigi-
bert were again at odds. Chilperic, aware that his brothers' alli-
ance was in jeopardy and that, even if the rift were healed, a
significant part of Guntram's forces would still be involved
against the Lombards, moved once more against Sigibert. Chil-
peric's son Chlodovech established himself at Bordeaux, but not
10
Gregory, Hist., IV, 30, 42, 44.
11
Ibid., IV, 14, 42, for Agricola, Celsus, Amatus, and Mummolus, all
patricians of Gallo-Roman origin. The patrician as a military commander
should not be confused with the patrician who was governor of Marseilles
(ibid., IX, 22).
41
long after Sigulf, one of Sigibert's supporters, drove him out.

Chilperic's son Theudebert was more successful: he took control
of Tours, Poitiers, and other cities of Sigibert's south of the
Loire. At Poitiers, Theudebert defeated Sigibert's Duke Gundo-
bald, drove his army from the field, slaughtered a large number
of the city's inhabitants, and compelled the city to submit to
Chilperic. He then plundered the area around Limoges and
Cahors.12
Sigibert did not immediately respond to Chilperic's attacks.
In 574, however, he gathered a force of pagan Germans from
beyond the Rhine and led them against his brother. Fearing the
oncoming attack of Sigibert, Chilperic arranged a treaty with
Guntram whereby the latter would stop Sigibert from crossing
the Seine and reaching his territory. But by threatening to move
against Burgundy, Sigibert pressured Guntram into allowing him
to cross the Seine. Chilperic retreated before Sigibert's advance,
refused to fight, and surrendered as the price for peace all that
Theudebert had taken. Sigibert, however, had difficulty con-
trolling the undisciplined warriors from beyond the Rhine and
they plundered the inhabitants and churches of the Paris area.13
In the following year, Chilperic arranged another alliance with
Guntram and moved his army against Sigibert, devastating his
brother's territory up to Rheims. When Sigibert learned of Chil-
peric's advance, he again called upon trans-Rhenish fighting men
and led them to Paris with Chilperic retreating before him. From
Paris Sigibert ordered the counts of Chateaudun and Tours to
lead their local levies against Theudebert. When the counts
ignored his orders, he sent the Dukes Godegisel and Guntram
Boso to call out the levies. They did so and moved against
Theudebert, defeating his army and killing him. Guntram broke
his alliance with Chilperic who, realizing that he could not
stand alone in the field against Sigibert, retreated to Tournai
where he prepared to withstand a siege. Sigibert's forces took
Chilperic's territory west of Paris up to Rouen, where he was
M
Ibid., IV, 47.
13
Ibid., IV, 49: " . . . Sigyberthus rex gentes illas quae ultra Renum
habentur commovit. . . ."
42
CLOVIS'S GRANDSONS
raised to the kingship of Chilperic's lands by the latter's erst-

while supporters. No sooner had this taken place, however, than
two armed retainers of Queen Fredegund, wife of Chilperic,
murdered Sigibert and wounded two of his followers, the Cham-
berlain Charegisel and the Goth Sigila. With this, the first phase
of the bella civilia came to an end.14
This phase of the bella civilia amply illustrates the diversity
of the units that formed the Merovingian military. Sigibert's
main military support seems to have come from pagan Germans
living beyond the Rhine. His Austrasian subjects had demon-
strated during the reigns of his uncle Theuderic and his father,
Chlotar, their unwillingness to obey their king and their dislike
for fighting outside their own territory. Early in his own reign,
they had further proved their ineffectiveness against the Avars.
Sigibert found the Austrasians of little help in his wars with
Chilperic, and this perhaps accounts for his inability to react to
Chilperic's attacks before he had arranged for the support of
trans-Rhenish troops in 574, It should be recalled that between
568 and 573 Chilperic attacked Sigibert's lands at least twice, but
that the latter on his own made no counterattack before 574. In
570, however, Guntram had sent an army to recover for Sigibert
what Chilperic had taken from him.
Sigibert's local levy of the Auvergne was loyal, but no match
for a large army. As has already been mentioned, his counts at
Tours and Chateaudun appear to have been disloyal and did not
call up their levies when ordered to do so. These local levies
only came into action when Sigibert provided them with new
commanders. A Frank named Sigulf led a force in defense of
Sigibert's cause at Bordeaux which proved to be of sufficient
strength to drive out Prince Chlodovech's following; this force
suggests a garrison which may have had its origin as a centena
established in 508 by Clovis. Sigibert's troops in the Poitiers
area, commanded by Duke Gundobald, were neither a local levy
nor a popular militia. Rather, they may have been composed of
Taifals who, as laetl established in the area, had been organized
" Ibid., IV, 50, 51.
43
under a duke and integrated into the Merovingian military

during the reign of Childebert, if not earlier.15
Sigibert's problems with the counts of Chateaudun and Tours
amply demonstrate the necessity of having the loyal support of
the magnates. The importance of loyal magnates to the func-
tion of royal power is further illustrated by what happened when
Godinus, a follower of Sigibert, defected and joined King Chil-
peric, who gave him lands from the royal fise. Gregory, who
tells the story, notes that as a result Sigibert went to war against
Chilperic. Godinus did not perform as had been expected of him
while serving Chilperic in a military capacity, and therefore
Chilperic deprived him of the lands previously granted to him
and took his life as well. It may be wondered why Chilperic
bought Godinus's support: Did Chilperic desire Godinus's loyalty
for personal military service only, or because he was an impor-
tant figure (one whose defection might cause a war) who led a
personal armed following? Did he, for example, have a following
like that of Titus who, as the leader of a band of bucellarii, was
of such value that the Emperor Leo invited him to come to the
East and made him a count? It seems reasonable to assume that
Godinus was such a man, especially in light of the consequences
of his defection and of Chilperic's willingness to buy his sup-
port with lands from the fisc. Men of military worth often may
have received such grants. Mummolus received estates from the
royal fisc to help support himself and his followers, as did Sigi-
vald when he was ordered to hold the Auvergne for Theuderic.16
Chilperic seems to have operated with three field forces dur-
** On the garrison of Bordeaux see ch. I, n. 29, For the levies of Poitiers
referred to by Gregory, Hist, V, 26: ". . . Pictavi . . . ex iussu Chilperici
regis abierunt"; VII, 2: "Berulfus vero dux cum . . . Pectavis. . . ."; and
VI, 26: "Pectavis super se inruentibus. . . ." In IV, 47, Gregory makes a
clear distinction between the people (populus) and Gundovald's force. On
the Taifal connection see ch. I, n. 30, above.
10
Gregory, Hist., V, 3: "Godinus autein, qui a sorte Sigyberthi se ad
Chilpericum transtulerat et multis ab eo muneribus locopletatus est, caput
belli istius fuit; sed in campo victus, primus fuga dilabitur. Villas vero, quas
ei rex a fisco in territurio Sessionico indulserat, abstulit et basilicae contulit
beati Medardi. Ipse quoque Godinus non post multum tempus repentina
morte praeventus, interiit." For Titus see ch. I, n. 45, and for Sigivald, ch.
II, n. 6. On Mummolus's land see Gregory, Hist,, IV, 44: "... Macho villam
44
CLOVIS'S GRANDSONS
ing this period: one led by himself, another by his son Chlodo-
vech, and a third by his son Theudebert. None of them seems to
have been very large or very effective. Mummolus, Guntram's
patrician, had little trouble defeating Theudebert's force, and
Sigibert captured Theudebert when he was commanding a garri-
son at Soissons. Theudebert finally met his death at the hands of
the local levies of Tours and Chateaudun. Sigibert's horde from
beyond the Rhine did not find it difficult to discourage Chilperic
from fighting, and Chlodovech was driven from Bordeaux by
what may have been a garrison force led by Sigulf. Chilperic's
Franks at Rouen in the western part of Gaul may perhaps have
been a garrison or centena for that area. Unfortunately little can
be ascertained with certainty about the composition of Chil-
peric's forces. They do not seem to have been local levies, but
rather personal followings, such as that which Godinus probably
had and the one which protected Queen Fredegund, known as
pueri reginae.17
The reigns of the three Merovingian monarchs, Sigibert, Chil-
peric, and Guntram, from 561 to 575 amply demonstrate the
importance of diverse military institutions to a ruler's success.
Sigibert's Austrasian Franks were of little military value and he
had to rely upon undisciplined hordes from beyond the Rhine.
His local levies, however, in the old Gallo-Roman cities were
effective. Chilperic, using private warbands, could gain brief
successes but was not able to sustain his efforts against Sigi-
bert's trans-Rhenish warriors or Guntram's disciplined Burgun-
dian forces of Roman origin. The armies of the Burgundian king-
dom, which owed much to imperial institutions, were the most
successful during this period.
575-584
After the death of Sigibert at the hands of Queen Fredegund's
armed retainers, Chilperic seems to have dominated military
Avennici territorii, quam Mummolus munere meruerat regio. . . ." See
Bachrach, "Charles Martel," pp. 71-72.
17
For the pueri reginae see Gregory, Hist., VI, 32. On the centena of
Rouen see Musset, Les invasiones, p. 187.
45
activity in Gaul. By early 576, Chilperic had taken Paris, cap-

tured Sigibert's widow, Queen Brunhild, and seized some of the
dead king's treasure. Brunhild's infant son, Childebert, had been
saved from his uncle in the preceding year by Duke Gundobald,
who took him from Paris to Austrasia. There the boy was
crowned king on Christmas day, 575. While Chilperic was at
Paris, he sent Count Roccolen of Le Mans with the local levy
of that city against Tours. This force ravaged the area, tried
unsuccessfully to capture Guntram Boso (who had led the
local levy of Tours victoriously against Theudebert earlier and
who, after the death of Sigibert, took sanctuary with his family
in the Church of St. Martin at Tours), but finally brought the
city under Chilperic's control. As a part of his efforts to take
possession of western Gaul, Chilperic ordered his son Merovech
and Roccolen to take Poitiers. Merovech, however, disobeyed
his father's orders and went to Tours where he spent Easter.
Roccolen died late in February before accomplishing his task
at Poitiers.18
The striking feature of these activities is that they were carried
out in the dead of winter. It should be emphasized that they
were not defensive actions but offensive campaigns designed
for the conquest of western Gaul. The most effective force seems
to have been the local levy of Le Mans, serving under Roccolen,
but perhaps the most significant aspect of the situation was the
complete inability of Childebert II's followers to organize some
kind of effective defense.19
The disobedience of Merovech and the death of Roccolen
slowed Chilperic's campaign, and he was soon to receive another
blow. After Easter, Merovech went to Rouen where he married
his aunt, Queen Brunhild, his father's enemy. When Chilperic
learned of the marriage, he went to Rouen as quickly as possible,
took Merovech into custody, and brought him to Soissons.20
18
Gregory, Hist., V, 1,2,4.
19
Roccolen was at Tours during Epiphany which was in February of 576.
He had been there for some time before that, having gone to Tours at about
the same time Chilperic went to Paris. See n. 18 above.
20
Gregory, Hist., V, 2.
46
CLOVIS'S GRANDSONS
Brunhild had escaped to Austrasia. By this time (it was late

spring or early summer of 576), at least some of the infant king's
followers were ready to go into action. Duke Lupus of Cham-
pagne, a loyal supporter of Sigibert's house, seems to have been
responsible for organizing a force of men from his territory to
go against Soissons. They managed to drive Fredegund and
Chlodovech out of the city, but Chilperic, who was camped not
far from Soissons with an armed force, came to the rescue and
decisively defeated the men of Champagne.21
After the Champagne offensive, Chilperic became convinced
that Brunhild had turned Merovech against him and he had the
prince tonsured and put into a monastery at Aninsula near
Angers. At the same time, Chilperic sent Chlodovech to take
Saintes, which belonged to Guntram. Saintes fell without note-
worthy resistance, but Guntram retaliated by sending a force
under Mummolus into the Limousin, which he devastated. Duke
Desiderius, who moved against Mummolus, was defeated and
barely escaped with his life.22
Practically nothing can be said with certainty about these
forces which campaigned during the remainder of 576. It might
be suggested that the force which attacked Soissons was a part
of the levy of Champagne, a group whose activities are noted
on several occasions by Gregory.23 It may also be inferred that
Mummolus led a part or all of the standing army of Burgundy,
a force which has been discussed in some detail above (pp. 22-
25). Chlodovech's troops may have been the local levies of Tours
and Angers ("congregate exercitus, in terminum Toronicum et
Andecavum").
In contrast to earlier campaigns, these later ones are curiously
devoid of sieges. Perhaps the unsettled events following Sigi-
bert's death created a situation in which his important followers
were unsure whether to give their support to his son Childebert
or to Chilperic. For example, Sigibert's referendary, Siggo, be-
came a follower of Chilperic, but then switched his allegiance
21
Ibid., V, 3: "... collect! aliqui de Campania. . . ."
22
Ibid., V, 3, 13, 14.
33
See n. 59 below.
47
to Childebert. In this crisis in allegiance there was a lack of

strong commitment to defend Childebert II's cities against Chil-
peric's attacks — which would explain in part the lack of sieges
during this time. Sigulf, who commanded Sigibert's garrison at
Bordeaux, tried to seize control of the city for himself, but lost
it to Chilperic's forces and then completely deserted Childebert's
cause. This kind of tergiversation seems to have been reasonably
common among Sigibert's erstwhile followers. Gregory also notes
that Ciuciolo, Sigibert's former count of the palace, and Bishop
Praetextatus, instead of supporting Childebert II, joined forces
with Merovech who was a rallying point for magnates seeking a
profitable connection.24
In late November or early December of 576, Merovech es-
caped from the monastery of Aninsula and went to Tours where
he joined Guntram Boso. Each stayed in sanctuary at the Church
of Tours with his armed followers. Count Leudast of Tours was
foiled in his efforts to capture or kill them, though he did on
one occasion manage to kill some of Guntram Boso's followers.25
After spending about two months at Tours, Merovech and Gunt-
ram Boso left the city with some 500 men to go to Austrasia.
Meanwhile, Chilperic learned that Merovech had taken sanctuary
at the Church of Tours and ordered Gregory to cast him out.
When the latter refused, Chilperic led an army against the city.
This force, arriving after Merovech had departed, ravaged the
area around Tours.26
While Merovech was passing through Auxerre in the dead of
winter, presumably with his following, he was captured by
King Guntram's Duke Herpo, but was released. Chilperic with
his forces hunted throughout Champagne for his son but could
not find him. Merovech joined Brunhild but was kept from
remaining with her by the magnates of Austrasia; he then went
into hiding in Champagne and remained there until some mag-
nates of the Therouanne area invited him to their territory,
indicating that they preferred his rule to that of his father. But
24
Gregory, Hist., V, 2, 18; VII, 9.
26
Ibid., V, 14.
"Ibid.
48
CLOVIS'S GRANDSONS
when he arrived at the town of Therouanne with a chosen group

of his followers, they were surrounded in their quarters by a
force of armed men who sent word to Chilperic that his son
was trapped. Merovech realized the hopelessness of his position
and ordered one of his followers, Gallien, to kill him; the rest
were taken prisoner by Chilperic, who had them executed, some
by torture.27
During the summer of 577 and before the death of Merovech,
King Guntram and King Childebert II made a pact to stop
Chilperic's conquest of their territories. In the autumn of the
same year, Guntram Boso, apparently acting on his own, led a
small group of his armed followers (armati) into Tours and,
despite the opposition of the count, carried off his daughters
from the church there and brought them to Poitiers.28
Shortly thereafter, Chilperic, undeterred by his brothers' alli-
ance, attacked Poitiers, drove out Childebert II's garrison, de-
posed the count, and took control of the city. When Guntram
Boso learned that Poitiers was under attack, he left his daughters
at the Church of St. Hilary and fled from the city. The next
year, on his way back to Poitiers to get his daughters, Guntram
Boso encountered Chilperic's Duke Dragolen leading an armed
force, and fighting ensued. Dragolen spurred his horse in a pell-
mell charge against Guntram who unseated his opponent with a
lance thrust to the throat. While Dragolen was hanging from his
mount, cne of Guntram's men killed him with a second lance
thrust. Dragolen's followers were then put to flight and Guntram
gathered up his daughters and went home.29
The activities of such magnates as Guntram Boso could cause
no end of trouble for the monarchs. Perhaps this is best demon-
strated by the career of Count Leudast of Tours. Gregory, who
was Leudast's contemporary and neighbor, knew a great deal
about him and discusses his activities in detail. Gregory's account
of Leudast illustrates not only the relative power that a magnate
might wield, but the diversity of the organization of armed force
27
Ibid,, V, 14,18.
s
"-
28
lbid., V, 17, 24, 25.
Ibid., V, 24, 25.
49
within the Merovingian kingdoms. Leudast became count of
Tours during the reign of Charibert. When that king died and
the city was supposed to pass under Sigibert's control, Leudast
gave his support to Chilperic but lost his position as count. After
gaining control of Tours, Chilperic restored the countship to
Leudast, who, in this position, spent much of his time and effort
in plundering and abusing the people whom he was charged to
protect. Among those he maltreated were soldiers (milites) sta-
tioned in the city, perhaps because they owed their loyalty
directly to the king.30
Leudast's activities eventually incurred the wrath of both Chil-
peric and Fredegund. When they deprived him of his courtship
and determined to eliminate him, Leudast fled to Paris but con-
tinued to plot, causing suspicion to be thrown upon Bishop
Gregory. Duke Berulf, who governed the territory of Tours and
Poitiers, and Eunomius, the new count of Tours, posted guards
(custodes) at the city gates, allegedly to watch Gregory. Mean-
while, Chilperic ordered that no one might receive Leudast into
his house. Leudast returned to Tours, gathered up much of his
treasure, and proceeded to Bourges, the possession of King Gunt-
ram. Chilperic's pueri pursued Leudast, but still he managed to
avoid capture. At Bourges the count and a band of his armed
followers robbed Leudast who barely escaped with his life. Leu-
dast then rallied his followers, some of whom were from Tours,
returned to Bourges, and retook part of his treasure. Berulf had
learned of Leudast's activity and sent some of his own pueri to
seize him. Leudast again escaped, this time to Poitiers where he
took refuge at the Church of St. Hilary. From his sanctuary at
St. Hilary, Leudast led his followers in plundering the people
and property nearby. For these depredations he was expelled
from the church. Shortly thereafter he went to Chilperic, who
was camped with an army at Melun, to seek his pardon; then
he traveled with the king to Paris to seek a pardon from the
queen as well. She refused and sent her pueri to seize him.
30
Ibid., 47, 48: ". . . milites fustibus verberari. ..."
50
CLOVIS'S GRANDSONS
Armed with swords and shields, they wounded and captured
Leudast, who was first tortured and then executed.81
Among the armed forces which were operating during the
episode described above was a band of pueri regis who pursued
Leudast from Tours to Bourges, while the king himself was some
two hundred miles to the north at Bernay. This force may be
compared to one which Chilperic sent to Limoges a short time
later. In March of 579, the people of Limoges attacked Chil-
peric's tax collectors who had to flee to save their lives. Chilperic,
upon learning of this, sent a band of men from his side (de
latero suo) to punish the inhabitants of Limoges. These men
inflicted heavy losses on the people, killing many and exacting
severe reprisals. These pueri regis further illustrate the diversity
of Chilperic's forces as do the milites, stationed at Tours, who
seem to have been responsible to an authority other than that
of the count. This is suggested by Leudast's hostility toward
them and by texts which indicate that milites served under a
tnbunus. A certain Medard served as tnbunus at Tours at about
that time. Also at Tours were men identified as custodes, who
are apparently described by their function, although this may
have been their title as well. It is reasonable to suggest that the
custodes formed some kind of force other than a local levy, for
Gregory speaks specifically of the local levy of Tours, but fails
to indicate in any way that the custodes may have served in this
same capacity. Gregory does mention that Chilperic used men
called custodes to guard strategic positions such as the bridge
across the Orge south of Paris. That local levies were not the
only troops in some cities is illustrated by the milites at Tours
and the distinction drawn by Gregory between these levies and
the garrison at Poitiers.82
Like the magnates under previous Merovingian kings, many
of Chilperic's important supporters—Duke Berulf and Duke
81
Ibid., V, 49; VI, 32.
Îbid., V, 28: "Unde rmiltum molestus rex, dirigens de latere suo per-
sonas, imnensis damnis populum adflixit suppliciisque conteruit, morte
multavit." On the tribunus see Gregory, Hist., VII, 23, and ch. IV, pp. 78-
80. Gregory, Hist., VI, 19, for custodes; for the levy of Tours, IV, 50; V, 26;
VI, 12, 31; arid for the local levy and garrison at Poitiers, IV, 47; V, 24.
51
Dragolen among them—had bands of private armed retainers

which were of crucial military importance. Gregory refers to
Berulf's followers as "pueri" and to Dragolen's as "socii," sug-
gesting the interchangeability of the terms. Childebert's Duke
Guntram Boso had followers whom Gregory calls "armati" and
"amici." Merovech also had a band of armed retainers which
at one time, according to Gregory, numbered some 500 men.
Among Merovech's following were important men like Sigibert's
former count of the palace, and pueri like Gallien. If at the time
Merovech and Guntram Boso left Tours together the latter can
be considered Merovech's follower, it might be concluded that
Merovech had at least one not so loyal supporter with his own
armed band of retainers. The varied status of the men serving
Merovech did not deter Gregory from calling them all — lower
class pueri and magnates alike — viri fortes. The importance of
these armed bands to a monarch, even a relatively successful one
like Chilperic, is illustrated positively by the activities of Drago-
len and Berulf, and negatively by the trouble which Merovech
and Leudast caused the king.33
Local levies also played a significant role in royal attempts to
win campaigns and gain power. For, like the personal armed
followings, the local levies could be either beneficial or detri-
mental to the ruler who needed their support. It was not always
easy to gain their loyalty because their leaders generally sought
to serve on the winning side and could not always be counted
upon to serve their king faithfully. In 579 Chilperic called up
the local levies of Tours, Poitiers, Bayeux, Le Mans, and Angers
to go against the Bretons. He also called up the Saxons of
Bayeux, whose forebears had been settled in the area more than
a century before by the Roman empire as military allies. The
Saxons took severe losses in a sneak attack led by the Breton
leader, Waroch. The Bretons nevertheless were forced to make
peace rather than face Chilperic's full army of local levies rein-
forced by Saxon military colonists. As a result of this peace
33
For Guntram see Gregory, Hist., V, 24, 25; VI, 26. For Dracolen, V,
25; for Berulf, V, 49; and for Merovech, V, 14, 18.
52
CLOVIS'S GRANDSONS
Waroch swore fealty to Chilperic and gave him his son as a
hostage.34
Chilperic's continued efforts to conquer western Gaul led him,
in 581, to order Duke Desiderius to go against Guntram's cities
in the southwest. Desiderius defeated Guntram's Duke Ragno-
vald and seized control of Agen, Perigueux, and Angouleme.
While Desiderius was in the field against Guntram's forces, Chil-
peric's Duke Bladast attacked the Gascons, but was decisively
defeated. Guntram planned to retaliate by sending the levy of
Bourges against Tours, but Duke Berulf called out the local
levy of Tours for defense of the city, and Guntram's forces made
no permanent gains.35 Two years after this successful defense of
Tours, Chilperic again employed the levies in a two-pronged
attack against Bourges, Guntram's stronghold in western Gaul.
Duke Berulf, with the levies of Tours, Poitiers, Angers, and
Nantes, attacked from the north, while Desiderius and Bladast
with the forces from Chilperic's territory in southern Gaul at-
tacked from the south. This army probably included the troops
Desiderius had commanded when he conquered Guntram's
cities two years earlier, the force Bladast had led against the
Gascons, and perhaps the local levies of the cities of Agen, Peri-
gueux, Saintes, Bordeaux, Toulouse, and Angouleme, which were
all at that time under Chilperic's control. A bloody battle ensued
between the local levy of Bourges and the force led by Desiderius
and Bladast. Chilperic's army compelled the local forces to
retreat and they took refuge in Bourges which Chilperic's dukes
then put under siege.36
Meanwhile, Chilperic attacked and burned Guntram's strong-
hold at Melun. But Guntram, leading a force against his brother,
defeated him decisively. Chilperic had to retreat to Paris and
84
Ibid., V, 26: "Dehinc Toronici, Pictavi, Baiocassini, Caenomannici et
Andecavi cum aliis multis in Brittania ex iussu Chilperici regis abierunt et
contra Varocum . . . resedent. Sed ille dolose per nocte super Saxones Baio-
cassinos ruens, maximam exinde partem interfecit." Note Gregory's dis-
tinction between the men of Bayeux and the Saxons of Bayeux. For other
references to these Saxons see X, 9, and Longnon, Geographie, pp. 172—175.
85
Gregory, Hist., VI, 12. Longnon, Geographie, pp. 145ff.
88
Gregory, Hist., VI, 31; IX, 31.
53
order his troops besieging Bourges to withdraw. For Guntram

to take the field in person was very unusual and both this action
and his failure to react sooner to Chilperic's attacks on his terri-
tory were probably due to his lack of a talented commander to
lead his forces, for Mummolus had defected to Childebert II.3T
Although the most striking aspect of Chilperic's forces during
the campaign against Bourges is the considerable number of
local levies which were at his command when he needed them,
it should be noted that Chilperic could muster large forces with-
out calling up the levies. For example, in the autumn of 584
Chilperic sent a military escort, which according to a con-
temporary numbered about 4000 armed men, to protect the im-
mense dowry that his daughter was taking on the road to Spain.38
While Chilperic was securing a more powerful position in
Gaul, his brother Guntram and his nephew Childebert II were
beset not only by his attacks, but by difficulties at home. After
Mummolus, Guntram's talented commander, switched his support
to Childebert, a group of Childebert's magnates, including Mum-
molus, arranged to support a pretender to Guntram's throne. In
582, with the aid of the Byzantine Emperor Tiberius, a certain
Gundovald, who had a long-standing but unaccepted claim to
royal blood, landed at Marseilles with a small following and a
large treasure. Leaving his treasure there, he obtained horses
from Bishop Theodore and joined Mummolus at Avignon. Gunt-
ram Boso, though a party to Gundovald's plot, went to Marseilles
where he arrested Theodore and seized the treasure. This altered
Gundovald's plans and he retreated to an island in the Mediter-
ranean. Soon after, Guntram Boso was taken prisoner by King
Guntram's men and condemned for his part in bringing Gundo-
vald to Gaul. In an effort to clear himself and save his family,
Guntram Boso promised to capture Mummolus. At the head of
the local levies of Velay and Auvergne, Guntram Boso advanced
on Avignon. Mummolus, however, arranged that all the boats
which might be used to take Guntram Boso and his troops across
the Rhone be hidden and that only unworthy craft be made
»IWd.,VI,31.
38
Ibid., VI, 42.
54
CLOVIS'S GRANDSONS
available. Thus when Guntram Boso and his men were in mid-
stream, their boats began to fall apart and sink and many fighting
men were lost in the river. This was not the only stratagem that
Mummolus used to prepare for Guntram Boso's attack, for he
also dug a moat on the eastern side of Avignon into which he
directed water from the Rhone. To make the moat an even more
formidable obstacle, deep pits were dug at random so that al-
though at certain places an attacker might be able to wade, with
his next step he might sink over his head. These tactics did not
prevent Avignon from being besieged, although Guntram Boso
nearly drowned in the treacherous pits in the moat and the life
of a follower (amicus) of his was lost when his attempts to
escape from a pit were foiled by his heavy armor. When Childe-
bert II learned that two of his supporters, Guntram Boso and
Mummulos, were thus engaged, he ordered Duke Gundulf to
bring peace among the king's followers at Avignon, which he
succeeded in doing.39
Although Childebert successfully prevented this incident from
becoming a full-scale war, his youthfulness caused difficulties, as
did the machinations of the magnates. Not only did many of
them connive to bring Gundovald into Gaul, but some worked
secretly for King Chilperic. Others overtly sought aggrandize-
ment at their king's expense. Two such magnates, Ursio and
Berthefred, with their followers, plundered the lands of Duke
Lupus of Champagne, a loyal supporter of the king, and drove
him out of his duchy despite the support he received from the
royal house.40
Nonetheless, Childebert did enjoy a strong measure of sup-
port on at least one occasion from the rank and file (minor
populus) of his forces. This occurred when Bishop Egidius of
Rheims and some of Childebert's other magnates arranged a
treaty with Chilperic and prevailed upon the youth to march in
his uncle's support. The minor populus refused, however, say-
ing that the king's magnates, who allowed his cities to be taken
over and his people to be subjugated, were selling him out to
89
Ibid., VI, 1, 24, 26.
40
Ibid., X, 19; VI, 4.
55
Chilperic. The minor populus attacked Egidius, who barely es-

caped in a precipitous flight in which many of his followers
(socii) fell along the way because their horses were exhausted.
In 584 Childebert managed to secure sufficient support from his
subjects (or perhaps from trans-Rhenish warriors) to muster an
army large enough to invade Italy and obtain tribute from the
Lombards. Such a lengthy and dangerous venture had not been
undertaken by an Austrasian monarch since Theudebald's Italian
campaign in 554.41
The uprising of the minor populus in Childebert's support as
well as the successful expedition into Italy in 584 indicates that
he did have some following, yet he was generally at the mercy
of the magnates from his accession to the throne until the death
of Chilperic in 584. Guntram, on the other hand, was in a consider-
ably stronger position. The local levy of Bourges remained loyal
to him, as did the standing army of Burgundy. Even without the
capable leadership of Mummolus, this force under Guntram's
personal command was able to defeat Chilperic's troops in the
field.
Chilperic had dominated military affairs in Gaul after the
death of his brother Sigibert, but when he attained the pinnacle
of power and success, the same fate that had befallen his brother
also became his: he was struck down by the hand of an assassin.
This led to struggles that involved his lands and his erstwhile
supporters. It should be pointed out that Chilperic had only been
moderately successful when he relied essentially upon warbands
between 561 and 575, but as he expanded his holdings into more
Romanized and more urbanized areas between 575 and 584
and utilized local levies his cause prospered considerably more.
584-593
When the news of Chilperic's murder spread across the land, the
magnates and monarchs began vying for power and support,
resulting in conflicts that resembled the chaos following the
death of Sigibert a decade earlier. Although a number of Chil-
°- Ibid., VI, 31, 42.
56
CLOVIS'S GRANDSONS
peric's adherents remained loyal to Queen Fredegund, and his
son Chlotar II was even crowned king of Neustria, other sup-
porters of Chilperic declared their allegiance to either Childebert
II or the pretender Gundovald. The major conflict which flared
up after Chilperic's death was not, however, a struggle for per-
sonal supporters, but a struggle for the deceased ruler's cities.
Soon after his death, Childebert's Duke Gararic seized Limoges
and Poitiers, the local levies of Blois and Orleans attacked Cha-
teaudun, the men of Bourges subjected Tours to King Guntram's
rule, and then together the levies of Bourges and Tours took
Poitiers for Guntram. In this struggle for power and this vying for
position, magnates led their armed bands in all kinds of activities:
for example, Avius, who had been raiding and plundering in the
area around Poitiers with his followers, was killed by Childeric
the Saxon who at that time was leading his own armed retainers
(pueri)*2
Childebert and Guntram were not the only ones who hoped to
gain by Chilperic's death, for Dukes Desiderius and Bladast
joined Gundovald's cause. Desiderius, upon learning of his king's
death, led a band of picked men into Toulouse and seized the
treasure that Princess Rigunth had been taking to Spain. He
placed the treasure under guard in Toulouse and sped to Avig-
non where he met Gundovald and Mummolus. In November of
584 Desiderius, Gundovald, Mummolus, and their retainers
quickly moved to Brives-la-Gaillarde near Limoges, where Gun-
dovald was raised to the kingship. Their plans to go north and
occupy Poitiers were thwarted, however, by the levies of Orleans
and Bourges which had retaken the city after the Poitivins re-
turned to Childebert's side following their submission to the men
of Bourges and Tours. Therefore Gundovald went south, receiv-
ing the submission of Angouleme, Perigueux, and Toulouse, and
proceeded to Bordeaux. In January or early February of 585,
*2 For Chilperic's supporters defecting to Childebert see Gregory, Hist.,
VII, 4; for Gundovald's supporters, VII, 9, 10, 27; for Duke Gararic, VII,
13; for the levies of Blois, Orleans, Chateaudun, and Chartres, VII, 2; for
Bourges, VII, 12; for those of Chilperic's followers who remained loyal to
his son, VII, 7; and VII, 3: " . . . unus ex pueris Childerici Avonem hasta
transfixit."
57
when Guntram learned of Gundovald's activities in the south, he

arranged a pact with Childebert and sent those troops previously
ordered to Poitiers to seek out the pretender and capture him.
Guntram's force, under the command of Duke Leudegisel and the
patrician Aegilen moved south to the Dordogne River and waited
there for news of Gundovald.43
Meanwhile, when Gundovald learned of Leudegisel's advance,
he retreated rapidly toward the fortress of Con venae (modern
Saint-Bertrand-de-Comminges). The remains of his treasure were
being transported by camels; because they could not keep up
with his retreating army, many were abandoned, as were horses
exhausted by the rigorous pace. Realizing that a showdown with
Guntram's forces was near at hand, Gundovald sent messengers
to the magnates of Childebert's kingdom whose support he had
been intending to buy with the great treasure he was bringing
with him. His emissaries, however, were captured by Guntram's
counts and were unable to carry out their mission. Early in Feb-
ruary, probably before the twentieth, Gundovald made plans for
the winter, gathering and storing provisions at Convenae, a hill-
top fortress with an internal water supply. He then tricked the
armed inhabitants of the fortress into leaving, relying upon the
strength of his position and the ability of his supporters to hold
out until the magnates of Childebert's kingdom and their fol-
lowers came to his rescue.44
While Gundovald was in retreat, Leudegisel received word
that the pretender had crossed the Garonne and was heading
south, so he sent his cavalry in hot pursuit. Some of the men
were drowned while crossing the Garonne, but those who sur-
vived found the exhausted horses and treasure-laden camels and
thus picked up Gundovald's trail. When Leudegisel learned that
Gundovald had reached Convenae, he ordered his slow-moving
baggage train and infantry to converge on the fortress. He also
commanded his horsemen to ride to Convenae with all possible
43
Gregory, Hist., VII, 9, 10, 24, 26, 27, 28, 31, 33, 34; VII, 24: "Anno
igitur decimo Childeberthi regis rex Gunthramnus, commotis gentibus regni
sui, magnus iuncxit exercitum. Sed pars maior cum Aurilianensibus adque
Biturigis Pectavum petiit."
" Ibid., VII, 30, 34.
58
CLOVIS'S GRANDSONS
speed, presumably to keep Gundovald from provisioning the
fortress for a long siege. Fortunately for Gundovald, Leudegisel's
cavalry stopped to plunder the Church of St. Vincent near Agen
and did not carry out its mission.45
By the end of February or early March, all of Leudegisel's
forces were finally encamped around Convenae with their tents
pitched, their foragers in the field, and access to and from the
fortress blocked. After some two weeks, Leudegisel found it
necessary to prepare new siege engines; battering rams were
mounted on wagons covered with sheds to protect the attackers
from missiles hurled from the walls of the fortress. Leudegisel's
troops also tried to fill up the moat on the eastern side of the
fortress, but their efforts were to no avail. Each attempt to storm
or breach the walls was met successfully by the defenders, who
hurled down loads of stones and cauldrons of burning oil and
pitch from the walls.46
Though Leudegisel's elaborate siege operations were unsuccess-
ful, Convenae nevertheless fell to the attackers. When Mummolus
and Gundovald's other supporters learned that help would not
be forthcoming from the magnates of Childebert's kingdom, they
concluded that their position was hopeless and in return for a
promise of pardon from Leudegisel they pressured Gundovald
into surrendering. The pretender, knowing he had been betrayed,
had no choice but to surrender since the magnates' followers
were the only troops he had. Perhaps if Gundovald had not
tricked the resident fighting men of Convenae into leaving the
fortress, he would have had a force capable of counteracting that
of Leudegisel. At any rate, more was lost than the fortress, for
despite assurances to the contrary, Gundovald, Mummolus, and
45
Ibid., VII, 35.
M
Ibid., VII, 35, 37: "Quintus et decimus in hac obsidione effulserat dies,
et Leudeghiselus novas ad distruendam urbem machinas praeparabat. Plaus-
tra enim cum arietibus, cletellis et axebus tecta, sub quae exercitus pro-
peraret ad distruendos muros. Sed cum adpropinquassent, ita lapidibus
obruebantur, ut omnes adpropinquantes muro conruerint. Cupas cum pice
et adipe accensas super eos proicientes, alias vero lapidibus plenas super
eos deiciebant. Sed cum nox certamina prohiberit, hostes ad castra regressi
sunt."
59
Bishop Sagittarius were killed. Some of the other magnates, how-

ever, managed to escape.47
Unlike Gundovald's army, Guntram's forces, described by
Gregory as a huge army, were called out from throughout his
kingdom and consisted of all those men who owed service. Not
only were the fighting men of the old kingdom of Burgundy,
under their patrician, Aegilen, called out, but so too were the
local levies of many of Guntram's cities, such as Bourges, Poi-
tiers, and Tours. Duke Leudegisel, the overall commander, had a
substantial body of cavalry, presumably strong enough to keep
Gundovald from provisioning Convenae and perhaps even large
enough to engage the pretender's entire force if necessary. If
these mounted troops were not thought capable of accomplish-
ing at least the former, there seems to be no reason for their
being sent in hot pursuit of the enemy. There is no way, how-
ever, to identify the role which these horsemen played within
the organization of Leudegisel's forces. In addition to a cavalry
force of note, Leudegisel's army carried a substantial quantity
of materiel, including tents and probably siege engines. He also
employed "engineers" capable of building and perhaps even
designing siege engines.
The deaths of Chilperic and Gundovald and the pact between
Childebert and Guntram ended the bella civilia for a time. Both
kings now sought to subdue those magnates whose loyalty they
questioned and to extend their efforts in the realm of foreign
affairs. Childebert, for instance, sent armies into Italy on several
occasions. Each effort, however, seems to have fared worse than
the previous one. The only Italian campaign described in con-
siderable detail is that which twenty dukes, serving under
Childebert, led in 590. Part of their force included the levy of
Champagne commanded by Wintrio, the duke of Champagne,
and Audovald, who was also a duke. With six of the other dukes,
Audovald moved on to Milan, but could find neither a Lombard
army to fight nor the Byzantine army with which he was sup-
posed to join. When OIo, another of these dukes, besieged Bel-
"Ibid., VII, 30, 32, 36. For personal armed followings, VI, 20, VII, 38,
IX, 35, and Bachrach, "Charles Martel," pp. 70-72. Gregory, Hist., VII, 39.
60
CLOVIS'S GRANDSONS
linzona, he was killed and many of his men were cut down
while foraging. In northwestern Italy Cedinus, with thirteen
dukes, took at least five Lombard fortresses. In general, however,
Childebert's troops were unable to coax a Lombard army into
meeting them on the field, and the Merovingian forces were un-
able to subjugate any of the larger Lombard cities, such as Pavia.
After ravaging northern Italy for some three months and being
wracked by fever and disease, Childebert's army retreated into
Gaul. Many of the soldiers were so poorly off that they had to
sell their arms and other possessions in order to obtain food.48
In the spring of 585, immediately following the Gundovald
affair, Guntram ordered the fighting forces of his entire kingdom
against the Goths in Septimania. Men from north of the Saone,
west of the Rhone, and north of the Seine, along with men of
the old Burgundian kingdom, advanced on Nimes, plundering
all along the way, even in Guntram's own territory. The local
levies of Bourges, Saintes, Perigueux, and Angouleme, Childe-
bert's levy of Auvergne, and people from other cities in Gun-
tram's kingdom moved on Carcassonne, also plundering as they
traveled. Neither of these forces was able to take those Gothic
cities which closed their gates. No effort was made to mount a
siege; only the surrounding areas and the cities whose citizens
were foolish enough to open then: gates were plundered. By the
end of July it became clear that this army was not going to sub-
jugate Septimania; not only were Guntram's troops finding it
difficult to forage, but the hostile populace was continually
harassing them. By the middle of August when the campaign
was over, many of the participants — Gregory says 5000 — had
died of hunger or had been killed by the people in the areas
they traversed.49
The failure of Guntram's forces in Septimania seems to have
encouraged Reccared, the Visigoth king, to attack the Merovingi-
an frontier. He plundered the area around Toulouse and raided
up to the Rhone; he captured the garrisons of the fortress at
48
Ibid., VI, 42; VIII, 18, and Paul, Hist., Ill, 29. Also Gregory, Hist., X,
3: 49"Audovaldus vero dux cum Vinthrione, commoto Campaniae populo. . . ."
Gregory, Hist., VIII, 30.
61
Cabaret and Beaucaire, plundered within the walls, and then left
them deserted. After accomplishing this, he shut himself up
within the fortress of Nimes. Learning of Reccared's raid, Gun-
tram appointed Leudegisel as patrician in place of Aegilan. Leu-
degisel then proceeded with troops to regarrison the fortresses
devastated by Reccared, while Duke Nicetius of Auvergne led a
force to patrol the frontier.50
It is worth noting that even before the Burgundian kingdom
had fallen under Merovingian rule, garrisoned fortresses played
an important role in the defense of the area. Using erstwhile Ro-
man fortifications and garrisons often descended from laeti and
milites, the Burgundian monarchs and their Frankish successors
provided for the protection of the frontiers, the imprisonment of
enemy captives, and perhaps the bases for standing military
forces. The fortresses attacked by Reccared and regarrisoned by
Leudegisel were part of Guntram's frontier defenses, which
probably included the fortified cities of Orange and Lodeve be-
fore the latter fell into Gothic hands. The men called to regar-
rison the forts at Cabaret and Beaucaire served under the patri-
cian and may perhaps have been detached from the Burgundian
standing army. The men whom Duke Nicetius led to patrol the
border of Childebert's kingdom were not of the levy of Au-
vergne, but custodes of some kind and perhaps a permanent
military body.51
In 587 Duke Desiderius, who had become a supporter of King
Guntram after Gundovald's death, and Count Austrovald of Tou-
louse, who served as his second in command, led the local levy
of Toulouse against Carcassonne. The Gothic garrison at Carcas-
sonne learned of the impending attack, however, and went out
to meet Desiderius's force, but when the battle began the Goths
turned and fled. Desiderius attacked their rear guard and pur-
50
Ibid., VIII, 30; IX, 7: "Unum etiam castro Ugernum nomen cum rebus
atque habitatoribus desolantes, nullo resistenti, regressi sunt." This last at-
tack was in 587, indicating that Beaucaire was regarrisoned by that time.
For the Visigothic side of this campaign see Thompson, The Goths in Spain,
pp. 92-94.
51
V. Epiphanii, 171; Procopius, H.W., V, xii, 29; xiii, 3; Gregory, Hist.,
IV, 42. Also Longnon, Geographic, pp. 441, 613, and Thompson, The Goths
in Spain, pp. 92-94.
62
CLOVIS'S GRANDSONS
sued it hotly, but when he reached the walls of the city only a
few of his men were still with him; the others had been unable
to keep up because their horses were exhausted. Seeing Desideri-
us's precarious position, the men of the city rushed out and
killed the duke and most of those with him.52
Austrovald, who replaced Desiderius as duke in the area of
Toulouse, later subjected Carcassonne to Guntram's control, but
Guntram's desire to gain control of the Gothic-held cities in Sep-
timania did not end with this victory. In 589, four years after his
first futile attempt to subjugate Septimania, he again sent an
army into the area, this time made up of the local levies of Peri-
gueux, Bordeaux, Agen, and Toulouse. Duke Boso commanded
the force, with a certain Antestius as second in command. Clau-
dius, the Gothic leader in Septimania, prepared an ambush for
Boso's levies. He sent a small force against the invaders' camp
and kept the remainder of his men hidden some distance away.
The small attacking force was easily repelled and as it retreated,
Boso's men pursued it. Soon they found themselves between the
two Gothic forces and the ensuing attack almost wiped them out.
Boso's camp was taken and only those who were able to reach
their mounts escaped; those less fortunate were either killed or
captured.53
Guntram's efforts to gain control of Septimania were curtailed
by these failures and also by his need to defend other parts of his
kingdom against the Bretons who had been raiding in the west.
In response to these raids, Guntram sent a force under Dukes
Beppolen and Ebrachar. On this campaign Beppolen was to fall
victim to Ebrachar's plotting and to Queen Fredegund's animosity.
The queen ordered her Saxons from Bayeux to cut their hair in
the Breton style and to join them as allies. While Beppolen was
fighting against both the Bretons and the Saxons, Ebrachar re-
treated with the larger part of the force, leaving his colleague at
the mercy of the enemy who murdered him and his personal
followers in the swamps. Learning that Beppolen was dead, Eb-
53
Gregory, Hist., VIII, 45.
Îbid,, IX, 31; John Biclar, Chron., 589; Fred., IV, 10. Bachrach,
"Feigned Retreat," p. 266.
63
rachar returned and made peace with the Breton leader. But to
return to their own territory Ebrachar's forces had to cross the
Vilaine River. The mounted troops were able to swim their
horses across the river, but the less important fighting men (in-
feriores and pauperes), who were on foot, were not as success-
ful. Some tried to swim across on the backs of horses (cabcdU)
from the baggage train, but most of the untrained mounts and
their equally unskilled riders were drowned. Those who were
not drowned but remained on the west bank of the river were
either killed or captured by the Bretons who thus broke the
peace.54
Aside from these major military campaigns during the last
decade of Guntram's life, there were many relatively minor en-
counters involving smaller groups. Guntram himself, according
to Gregory of Tours, never went anywhere, even to church,
without a large following of armed men. He also sent men from
the court to carry out paramilitary missions, one of which was
the killing of a certain Boantus by a detachment of pueri be-
cause he had been unfaithful to Guntram. Two of the more im-
portant men who served the king in this capacity were Antestius
and Claudius; neither seems to have had an official title, but
both led missions of a military nature from the court. As has al-
ready been mentioned, Antestius acted as second in command
to Duke Boso in the campaign of 589; he also served Guntram
by punishing troublemakers of Angers and Nantes. Claudius,
too, took part in such activities; his service cost him his life. He
was sent with his own pueri to capture Eberulf, one of Chil-
peric's high officials accused of his murder. Claudius and his
pueri fought it out with Eberulf and his followers, resulting in the
death of both leaders and some of the pueri.55
Many of these paramilitary activities were directed against
Childebert, who not only had to deal with such minor harass-
ments but was faced with a more serious conspiracy. For in the
autumn of 587 three of his magnates — Berthefred and the Dukes
54
Gregory, Hist., X, 9.
66
Ibid., VII, 8, 18; IX, 3, for Guntram's personal guards; VIII, 11, for
Boantus; VIII, 43; IX, 31, for Antestius; and VII, 29, for Claudius.
64
CLOVIS'S GRANDSONS
Rauching and Ursio — plotted to kill the king and then rule the
kingdom in the names of his sons. Childebert learned of the
plot, however, and invited Rauching to the court where the
king had the duke murdered. At the same time, Childebert sent
some of his pueri from the court to seize Rauching's extensive
estates. When the pueri who had accompanied Rauching to the
court learned of their lord's death, one of them rode off with
word of the disaster to his widow who received the news as she
was parading through the streets of Soissons with a following of
her own armed guards.56 While Rauching went to see Childe-
bert, Ursio and Berthefred prepared their forces. When they
learned of Rauching's death, they took refuge with their follow-
ers and families at a Roman fortification which had been con-
verted into a church. In December of 587 Childebert sent a
force against them. Godegisel, Childebert's commander on this
occasion, laid siege to the fortified place and then tried to burn
out the conspirators. Ursio sallied forth with his followers and
cut down many of the besiegers before he himself was killed.
Berthefred escaped on horseback with some of his armed re-
tainers, but they were eventually caught and slain.57
MILITARY COMPLEXITY: 561-593

The period during which one or more of Clevis's grandsons ruled
in Gaul is the most thoroughly documented era in Merovingian
history. Because of the comparative abundance of evidence, the
heterogeneous nature of the Merovingian military can be demon-
strated in considerable detail. One of the most important innova-
tions by the Merovingian monarchs, which contributed to the
complexity and diversity of their military organization, was the
development of local levies in many of the cities in Gaul.
These local levies, either acting singly or in groups, played a
significant role in military affairs. The value of these levies to the
king depended in part on his ability to hold the loyalty of the im-
portant men who commanded them, as is illustrated by Sigibert
when he had to send loyal commanders to get the levies of Cha-
86
w
Ibid., IX, 9.
Ibid., IX, 12.
65
MEROVINGIAN MILITABY ORGANIZATION
teaudun and Tours to support him in 575. His son Childebert
found it difficult during the first decade of his reign to keep the
loyalty of his magnates and was therefore unable to use the
local levies of his cities effectively. Similarly, King Guntram had
trouble with his magnates during the last years of his conflict
with Chilperic and consequently had difficulty in mobilizing the
support of his own local levies. Conversely, Chilperic utilized
the local levies of his cities effectively in his expansion at the
height of his power because he had the support of his com-
manders.
It was only in cities in Neustria and Aquitaine that the locally
based levies appeared during the reigns of Clovis's grandsons.
Gregory notes their operation, between 568 and 591, at Agen,
Angers, Angouleme, Auvergne, Bayeux, Blois, Bordeaux, Bourges,
Chartres, Chiteaudun, Le Mans, Nantes, Orleans, Perigueux,
Poitiers, Saintes, Toulouse, Tours, and Velay.58 In Champagne
a levy seems to have been organized on a provincial basis, rather
^ Ibid., IV, 30: "Sigyberthus vero rex Arelatinsim urbem capere cupiens,
Arvernus commovere praecipit. Erat enim tune Firminus comes urbis illius,
qui euro ipsis in capito abiit." IV, 50: "Dolorem enim ingerit animo ista
civilia bella referre. . . . Sigyberthus . . . mittens nuntius Dunensibus et
Toronicis, ut contra Theodoberthum ire deberent." V, 1: "Tune Roccolenus
cum Cinomannicis Toronus venit et praedas egit et multa scelera fecit. . . ."
V, 26: "Dehinc Toronici, Pictavi, Baiocassini, Caenomannici et Andeacavi
. . . ex iussu Chilperici regis abierunt et contra Varocum . . . ad Vicinonam
fluviam resedent." VI, 12: "Berulfus vero dux, cum Bitorigus rmisitare, quod
Toronicum terminum ingrederentur, audisset, exercitum commovet et se in
ipsos fines statuit." VI, 26: "At ille, adsumptos secum Arvernis atque Villa-
vis, Avennione abiit." VI, 31: "Berulfus vero dux cum Toronicis, Pectavis,
Andecavisque atque Namneticis ad terminum Bitoricum veriit." VII, 2:
"Aurilianensis cum Blesensibus iuncti super Dunenses inruunt eosque in-
opmantes prosterunt. . . . Quibus discedentibus, coniuncti Dunenses cum
reliquis Carnotenis. . . ." VII, 13: ". . . exercitum contra Pectavos commovit,
ut scilicet ab una parte Toronici, ab alia Biturigi commoti cuncta vastarent."
VII, 21: "Tune data occasione, ut custodiretur, Aurilianensis adque Ble-
sensis vicessim ad has excubias veniebant, impletisque quindecim diebus,
cum praeda multa revertebantur, adducentis iumenta, pecora, vel quodcum-
que derepere potuissent." VII, 24: "Sed pars maior cum Aurilianensibus
adque Biturigis Pectavum petii." VIII, 30: "Similiter et Byturigi, Sanctonici
cum Petrocoricis, Ecolesenensibus vel reliquarum urbium populum. . . .
usque Carcasonam urbem devicti, similia mala gesserunt. . . . Tune et
Nicetius dux cum Arvernis in haec expeditione eommotus, cum reliquis
urbis adsedit." IX, 31: ". . . ipse cum Sanctonicis, Petrocoricis Burdegalen-
sibusque, Agennensibus etiam ac Tolosanis illuc direxit."
66
CLOVIS'S GRANDSONS
than on the more limited scope of the other local levies.59 In
neither Austrasia nor the lands which before 534 formed the
Burgundian kingdom do locally organized levies seem to have
been developed during this period. Gregory makes no mention
of forces at Cologne, Zulpich, Metz, Mainz, Verdun, Strasbourg,
Worms, Spire, Toul, or Maastricht in a manner which might be
interpreted as a force organized on a local basis. Neither does
Gregory mention levies at Lyons, Nevers, Autun, Chalons,
Geneva, Grenoble, or Valence. It is true that Gregory lived at
Tours and perhaps showed a greater interest in the western and
more Roman parts of Gaul than in Austrasia and Burgundy. Yet
Lyons and Vienne are closer to Tours than either Toulouse or
Bordeaux, both of which Gregory indicates as having a local
levy; and it would be difficult to argue that the latter were more
Romanized than the former. In addition, it must be noted that
among the dozens of military encounters which Gregory de-
scribes many involved forces from Austrasia and the erstwhile
Burgundian kingdom.
The levy of a city was usually commanded by the count of the
tivitas, although in several instances a group of cities, with their
counts, was brought together under the overall administrative
and military control of a duke. For example, Duke Ennodius
was responsible for Tours, Poitiers, Aire, and Lescar; Duke Berulf
was his predecessor in Tours and Poitiers; Nicetius was the duke
of Clermont, Rodez, and Uzes; Duke Beppolen administered the
cities of western Gaul belonging to Chlotar II, but controlled
by King Guntram. To these commands might be added the re-
gional levy of Champagne which was led by the duke. These
dukes who served as the administrative and military heads of
several cities should be differentiated from those dukes who
held ad hoc command of one or more local levies for a single
campaign. For example, Duke Guntram Boso took command of
the levies of Auvergne and Velay in 582 in an attempt to rout
Mummolus who was ensconced at Avignon, and Duke Boso (not
69
Ibid., V, 3: ". . . collecti aliqui de Campania. . . ." X, 3: "Audovaldus
vero dux cum Vinthrione commoto Campaniae populo. . . ." X, 27: "Com-
motns autem pro hac causa Campanensis populus. . . ."
67
to be confused with the former) commanded the levies of P&i-
gueux, Bordeaux, Agen, and Toulouse in 589 for an invasion of
Septimania.60
During the latter part of the sixth century, these locally or-
ganized levies, as well as the provincial levy of Champagne,
were used in both offensive and defensive actions. As defensive
measures they protected their area from invasion and performed
police duties of a substantial nature. Offensively, levies went
against the Goths in Septimania, the pretender Gundovald, and
the Bretons. None of the city levies were used outside of Gaul,
however; they did not go south of the Alps, south of the Pyr-
enees, or east of the Rhine. In fact, ventures of more than 300
miles from their home base seem to have been beyond their com-
petence, although they did remain away for more than three
months at a time during all seasons.81
The personnel composition of these levies is difficult to ascer-
tain. Neither all men, nor even all freemen, were required to
serve. For example, on occasion a son might ask and receive per-
mission to serve in place of his father, who would owe service,
whereas the son, who was in some cases more fit to serve, did
not.62 The selective nature of the local levy is further emphasized
by an episode which took place in Guntram's army when it went
against the pretender Gundovald during the autumn of 584. In
addition to the levy of Tours, a large number of other inhabi-
tants of the city (whom Gregory clearly distinguishes from the
levy) followed along in the hope of obtaining booty.63
"'Ibid., VI, 18, 21, 26, 31, 42; IX, 31.
81
The levy of Poitiers which traveled south to fight at Convenae in 585
seems to have gone further from its home base than any of the other local
levies. The provincial levy of Champagne went beyond the borders of Gaul
and fought in Italy (Gregory, Hist., X, 3).
88
Form. Andec., 37 (MGH Form. Merov. et Karol.): "Dum in omnibus
et per omnia et super totum nobis fidiliter servire videras, multa penurias et
iniurias per deversa loca pro nostra necessitate successisti, et in utilitate
domnorum partibus Brittanici seu Wasconici austiliter ordine ad specie mea
fuisti, proinde convenit nobis, ut aliquid de facultatis nostra te emeliorare
deberent; quod ita et fecerunt." See the discussion of this by Boutaric,
Institutions militaires de la France, p. 59, and N. D. Fustel de Coulanges,
Histoire des institutions politiques de I'ancienne France (Paris, 1888), IV,
293, n. 2.
83
Gregory, Hist., VII, 28: "Post haec [autem] exercirus ab urbe Pectava
68
CLOVIS'S GRANDSONS
Any suggestion that service in the levies was universal is cast
in doubt by the eloquent silence of the Prankish law codes on
this point. These codes were drawn up and revised throughout
the sixth century and do not mention an obligation for all to
serve. At one time in the sixth century, however, both poor men
and economic dependents seem to have been included among
those who owed service in the local levies. But by the last
quarter of that century it was customary, at least in northwestern
Gaul, for the poor to be exempt from such service. Some eco-
nomic dependents (particularly those of the Church of St. Mar-
tin) were exempt as well.64 These exemptions were granted not
out of benevolence, however, but rather because of the economic
repercussions that would have resulted from the use of such
people in the local levies. For economic dependents, burdened
with labor services which they owed to their lords as well as
with the upkeep of their own holdings, could hardly perform in
even the most limited offensive campaigns (a week or two)
without causing economic dislocations more damaging than their
absence from military campaigning. Fields left unplanted or
crops left to rot unharvested could not possibly have been per-
mitted as an institutionalized consequence of the mode of mili-
tary organization which flourished in Neustria and Aquitaine for
a generation and more. In short, the demands of the division of
labor caused certain limitations in the recruitment of men for the
local levies. Whereas the requirements of economic dependence
limited mobility and thus service, poverty had similar effects.
The men who served in these local levies seem to have had to
rely upon their own resources to provide the food and materiel
necessary for their service. The inability of the poor to provide
their own weapons and sustenance for several months of cam-
paigning may well have been the reason for their eventual ex-
clusion from the local levies.
remotus inantea post Gundovaldum profiscitur. Secutique sunt eum de Toro-
nici multi lucri causa; sed Pectavis super se inruentibus, nonnulli interempti,
plurimi vero spoliati redierunt. Hi autem qui de his ad exercitu prius
iuncxerant pariter abierunt."
64
Ibid., V, 26: "Post haec Chilpericus rex de pauperibus et iunioribus
eclesiae vel basilicae bannos iussit exigi, pro eo quod in exercitu non
69
In negative terms, it has been ascertained that the personnel of

the local levy was not the same as all the male or the free male
inhabitants of the civitas, and that both the poor and economic
dependents were exempted. In positive terms, it seems reasonable
to assume that the lowest possible economic status from which
local levies might be recruited regularly were owners or holders
of land with a sufficient number of economic dependents or
slaves to perform the necessary agrarian services so that pro-
duction would not suffer. This conclusion is supported by the
conflict between die demands of the agricultural cycle and the
logistical demands on an individual in a local levy. If the rank
and file of the local levy were small landholders with a few
slaves or economic dependents, it was also true that important
personages (magni viri) served in the local levy, as can be demon-
strated by the one from Auvergne.65
The Arvernian levy led by Firminus in 568 seems to have con-
sisted, at least in part, of horsemen wearing helmets and carry-
ing shields. Two difficulties arise, however, in drawing this
conclusion from Gregory's account. Gregory uses a quotation
from Virgil — "scuta virorum galeasque et fortia corpora volvit"
— to indicate that they used helmet and shield. By using a quo-
tation from Virgil, Gregory's remarks about the helmets and
shields are less reliable than they would have been if he had
used his own words. An even greater problem exists when at-
tempting to ascertain if the levy of the Auvergne was mounted
and had helmets and shields, because not only the levy but
Audovar's troops were involved in this encounter.
Gregory's remarks concerning horses and armor therefore may
refer to Audovar's group which was not even a local levy let
ambulassent. Non enim erat consuetudo, ut hi ullam exsolverent publican!
functionem." VII, 42: "Post haec edictum a itidicibus datum est, ut qui in
hac expeditione tardi fuerant damnarentur. Biturigum quoque comes misit
pueros suos, ut in domo beati Martini, quae in hoc termino sita est, huius-
modi homines spoliare deberent. Sed agens domus illius resistere fortiter
coepit, dicens: 'Sancti Martini homines hii sunt. Nihil eis quicquam inferatis
iniuriae, quia non habuerunt consuetudinem in talibus causis abire.'"
85
Ibid,, IV, 30, quoted in n. 58. The origins of this levy may go back
to the Visigothic period when the men of the Auvergne, including some of
the more important people, served at VouilM {Gregory, Hist,, II, 37). R.
Latouche, The Birth of the Western Economy (London, 1961), pp. 59-116.
70
CLOVIS'S GRANDSONS
alone a part of the levy of Auvergne. There are other instances

of levies serving in armies with strong mounted contingents: the
levies of Bourges and Poitiers served in Leudegisel's army which
had a substantial force of horsemen; the force led by Desiderius
against Carcassonne in 587 included the levy of Toulouse and
had enough horsemen to rout the rear guard of the Gothic gar-
rison; and the force commanded by Duke Boso in 589 when he
attacked Septimania included the levies of Perigueux, Agen, Bor-
deaux, and Toulouse and had a noteworthy number of horsemen.66
The ambiguity of the evidence is such that it is not possible to
conclude positively that substantial elements in these local levies
served on horseback and had armor. Nevertheless, the frequent
mention of mounted forces in the same context with levies does
make such a conclusion reasonable.
At times the Merovingian kings supplemented their forces with
what might be called a general levy, including apparently just
about anyone who could hurl a stone or swing a club. Among
the terms used by Gregory to describe members of such levies
are inferiores and pauperes. The military value of these elements
seems to have been negligible and they do not seem to have seen
very much action. The term minor populus, on the other hand,
appears to refer to a stratum of society above that of the inferi-
ores and pauperes. The minor populus was inferior to the mag-
nates, and it may have formed the rank and file of the city
levies.67
The general levies, though infrequently used, and the local
levies can both be best described as a militia of sorts, a part-time
army. In contrast, there were standing forces such as the patri-
cian's army in Burgundy, the laeti and milites who served in gar-
risons, the erstwhile Roman allies like the Saxons who lived in
colonies, and the antrustiones who served in part in the centenae.
The custodes frequently mentioned by Gregory may have been
86
Gregory, Hist., IV, 38, for the Auvergne; VII, 35, for Leudegisel; VIII,
45, for Desiderius; and IX, 31, for Boso.
87
Ibid., VII, 24: ". . . rex Gunthramnus, commotis gentibus regni sui.
. . ." VIII, 30: ". . . commoto omni exercitu regni sui, illuc dirigit. . . ."
X, 9: " . . . inferiores et pauperes, qui cum his erant, simul transire non
potuerunt." On the minor populus see VI, 31.
71
organized in special groups of some kind or they may have been
drawn from one of the other standing forces. The contexts in
which they are mentioned make it clear that they are not local
levies. Most of these standing forces seem to be overwhelmingly
imperial in origin — the laeti were Roman military colonists, the
milites were descendants of Roman soldiers, the centena was
commanded by a centenarius who held a Roman title. The cen-
tena was placed on lands of the royal fisc which was often the
former imperial fisc. The antrustiones who served on the cen-
tenae were drawn not only from among the Franks, but from
among the Romans and laeti as well. Those who served in gar-
risons performed their duties in fortified places built by the
empire.
Along with these local and general levies, the Merovingian
monarch had at his disposal a personal armed following which
served him at the court. These fighting men were employed to
quell minor uprisings, deal with unfaithful magnates, and of
course to protect the king. Most often the rank and file of these
groups are called pueri, which seems to indicate low estate. Pueri,
nevertheless, were members of the royal trust and served as
antrustiones.
As has already been mentioned in chapters I and II, the per-
sonal armed followings of the magnates also played an important
role in the complex Merovingian military. The shifting loyalty of
these powerful men could result in critical situations such as the
Gundovald affair, Rauching's conspiracy, and Merovech's ven-
ture. In recognition of the problem caused by the shifting loyalty
of these magnates, whom Gregory sometimes calls leudes, Gun-
tram and Childebert II agreed not to "invite" each other's leudes
to join them and not to accept those who volunteered to switch
sides.68 The example of Godinus, whose defection from Sigibert
to Chilperic in exchange for lands from the Neustrian fisc caused
a war between the two monarchs, is worth recalling in this
context.
68
Ibid., IX, 20, for the Pact of Andelot: "Similiter convenit, ut nullus
alterius leudis nee sollicitet nee venientes expipiat."
72
CLOVIS'S GRANDSONS
On the whole, the Merovingian military in the era of Clevis's
grandsons is a confusing mosaic of heterogeneous elements. The
legacy of imperial institutions — milites, laeti, allies, and gar-
risoned Roman fortifications — coexists with pagan hordes from
beyond the Rhine, centenae of antrustiones recruited among Ro-
mans and barbarians, local levies from the cities of the Roman-
ized parts of Gaul, the personal followings of the magnates, and
the pueri regis. When a monarch was able and fortunate, these
various groups tended to serve him with a significant measure
of loyalty, but during minorities, regencies, and interregna, or
other times of great stress and upheaval, many of these groups
served nonroyal interests. The monarchs' struggle to maintain
royal power cannot be separated from their efforts to control
these many elements of the Merovingian military. Such efforts
were constant and costly and the loyalty of the magnates, which
was perhaps most important, was often for sale to the highest
bidder.
73
CHAPTER IV
The Last of the

Ruling Merovingians:
593-638
T HE PERIOD of the last ruling Merovingians witnessed internal
fighting similar to that which had characterized the reigns
of Clevis's grandsons. Soon after the death of King Guntram in
593, some of the magnates of Childebert II's kingdom attempted
to despoil Chlotar II's holdings since the young king no longer
had the strong hand of Guntram to protect him and his lands.
The first phase of this effort — undertaken by Duke Wintrio of
Champagne and magnates (superiores) from Austrasia and Bur-
gundy — was directed against Soissons. The force was neither a
general levy nor the territorial levy of Champagne.1
When Queen Fredegund learned of Wintrio's maneuvers, she
called together a number of dukes (duces) to decide what should
be done to prevent Wintrio's men from seizing part of Chlotar's
territory. (The word dux, as used here by the author of the Liber
Historiae Francorum, seems to mean, in a nontechnical sense,
simply the leader of a band of men.) The force which went to
deal with Wintrio was led by Fredegund and Landri, mayor of
the palace of Neustria, and was neither a general levy nor a local
one as can be discerned from the terms used to describe it.2
1
For the terms used to designate the levy of Champagne see ch. Ill, n.
59, above. L.H.F., ch. 36.
2
Longnon, Geographie, pi. X, for the cities held by Chlotar, and ch. Ill,
n. 58, for the local levies in Gaul. See also Eugen Ewig, "Die frankischen
Teilreiche im 7 Jahrhundert (613-714)," Trierer Zeitschrift, XXII (1953),
86-105. See also ch. Ill, n. 60, above.
74
THE LAST OF THE RULING MEROVINGIANS
As a general stratagem for the battle, the queen suggested a

surprise attack to compensate for her forces' weaknesses and to
lessen the chances of an open battle, which would have placed
her men at a severe disadvantage. In accordance with this plan,
the Neustrian horsemen slowly advanced toward Wintrio's camp
in the predawn behind a line of warriors carrying tree branches
to camouflage their movements. The author of the Liber, who
has described these events in considerable detail, contends that
one of Wintrio's sentries heard the tinkle of a horse's bell and
upon seeing the camouflaged horsemen, thought a forest had
appeared where none had been before. When he mentioned the
incident to his fellow sentry, he was told the bells belonged to
their own mounts pastured near the camp and, as for his remark
about the forest, he must be drunk his colleague concluded. In
any event, a warning was not given; at dawn the Neustrians
charged into Wintrio's camp killing a large number of men and
scattering the rest. The Neustrians then took Rheims and ravaged
Champagne before returning to Soissons with their booty.3
Although Childebert's prestige may have been diminished by
the failure of his supporters, he was able to crush a Thuringian
revolt in 595, suggesting that his army had recovered from the
disaster two years earlier. He died soon after, however, and his
sons Theuderic and Theudebert inherited his kingdom, receiving
Burgundy and Austrasia, respectively. Soon after Childebert's
death, Chlotar took over Paris and several other cities belonging
to his cousins. In retaliation Theudebert and Theuderic led a
force against the Neustrian kingdom, but Chlotar's followers
were victorious once again, this time at Laffaux near Soissons.
Little is known about this battle except that the Neustrians won
with a charge.4
3
L.H.F., ch. 36: "Burgundiones et Austrasii superiores Franci simul com-
moti grande exercitu, valde Campanias digress!, paygo Suessionico cum
Gundoaldo et Wintrione patriciis vastantes ingrediuntur. Haec audiens
Fredegundis, cum Landerico et reliquos Francorum duces hostem congre-
gat. Brinnacum villa veniens, multa dona et munera Francis ditavit, eosque
ad pugnandum contra inimicos eorum coortans." See Fred., IV, 14. Fredegar
and the Liber use the term patrician to describe both Wintrio and Gun-
doald. Gregory called the leader in Champagne a duke.
* L.H.F., ch. 37, confuses two campaigns, one in the year 596 or 597
75
However, Theuderic and Theudebert had extensive military
resources to draw on, and when fully exploited, these would
place the less fortunate Chlotar at the brothers' mercy. Not only
were the combined kingdoms of Theudebert and Theuderic
more than three times larger than Chlotar's holdings, but the lat-
ter had only one local levy at his command — that of Bayeux.
Theuderic could call on the levies of Toulouse, Agen, Nantes,
Angers, Saintes, Angouleme, Perigueux, Orleans, Blois, Chartres,
and Le Mans, as well as the standing army of Burgundy. Theude-
bert controlled the territorial levy of Champagne and the local
levies of the Auvergne, Poitiers, Tours, Vellay, Bordeaux, and
Chateaudun; he also had access to the general levy of Austrasia,
pagan Germans from beyond the Rhine, and Alamans from the
Transjura area.5
Despite this significant military advantage, it took the brothers
more than four years to mobilize their superior forces and defeat
Chlotar. In 600 Theuderic and Theudebert invaded Chlotar's
kingdom and massacred his men on the banks of the Orvanne
near Dormelles. They followed up this victory by attacking the
towns along the Seine which Chlotar previously had conquered.
The brothers' army cut gaping holes in the walls of the towns,
carried off a large number of the populace, and devastated the
area. Chlotar was so decisively defeated that his sole option was
to surrender most of his territory; he was left with only twelve
pagi between the Seine, the Oise, and the sea.6
Theudebert and Theuderic, two years after their triumph over
Chlotar, conducted a successful invasion of Gascony, concerning
which there is no information.7 In the winter of 605 Theuderic
fought against a force under the nominal leadership of Merovech,
Chlotar's son; the real commander, however, was Landri. This
campaign began in the autumn of 604 when Bertoald, mayor of
and the other in 600. Fred., IV, 17, treats the former. For Childebert see
Fred., IV, 15, 16.
5
Longnon, Geographic, pi. X, and the last section of ch. Ill above.
'Fred., IV, 20, and Longnon, Geographic, pi. X. L.H.F., ch. 37, indi-
cates that Theuderic leading a "hoste maximo ex Burgundia" defeated
Chlotar's "commoto Francorum exercitu."
7
Fred., IV, 21.
76
the palace for Burgundy, set out with a following of 300 men to
inspect the royal domains along the Seine (from Paris to the
Channel) which Theuderic had recently taken from Chlotar. At
Arele, on the Seine not far from the coast, Bertoald stopped to do
some hunting. When Chlotar learned where Bertoald was, he
sent Merovech and Landri against him. Bertoald fled south to
Orleans, and the Neustrians laid siege to the city. It was then
almost mid-November and the siege was soon raised. Theuderic
learned that Chlotar's forces were in the field and moved against
them toward the end of December. Bertoald and his followers,
in the van of Theuderic's army, were killed by Landri's men
when they advanced too far beyond the main body of troops.
Theuderic, however, was victorious; his troops captured Mero-
vech, but Landri managed to escape. Although Theudebert and
Chlotar confronted each other at Compiegne, peace was nego-
tiated and both armies returned home without fighting.8
In the next year Theuderic decided to go to war against his
brother Theudebert. The important men (leudes) of Theuderic's
kingdom were not anxious to fight, however, and while some of
them kept Theuderic occupied in conversation, a number of war-
riors murdered Protadius, the mayor of the palace, who had en-
couraged the war. As a result of this setback, Theuderic con-
sented to make peace with his brother. This incident indicates
once again the great power the magnates exercised in military
activities.9
In 609 Theuderic, at the urging of his mother, Brunhild, or-
dered the Irish missionary Columbanus to leave his kingdom.
Columbanus was driven out of his monastery at Luxeuil by
8
Ibid., IV, 25; Longnon, Geographic, pi. X; and F. Lot, L'Impdt fancier
et la capitation personelle (Paris, 1928), p. 102.
9
Fred., IV, 27: "Quod cum loco nomen Caratiaco Teudericus cum exercito
castra metasset ortabatur a leudibus suis ut cum Theudeberto pacem iniret.
Protadius singulos ortabatur ut prilium committeretur. . . . Protadius in
tenturio Teuderici regis cum Petro archyatro tabulam ludens sedebat. Cum
eum undique iam exercitus circumdasset et Teudericum leudis suae tenebat
ne illuc adgrederit. . . . Inruentes super eum, tenturium regis gladio undique
incidentis, Protadium interficiunt. Teudericus confusus et coactus cum
fratri Theudeberto pacem inivit et inlesus uterque exercitus revertit ad pro-
priis sedibus."
77
Baudulf, one of Theuderic's magnates, and was taken to Besan-

con. While in exile there, he helped a group of captives escape
from Theuderic's prison. The tribunus who commanded a garri-
son of milites in the city tried to recapture the fugitives but was
unsuccessful. Soon after, Columbanus himself escaped from Be-
san?on and returned to his monastery. Theuderic, upon learning
of these events, sent a troop of soldiers to the monastery to cap-
ture Columbanus, but they failed. Theuderic therefore ordered
a band of men from the court under the leadership of Count
Berthar and the magnate Baudulf to take Columbanus prisoner.
When the missionary refused to come out from the monastery,
Berthar and Baudulf departed; they left behind a group of men
who dragged Columbanus from his sanctuary, and then a cer-
tain Ragamund in command of a troop of custodes escorted the
monk from Theuderic's kingdom.10
The journey into exile undertaken by Ragamund and Colum-
banus provides some insights into communications and paramili-
tary transport during this period. Since Ragamund intended to
travel from Luxeuil north and west to the mouth of the Seine,
where Columbanus would sail to Britain, the party went from the
monastery to Besancon and then north to Autun, stopping at the
fortress of Cavalo. They continued north from Cavalo along the
Roman road to the village of Chora, which had once served as
the base of a Sarmatian military colony. (The toponymy indi-
cates that the area continued to be influenced by these laeti
throughout the Middle Ages.) Near Chora, a functionary of
Theuderic's horse-breeding estates attacked Columbanus and
tried unsuccessfully to kill him. Despite, this and several other
incidents, the group arrived at Auxerre where Ragamund seems
to have reconsidered his plans; rather than continue north to
Paris and the coast, he decided that the party should backtrack
to Nevers where he could obtain boats to transport them to
Nantes. These river craft (scarfa) carried not only the custodes,
oarsmen, and prisoners, but tents and sufficient provisions for
the several days' trip from Nevers north to Orleans.11
10
V. Columbani, I, 19, 20. Fred., IV, 36, provides an abridged version
of the story.
* V. Columbani, I, 20, 21, 22.
78
Upon reaching Nantes, Ragamund handed his prisoner over

to Count Theudoald who was to arrange for Columbanus's de-
portation. The latter again escaped, however, and fled to the
court of Chlotar II where he was given an armed escort for a
journey to Italy. Going from Paris to Meaux and then to Pussy
on the Marne, Columbanus left Chlotar's lands and entered the
kingdom of Theudebert, who provided the missionary with
guards (comites e regio latere and custodes) and manned boats
for his travels on the Rhine.12
This account of the Columbanus affair provides additional
evidence on the diverse nature of the Merovingian military. Like
Theuderic I and Chilperic, Theuderic II utilized milites serving
under tribuni. The direct relation of these milites with the Ro-
man past is indicated in the sources, and therefore the inclination
to assume that both Gregory and Jonas are employing obsolete
imperial terms to describe a contemporary institution should be
avoided. These authors use the word milites very sparingly and
only when describing events about which they have reliable in-
formation. Gregory was at Tours when the milites there were
mistreated by Count Leudast; Gregory's uncle was involved in
the affair at Langres; and Jonas was a companion of Colum-
banus. Procopius, writing in the mid-sixth century, notes that
even in his day the descendants of Roman soldiers once stationed
in Gaul continued to perform the duties of their fathers.13 The
Merovingian monarchs regarded these descendants of imperial
milites and imperial laeti within the framework of the personality
of the law as quasi-national groups analogous in status to the
Romani. The milites and the laeti were treated in the Salian law
in this same manner as were their wives (milituniae and laetae)
and daughters of marriageable age (puellae laetae and puellae
milituniae}^
12
Ibid., I, 23, 24, 25, 26, 27.
13
Gregory, V.P., VII, 4; Gregory, Hist., V, 48; and Procopius, H.W.,
V, xii, 16ff.
11
Cap. Lex Sal. add., C, 4: "De puellas militu[n]ias vel litas haec lex
medietate servetur." CIV, 9: "Aut Romanum ingenuum vel tributarium aut
militem, solidos culpabilis iudicetur." See also Lex Sal., 41, 8-10, and Par-
dessus, I, 49, ch. 9.
79
The Merovingian interest in legislating for these people sug-

gests they were more important and more numerous than the
few narrative accounts involving them might indicate. The ra-
tionale behind the official legal concern with these groups was
based upon their importance as functioning armed forces within
the Merovingian military and upon their institutional position,
first under the empire and later under the monarchy. The laeti
were a corpus publicum of the empire and probably came under
the direct control of the Merovingian monarchs in the same man-
ner as did the fisc, also a corpus publicum.15 The similar treat-
ment of the milites and the laeti in the Merovingian legislation
implies that the former were also acquired by the monarchs. This
would perhaps tend to explain the hostility at Tours between a
count who sought to promote his own interests and milites who
had a special tie to the king.
If the hypothesis above is correct — that the milites in Mero-
vingian Gaul were descendants of imperial milites in title, voca-
tion, and blood — then the milites serving under their tribunus
at Besan?on may perhaps be identified as the descendants of
the milites Batavi who were stationed in that city during the
fifth century.16 Procopius remarks that in his day (about 550)
some Roman units had survived in Gaul in the border areas.17
(Besan£on, approximately seventy-five miles west of the Rhine,
would seem to fit that description.)
As has already been mentioned, Theuderic also used armed
retainers from the court, custodes, and a magnate with his own
followers against the intrigues of Columbanus. In addition, Theu-
debert provided custodes and Chlotar gave the exiled mission-
ary an armed escort of comites from the court. The custodes re-
ferred to in these texts are treated similarly to those mentioned
throughout the sixth century and seem to be some type of per-
manent armed group used by all the Merovingian kings. The
comites at Chlotar's court do not seem to have been the kind of
16
H. Pirenne, "Le fisc royal de Tournai," Melanges Ferdinand Lot (Paris,
1925), p. 648, and Wallace-Hadrill, Long-Haired Kings, p. 153.
lf>
Notitia Dignitatum, oc., XXXVI, 5, n. 1: "Milites Batavi, Vesontione."
17
Procopius, H.W., V, xii, 16ff.
80
counts who administered cities, but rather were royal com-
panions serving the king,18 As further proof of the diverse nature
of the Merovingian military organization, both Theuderic and
Theudebert had organized water transport on at least the most
important rivers in their territory, and Theuderic maintained
stopping places at erstwhile imperial fortifications along the
Roman roads in his kingdom.
In 610, the year following the Columbanus episode, Theude-
bert raided Alsace, which belonged to his brother. Soon after this,
the brothers agreed to meet at the village of Seltz to settle their
dispute. Theuderic arrived with two units — each of which Frede-
gar extravagantly asserts comprised 10,000 men — and Theude-
bert was accompanied by a large force of Austrasians with which
he surrounded his brother's troops, thus ensuring for himself a
favorable settlement. While the brothers were engaged at Seltz,
Theubert's Alaman supporters raided Theuderic's territory
around Avenches. The Counts Abbelin and Herpin, with other
counts from the district, led their followers against the Alamans
but were defeated and sustained severe losses.19
Though Fredegar exaggerates the size of Theuderic's force at
Seltz, it was probably impressive to elicit such a reaction; and
the Austrasian force which surrounded it must have been huge
indeed. The term used to describe Theudebert's army—magnus
exercitw Austrasiomm — might even suggest a special levy of
some kind. Theuderic's troops, by contrast, are called escaritas
and this is the only appearance of the term in medieval literature.
Escaritos seems to have been replaced by the word scant (the
members of a scara) .20 These scant, which were picked or special
troops, are mentioned in the sources throughout the later Mero-
vingian era. The escaritos commanded by Theuderic in 610 may
be identified as a part of the score Burgundiae which in turn may
have been a part or all of the standing army of Burgundy dis-
cussed in the previous chapter. Theuderic's force which defended
18
See the bibliography cited in ch. Ill, n. 60, above.
" Fred., IV, 37.
20
J. F. Niemeyer, Mediae Latinitatis Lexicon Minus (Leiden, 1963), pp.
943-944. Cf. J. F. VerBruggen, "L'Annee et la strategic de Charlemagne,"
Karl der Grosse (Dusseldorf, 1965), I, 421-422.
81
the pagus of Avenches is said to have included many counts

(comites), which implies that the term is being used here not in
the formal sense of administrative and military leader of a
civitas or pagus, but rather in the less technical sense of com-
panion — similar perhaps to the comites who served at the court
of Chlotar II.
In May of 612 Theudric assembled an army from all parts of
his kingdom at Langres, a long-established Merovingian military
base. Marching north from Langres through Andelot, he took the
fortress of Naix and the fortified city of Toul — successes which
should not be overlooked as evidence of his troops' capabilities
in this type of warfare. While Theuderic traveled north through
Andelot, Theudebert moved south with the forces of Austrasia,
and the brothers' armies did battle at Toul. Theudebert was
soundly defeated and he retreated by way of Metz and the
Vosges to Cologne. Theuderic led his army through the Ar-
dennes to Zulpich. Meanwhile, Theudebert, having been failed
by the armies of Austrasia like his grandfather Sigibert I almost
two generations earlier, summoned the tribes living beyond the
Rhine. With Saxons, Thuringians, and other trans-Rhenish war-
riors under his command, Theudebert met Theuderic's forces at
Zulpich. The carnage was so great that the chronicler Fredegar
contends "men had nowhere to fall." Theudebert's forces were
once again defeated; the king attempted to escape, but he was
betrayed by his supporters and taken into custody by Berthar, a
magnate who had pursued him in his flight.21
The mass scale of the battles at Toul and Zulpich has few if
any parallels in Merovingian history. Theuderic called out men
from all his lands, probably including the many local levies from
his cities in Aquitaine and Neustria. At Toul Theudebert com-
manded only a force from Austrasia (exercitus Austrasiorum)
which was apparently smaller and less strong than the one he
called out in 610 (magnus exercitus Austrasiorum'). Both these
21
Fred., IV, 38: "Anno XVII regni sui Lingonas de universas regni sui
provincias mense Madio exercitus adunatur, dirigensque per Andelaum,
Nasio castra ceptuna, Tollo civitate perrexit et cepit. Ibique Theudebertus
cum Austrasiorum exercitum obviam pergens, Tollensem campaniam con-
fligunt certamine." V. Columbani, I, 28, and cf. L.H.F., eh. 38.
82
armies would seem to exclude Theudebert's local levies from

Aquitaine and Neustria, as well as the territorial levy of Cham-
pagne. In any event, no details are known about the battle of
Toul except that Theuderic was victorious. Not much has been
recorded about the battle of Zulpich either, aside from the ac-
counts of the great carnage. If Fredegar is not unduly exaggerat-
ing when he says the dead stood in their ranks unable to fall
because there was no room, then it does not seem unreasonable
to conclude that significant numbers on both sides were fight-
ing on foot. This is a plausible assumption in light of the nature
of the forces involved. Theudebert probably commanded what
remained of the exercitus Austrasiorum which had been defeated
at Toul, but his primary support came from the trans-Rhenish
peoples, including Saxons who may well have provided the bulk
of his infantry. Theuderic had called up a vast force from
throughout his kingdom and there can be little doubt that many
pauperes and inferiores had served at Zulpich. Some of Theu-
deric's forces, however, were mounted, such as Berthar's contin-
gent which captured Theudebert after a hot pursuit. Berthar,
incidentally, was given Theudebert's horse as a reward for his
valuable service.
Shortly after his capture, Theudebert was put to death by his
brother, who soon followed him to the grave, leaving his king-
dom to his four young sons. Their great-grandmother Brunhild
tried to serve as their regent, but Chlotar II, the only remaining
adult Merovingian ruler, sought to take advantage of his cousins'
youth by acquiring the allegiance of a group of their Austrasian
and Burgundian magnates. Chlotar was able to occupy part of
Austrasia because several Austrasian magnates and their follow-
ers defected from Brunhild. Brunhild's efforts to secure trans-
Rhenish support were thwarted by the defection of Warnachar,
mayor of the palace for Burgundy and Brunhild's envoy to the
Thuringians. A number of the magnates, keeping their defection
secret, accompanied Brunhild and her great-grandsons into the
field against Chlotar. In the battle with the Neustrians some of
Brunhild's troops simply deserted, while others captured the
83
queen and her great-grandsons and handed them over to Chlotar

who had them killed.22
With the death of Theudebert, Theuderic, and their heirs,
Chlotar ruled all of Merovingian Gaul. Soon after achieving
this position he issued the Edict of Paris by which he hoped to
establish peace and order in his kingdom. Those magnates who
had helped him gain control in Austrasia and Burgundy were
rewarded with gifts and important offices.23 Aware that the op-
position of these powerful men could cause serious trouble, Chlo-
tar and later his son Dagobert strove to avoid such difficulties
by purchasing the loyalty of their leudes with offerings of offices,
estates, and at times even church lands. Dagobert, for example,
is alleged to have alienated church land so that he might ensure
or perhaps secure the military support of certain magnates.24 The
memory of 605 when Theuderic II's magnates had refused to
fight and killed the mayor of the palace, and of 612 when the
defection of the magnates had brought about the death of Queen
Brunhild and Theuderic's heirs may have motivated Dagobert's
largesse when it came to maintaining the military support of
the important men in Gaul. At any rate, Chlotar and Dagobert
proved to be more adroit in dealing with their magnates than
were Theudebert II and Theuderic II.
Chlotar and Dagobert were ruthless, however, in eliminating
magnates who were troublemakers. Chlotar saw to it that Godi-
nus — the son of Warnachar, the former mayor of the Burgun-
dian palace — was killed along with a band of his followers. The
task, which was first assigned to Duke Arnebert who failed, was
finally accomplished by a magnate named Chramnulf and a do-
mesticus by the name of Waldebert, when they and their follow-
ers cut down Godinus and his men at a villa near Chartres. The
magnates Brodulf and Erminarius were killed because they
plotted to overthrow Dagobert. Aighyna, a Saxon who later be-
23
Fred., IV, 38, 39, 40, 41, 42.
28
Ibid., IV, 42; Wallace-Hadrill, Long-Hcrired Kings, pp. 214-216.
"Mir. Martini Abbotts, ch. 7: "Rex Dagobertus Francorum rei publicae
princeps cum multis et variis bellorum eventibus premeretur, de coenobiis
sanctorum multa abstulit, quae suis militibus est." See Boutaric, Inst. milit.,
p. 68, and Bachrach, "Charles Martel," pp. 66-72.
84
came duke of Gascony, used his pueri to murder Erminarius,
while the Dukes Amalgar and Arnebert, and Willebad, the patri-
cian of Burgundy, did away with Brodulf. Arnebert also got rid
of Boso of fitampes. Aighyna, Amalgar, and Arnebert all held
the title dux, but were generally at the court and held no ad-
ministrative position at the time they were assigned to kill those
magnates whom Chlotar and Dagobert considered dangerous.25
From these examples of magnates who were eliminated it
should not be concluded that Chlotar and Dagobert were vic-
timized by rebellious potentes. On the contrary, both these mon-
archs enjoyed strong support from the important men of Mero-
vingian Gaul. Though the transfer of power after the death of a
ruler was usually a critical period, Dagobert succeeded his father
with relatively little trouble by soliciting the aid of his magnates
(leudes). This may call to mind the transition from the reign of
Theuderic I to that of Theudebert I in 534, when the latter
gained the support of his magnates (leudes) with gifts and was
thus able to secure his throne in the face of his uncles' efforts
to deprive him of it. Such comparatively easy transfers of power
were rare during the Merovingian era, since the magnates often
sought to take advantage of these situations by selling their sup-
port to opposition forces, or by creating an opposition themselves.
Upon learning of Chlotar's death, Dagobert gathered his follow-
ers and their retainers in exercito, rode into Neustria, and secured
the loyalty of the magnates there; shortly after, he repeated that
process in Burgundy. Dagobert gave Aquitaine to his brother
Charibert; Fredegar notes that this grant included the cities of
Toulouse, Cahors, Agen, Perigueux, and Saintes, all of which
had local levies with the possible exception of Cahors. Sometime
after obtaining Aquitaine from Dagobert, Charibert brought Gas-
cony under his control.26
Because Chlotar and Dagobert faced little internal opposition,
both were able to spend the bulk of their military energy beyond
the borders of Gaul. For example, in either 622 or 623 Dagobert
led a large force across the Rhine to crush the Saxon rebellion
25
Fred., IV, 54, 55, 58.
28
Ibid., IV, 56, 57.
85
against Merovingian domination. When Dagobert realized that

his men were imperiled by the Saxon army, he sent word to his
father for help. Upon receiving this request for aid, Chlotar
roused his troops and moved from his camp in the Ardennes to
the east. After traveling all night and part of the morning, Chlo-
tar's force reached Dagobert's camp and pitched its tents. Angered
by the remarks of the Saxon Duke Bertoald, Chlotar urged his
horse into the river separating the Merovingian and Saxon camps
and charged the enemy. Dagobert led his mounted followers
across the river in support of his father and together they de-
feated the Saxons and brought about the death of Bertoald.27
When Dagobert had firmly secured his position within his
own territory, he went to war against Samo, king of the Slavs,
in 630. The Merovingian monarch organized a three-pronged at-
tack against Same's kingdom: the Lombards advanced from Italy,
the Alamans from their Transjuran settlements, and the general
levy of the Austrasian kingdom from the north of Gaul. Although
the Lombards and the Alamans were successful, the Austrasians
laid siege to the Slav fortress of Wogastisburg where they were
decisively defeated in a three-day battle.28
The same year Dagobert arranged to send an army into Spain
to support Sisenand, the Visigothic king, in his struggle for the
throne; in return for this military aid he was to receive a 500-
pound gold dish. Dagobert ordered the general levy of Burgundy
and the local levy of Toulouse, which had come under his control
after the death of his brother Charibert, to be called up. The
levy of Toulouse, led by Venerandus and Abundantius, moved
toward Saragossa but the Goths, seeing their peril, recognized
Sisenand as king. Neither the general levy of Burgundy nor the
local levy of Toulouse seems to have seen any action.29
"
28
L.H.F., ch. 41.
Fred., IV, 68.
28
Ibid., IV, 73: ". . . Dagobertus . . . exercitum in ausilium Sisenandi
de totum regnum Burgundiae bannire precepit." Cf. IV, 68: "Dagobertus
. . . iubet de universum regnum Austrasiorum . . . movere exercitum. . . ."
The former group seems to be a general levy from Burgundy and the latter
a general levy from Austrasia. For the levy of Toulouse, IV, 73: "Abundan-
cius et Venerandus cum exercito Tolosano tanto usque Cesaragustam civi-
tatem cum Sisenando acesserunt. . . . Abundancius et Venerandus cum
86
In the following year Dagobert prepared to go against the
Wends and summoned a force from Austrasia to take part in the
battle. He had with him select detachments (scara) from Neustria
and Burgundy, as well as numerous dukes and counts, presum-
ably with their own personal followings. The campaign was
canceled, however, when Dagobert arranged for the Saxons to
be used against the Wends.30
Dagobert called up the general levy of Burgundy in 635 for
the purpose of crushing a Gascon revolt. This force, under the
command of Chadoind, a referendary, included ten dukes and
Willebad, the Burgundian patrician, who probably commanded
the Burgundian scara. Chadoind's men defeated the Gascons,
who retreated into the mountains where they hid among the rocks
and gorges while the Merovingian army devastated Gascony. The
troops would have returned home with minimal losses had it not
been for the negligence of Duke Arnebert, but he and his unit, in
which were a number of important people (seniores and nobili-
ores), were massacred in the valley of the Soule. (It should be
noted that none of the ten dukes who served under Chadoind
in this campaign held administrative positions in specific terri-
tories. Aighyna was subsequently appointed duke of Gascony,
but Arnebert, Amalgar, Chramnelen, and Barontus all served at
Dagobert's court. The other five dukes are mentioned in no
other Merovingian texts.)31 In the same year as this Gascon up-
rising, Dagobert also threatened to send a Burgundian army into
Brittany, but the mere threat brought the Bretons to terms.32
The heterogeneity of the Merovingian military persisted into
the seventh century. That this diversity continued is proven by
the number of different combinations of terms Fredegar uses to
describe the Austrasian military: magnus exercitus Austrasiorum,
exercitus Austrasiorum, exercitus de Auster, exercitus de univer-
exercito Tolosano munerebus onorati revertunt ad propries sedibus." See
Gregory, Hist., IX, 31, for an earlier mention of the levy of Toulouse. Cf.
Thompson, The Goths in Spain, pp. 171-172.
30
Fred., IV, 74.
31
Ibid., IV, 54, 58, for Arnebert; 58, 73, 90, for Amalgar; 67, for Barontus;
90, for Chramnelen; and 54, 55, for Aighyna.
32
Ibid., IV, 78.
87
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2008
MEROVINGIAN
For Evaluation Only. MILITARY ORGANIZATION
sum regnum Austrasiorum, and universi leudes in Auster in exer-

cito. The last of these phrases indicates an army of all the mag-
nates of Austrasia, whereas exercitus and magnus exercitus Aus-
trasiorum signify Austrasian forces of varying sizes. The exercitus
from the universum regnum Austrasiorum is an army composed
of inhabitants from all of Austrasia, and the exercitus de Auster
seems to be territorial in nature. These Austrasian groups may be
compared with the forces from Burgundy, which Fredegar calls
an exercitus de Burgundia (a territorial levy), exercitus de uni-
versum regnum Burgundiae, and a scara de elictis viris fortis de
Burgundiae (a group of select warriors which probably formed
part of the standing army of Burgundy traditionally commanded
by the patrician). The army of the entire Burgundian kingdom
appears to have been some kind of general levy. There was also a
scara de elictis viris fortis de Neuster which seems to have been
some type of standing force. In the 570s and early 580s Chil-
peric commanded field forces which were not local city levies,
general levies, custodes, or milites; for that matter they do not
seem to have had any particular designation, but in view of their
functions it is reasonable to conclude that they were a permanent
military unit.33 This force, so often seen serving under Chilperic,
may well have been a Neustrian scara which went unlabeled by
Gregory who was less apt than Fredegar to use germanisms.
The armies of Austrasia and Burgundy and the Neustrian
scara were not the only fighting forces utilized by the Merovin-
gian monarchs. The kings ruling in the northeastern parts of Gaul
were able to obtain the military support of peoples from beyond
the Rhine to compensate for the failures of the Austrasian armies
during this period. This policy of enlisting the help of trans-
Rhenish troops was followed by Theudebert I, Theudebald I,
and Sigibert I during the sixth century and endured to the
seventh. Theudebert's Alaman warriors from the eastern parts of
what is now Switzerland were probably the descendants of those
who had remained under Merovingian control after Clovis's vic-
tory of 506. Alamans, formed into a levy of some kind, fought
88
Gregory, Hist., IV, 23; V, 12, 24; VI, 31.
88
under the Alaman Dukes Leutharius and Buccelin in Theudebald
I's campaign against Italy in 554, and continued to function well
into the seventh century.
The local levies which were so prominent during the civil wars
of Clovis's grandsons continued to be used by their successors,
but in a much more limited manner. The reason for their limited
value is that not only does a large amount of military action
occur in the Austrasian and Burgundian kingdoms (which do
not seem to have developed local levies at this time), but the
extensive military action which took place beyond the borders
of Gaul seems to have been beyond the competence of these
these units. As has already been mentioned, the erstwhile im-
perial elements of the Merovingian military, especially the mill-
tes, continued to function, as did the custodies, whose basis of
organization seems impossible to ascertain, the river fleets on
the major waterways, and the fortified positions along major
roads. Fredegar and the author of the Liber Historiae Francorum
used the terms comes and dux much more loosely than did Greg-
ory, and from their reports it may be suggested that many of
the men who held such titles in this period functioned in much
the same way as the untitled leaders of armed bands like Clau-
dius who resided at the court of Clovis's grandson Guntram.34
The magnates and their bands of armed followers also provide
another example of continuity in the Merovingian military. Dur-
ing this period of the last ruling Merovingians, the magnates
still exerted as much influence on military and political affairs
as they had in the past. Often their loyalty and support had to be
purchased with expensive "gifts" as was the case when Chlotar II
needed his magnates' aid to oppose the attack of Wintrio in 593.
On occasion, however, the magnates were unwilling to lead their
armed followings in the king's support, as in 605 when Theu-
deric's leudes refused to fight, or in 612 when the magnates who
had promised their support to Theuderic II's sons and their re-
gent, Brunhild, were bribed by Chlotar II and deserted the battle-
34
See ch. Ill, n. 60, above.
89
field, thus causing the death of the queen and her great-grand-
children. This was not altogether different from the desertion,
slightly more than a century earlier, of Ragnachar's leudes which
had cost him his kingdom and his life.
The monarchs also continued to support bands of personal
armed followers: men who served them at the court wherever it
might move. Several armed groups from Theuderic II's court saw
action in the Columbanus affair, Chlotar II used his retainers to
restrain his son Dagobert, and Theudebert II is said to have lost
his life because of the faithlessness of his personal followers who
allowed him to be captured by one of his brother's magnates.35
These armed followings in attendance upon the king, including
men called duces and comites in the sources, may be compared
to the bands employed by the magnates, for the monarch after
all was the most powerful magnate in the realm.
There seems to have been a certain continuity in the tactical
maneuvers of the Merovingian military as well. Siege warfare,
though less common than during the previous periods of Mero-
vingian rule, was nevertheless used in the investment of Orleans
and Wogastisburg. The relatively limited number of sieges was
probably due to a change in the theaters of warfare from the
more Romanized parts of Gaul with their fortresses and walled
cities to the Austrasian hinterland and areas beyond the Rhine.
Mass infantry tactics, though generally uncommon, appear once
again when less Romanized barbarians from beyond the Rhine
are introduced into Merovingian military affairs. Mounted forces,
so conspicuous in Theuderic Fs conquest of Thuringia in 531 and
in the campaigns of Desiderius and Boso a generation or so later,
were still evident in the operations undertaken during the reigns
of the last ruling Merovingians.
The various types of levies, personal followings, and standing
forces and the use of infantry, cavalry, siege operations, fortified
positions, and organized transportation and communication in-
dicate that the Merovingian military establishment and its tactics
35
Gesta Dagoberti, ch. 8, and V. Columbani, I, 28.
90
were marked by both diversity and continuity during the period
of the last ruling descendants of Clovis. The imperial and non-
Frankish barbarian elements which lent support to the Mero-
vingian leaders also illustrate this same continuity with the past
as well as the debt the Merovingian military owed to groups in
society other than the Franks.
91
CHAPTER V
The Rois Faineants

and the Mayors
of the Palace: 638-751
N OT LONG after the death of King Dagobert in 638, his son

Clovis II succeeded him as ruler of Neustria; Sigibert II,
his elder son, had been king in Austrasia since 632. Because both
monarchs were children, their reigns were supported and con-
trolled by the magnates of their kingdoms. In Austrasia, Peppin
of Landen, the mayor of the palace, and Bishop Chunibert of
Cologne, who were the leaders among the magnates, assisted
Sigibert in his political and military duties as king. According to
Fredegar, the bishop and the mayor "skillfully and with suitable
inducements drew the Austrasian magnates to their cause, and
by treating them generously they gained their support and knew
how to keep it."1
When Peppin died early in the year 639, a struggle erupted
between Peppin's son Grimoald and Otto, the son of the domesti-
cm Uro, for supremacy among the magnates. The weakness of
the king's power in Austrasia was made manifest in 639 when
Duke Radulf of Thuringia rebelled against royal control. Grim-
oald, on behalf of Sigibert (who was about eleven years old at
that time), summoned the Austrasian magnates to put down the
revolt. In addition to the personal followings of these leudes, a
'Fred., IV, 75, 79, 85: ". . . Pippinus cum Chuniberto . . . omnesque
leudis Austrasiorum secum uterque prudenter et cum dulcedene adtrad-
gentes, eos benigne gobernantes eorum amiciciam constringent semperque
servandum."
92
Rois Faineants AND MAYORS OF THE PALACE
general levy of the trans-Rhenish peoples living under Austrasian

control and the local levies of the Saintois and of Mainz were
also called to serve.2 Grimoald, as Sigibert's commander, led the
heterogeneous force across the Rhine and attacked a magnate
named Fara, a supporter of Radulf and the son of the Agilolfing
noble Chrodoald who had been killed on Dagobert's orders in
629. Fara and most of his followers were massacred; the few who
survived were made slaves. Radulf, meanwhile, called up a
levy from throughout Thuringia to oppose the Austrasian invasion
and prepared for a siege in a stockade on a hill above the banks
of the Unstrut in Thuringia. Sigibert's forces pitched their tents
and laid siege to Radulf s position; however, Grimoald, who led
the young king's army, became aware that he did not have the
loyalty of all the magnates. Realizing that this lack of allegiance
might endanger not only their position, but the life of the king as
well, Grimoald and Duke Adalgisel made a great effort to guard
Sigibert. The Austrasian monarch apparently did not have his
own personal armed following to provide such protection.3
Despite the lack of solidarity in the army, Bobo, the duke of
Auvergne and a loyal supporter of Grimoald, commanding some
of Adalgisel's men, Count Innowales of the Saintois with the
local levy of his district, and a large number of the remaining
troops stormed the entrance to Radulfs fortifications. Radulf,
knowing that several of Sigibert's dukes would not go against
him, led his followers in a counterattack. Among those who did
not support the king were the local levies of Mainz. Sigibert's
forces suffered severe losses and had to obtain permission from
Radulf to withdraw safely.4
2
Ibid., IV, 86, 87: ". . . Radulfus cux Toringiae vehementer contra
Sigybertum revellandum disposuissit, iusso Sigyberti omnes leudis Aus-
trasiorum in exercitum gradiendum banniti sunt. Sigybertus Renum cum
exercito transiens gentes undeque de universis regni sui pagus ultra Renum
cum ipsum adunati sunt." For the levies of the Saintois and of Mainz see n.
4 below.
3
Ibid.
* Ibid.: "Bobo dux Arvernus cum parte exercitus Adalgyseli et Aenovales
comex Sogiontinsis cum paginsebus suis et citeri exercitus manus plura con-
tra Radulfum ad portam castri protenus pugnandum perrexerunt. . . .
Macansinsis hoc prilio non fuerunt fedelis."
93
In the struggle for power in the first few years following

Dagobert's death, the focus of the sources changes radically:
more emphasis is placed on the activities of the mayors and the
magnates than on those of the kings. In particular, Fredegar
indicates how various magnates became mayors of the palace
and acquired the support necessary to secure and defend their
positions. Aega, the mayor of the palace for Neustria and an able
man according to Fredegar, strengthened his position by alienat-
ing lands from the fisc and giving them to the magnates. After
Aega's death in 642, Erchinoald succeeded to the mayorship of
Neustria; in the following year, Erchinoald's friend and follower
Flaochad became mayor of the palace in Burgundy and gained
the support of the Burgundian magnates by promising in writing
and by oath to protect them all in their honors and dignity and
to give them his assistance. As noted above, Peppin made con-
cessions and presented gifts to the Austrasian magnates to gain
their loyalty. In short, Fredegar provides a glimpse of the process
by which a magnate purchased the support of powerful men
whose wealth and armed followers were vital to the exercise of
power on a large scale. Grimoald's inability to unite effectively
such men behind him was responsible in part for his defeat by
Radulf in 639 and ultimately for his death in 656.5
In Burgundy Flaochad's position was endangered by the pa-
trician Willebad, and in May of 643 Flaochad summoned a placi-
tum at Chalons where Willebad was to be condemned. However,
Willebad appeared with his followers, and discovering his peril,
took Amalbert, Flaochad's brother, as hostage and escaped. In
September of that year Flaochad returned from Paris to Bur-
gundy with the child-king Clovis and a group of Neustrian
6
Ibid., IV, 80: "Facultatis pluremorum, quo iusso Dagoberti in regnum
Burgundiae et Neptreco inlecte fuerant usurpate et fisci dicionebus contra
modum iusticiae redacte, consilio Aegane omnibus restaurantur." That Aega,
despite his avarice ("eo quod esset avariciae deditus"), should give lands
of the fisc which he controlled to those from whom the late King Dagobert
had taken them suggests that the mayor was obtaining some benefit from
such transactions. See IV, 84, for Erchinoald, and 89: "Flaochadus cumtis
ducibus de regnum Burgundiae seo et pontefecis per epistolas etiam et
sacramentis firmavit, unicuique gradum honoris et dignetatum seo amiciciam
perpetuo conservarit." See Ewig, "Teilreiche," pp. 114-120, and Bachrach,
"Charles Martel," pp. 68-72.
94
magnates and their followers. Willebad, who realized he would

have to fight, called together a general levy from his patriciate,
his followers among the ecclesiastical and lay magnates, and what
may have been a part of the Burgundian standing army which
traditionally served under the patrician's command.6
Willebad advanced with his men to Autun where, after some
machinations, battle was joined with the combined forces of
Flaochad, Amalgar, Chramnelen, and Wandelbert. Erchinoald,
his Neustrian followers, and the Burgundian magnates who had
promised to support Flaochad remained spectators to the battle.
Berthar, a Transjuran Frank and count of the palace, was the first
to attack Willebad, but a Burgundian named Manaulf and his
followers charged Berthar and badly wounded him. However,
before the coup de grace could be delivered, Berthar's son
Chaubedo rode full tilt to his aid, knocking Manaulf to the
ground with his lance and killing his followers. In the ensuing
battle, Willebad and many of his men were slain. Those mag-
nates who had not entered the battle plundered Willebad's camp
and seized a great deal of booty including a large number of
horses.7
In Neustria the struggle for power continued throughout the
second half of the seventh century. Erchinoald died in 657 and
was succeeded by Ebroin. After little more than a decade, Eb-
roin's support dwindled, and a group of Neustrian magnates
finally sent him into exile and chose as their leader Duke Wulf-
oald who was at that time the power behind Childeric II, the
Austrasian king and the son of Clovis II. Wulfoald, however,
alienated a faction of Neustrian magnates including Amalbert,
Ingobert, and Erchinoald's son Leudesius, who succeeded Wulf-
oald when he fled to Austrasia after a Frank named Bodilo
murdered King Childeric and his queen.8 A short time later,
Ebroin, exiled at Luxeuil, called together his erstwhile supporters
(socii) and with a large following moved north against Leudesius
* Fred., IV, 89, 90: "Willibadus . . . colligens seoum pluremam multi-
tudinem de patriciatus sui termenum, etiam et pontevecis seo nobelis et
fortis quos congrecare potuerat. . . ."
7
Ibid., IV, 90.
8
Fred, con't., ch. 2.
95
and his puppet, King Theuderic IV, who were residing at a villa
somewhere between Compiegne and Corbie. After a long hard
ride from Luxeuil, Ebroin reached Pont-Saint Maxence on the
Oise, found the unsuspecting guards (custodes] at the bridge
asleep, and had them killed. Once Ebroin's men had crossed the
Oise, word of their advance seems to have reached Leudesius
who sent out a force to delay them. Meanwhile, Leudesius and
his socii fled north with the king and the royal treasure. Ebroin
destroyed those sent against him and pursued the royal party.
At Basieux, some fifty miles from Pont-Saint Maxence, Leudesius
abandoned the treasure so that his party might move more ra-
pidly. But within another fifty miles Ebroin's men overtook and
captured the royal treasure and the king. Leudesius and many
of the magnates who supported him were put to death; others
fled into exile south of the Loire.9
Early in the last quarter of the seventh century, a Merovingian
king of strength and ability projected himself into this morass of
royal puppets and contending magnates. Dagobert II, who had
been banished to Ireland as a young man in the mid-650s, re-
turned in the 670s with the aid of Bishop Wilfrith of York and
with an armed following substantial enough to make credible his
long hair and his claim to the throne. His efforts to restore royal
power were opposed and, although he destroyed the strongholds
of several magnates, he ultimately fell before the swords of an
enemy faction. In this struggle for predominance Dagobert
seems to have been not so much a king as just another magnate
vying for supremacy — and not even the most powerful one as
his ancestors had been.10
It should be noted that Dagobert, by having his own armed
followers, seems to have been a rarity among the rois faineants.
8
L.H.F., ch. 45; Fred, con't., eh. 2.
10
V. Wilfrithi, ch. 28: "Et sic sanctus pontifex noster perfecit, suscipiens
eum de Hibernia venientem, per arma ditatum et virlbus socioram elcvatum
magnifice ad suam regionem emisit." See also ch. 33, in which Dagobert is
described as a "dissipator urbium" which Wallace-Hadrill, Long-Haired
Kings, p. 238, n. 3, conjectures to mean that he destroyed city walls to keep
them from being used as centers of resistance. For magnates in fortified
cities cf. Fred, con't., ch. 4. Also in ch. 33 of the V. Wilfrithi, a bishop is
described as leading a large warband.
96
Rots Faineants AND MAYORS OF THE PALACE
For example, we have seen that as early as 639 Sigibert II had

to rely on Duke Grimoald and Duke Adalgisel for his safety;
Childeric II and his wife do not seem to have been protected
by ubiquitous pueri like those who surrounded King Guntram.
Yet oddly, among the formularies which were used in the chan-
ceries of the rois faineants, there are documents referring to the
installation of antmstiones — armed followers including pueri
who swore to serve the king in a military or paramilitary capacity
in the centenae and probably at the palace as well. Since the
kings during this period did not benefit from the assistance which
these men were supposed to provide in return for support, then
perhaps the mayors of the palace who controlled the royal fisc
and chancery were served by these jideles cum. armis stii?,11
The struggle for power in Neustria had its counterpart in
Austrasia. Grimoald was killed by a faction of magnates be-
cause he tried to place his own son on the throne instead of a
Merovingian. After Grimoald's death, Wulfoald seems to have
been the most successful magnate in Austrasia, and, as noted
above, he even exercised considerable influence in Neustrian
affairs for a time. At his death, and with "the disappearance of
the kings [from Austrasia] Martin and Peppin, the younger son
of the deceased Ansegisel, ruled over Austrasia." The Arnulfings
once again established their supremacy as the premier magnates
of Austrasia, and in fact ruled without a king.12
Martin and Peppin led a large force of Austrasians (exerci-
tum plurimum Austrasiorum commotum) against the Neustrian
mayor Ebroin but, in a battle at Bois du Fays near Rethel (Ar-
dennes), Ebroin's men were victorious. Martin took refuge at
Laon with his supporters (saddles) and his followers (socii), but
they were massacred when they were tricked by Ebroin into
11
Marculf, form., I, 18: "Rectum est, ut qui nobis fidem pollicentur inle-
sam, nostro tueantur auxilio. Et qui illi fidelis, Deo propitio, noster veniens
ibi in palatio nostro una cum arma sua in manu nostra trustem et fidelitatem
nobis visus est coniurasse: propterea per presentem preceptum decernemus
ac iobemus, ut dienceps memoratus ille inter numero antruscionorum con-
putetur. Et si quis fortasse eum interficere presumpserit, noverit se wire-
gildo suo soledos sexcentos esse culpabilem."
13
Fred, con't., ch. 3; L.H.F., ch. 46.
97
leaving. A Frank named Ermanfred and his socii then murdered

Ebroin and fled to Austrasia where they were given protection
by Peppin.13
After this campaign, the Neustrian magnates appointed War-
atto mayor; he sought a policy of peace with the Austrasian
mayor, but his son Ghislemar would not accept this policy, Ghis-
lemar seized the fortified town of Namur, one of Peppin's strong-
holds, and killed many of Peppin's supporters who garrisoned
the town.14
Shortly after this futile effort to establish peace, both Ghisle-
mar and Waratto died, and Berchar, the latter's son-in-law, be-
came mayor. Berchar, however, angered a number of Neustrian
magnates, some of whom defected and joined Peppin in Austra-
sia. With the aid of these defectors and his own followers, Peppin
was able to defeat Berchar, thus extending his authority over
Neustria as well as Austrasia. St. Boniface, writing toward the
middle of the eighth century, argues that church property had
been taken over by secular magnates for at least sixty or seventy
years and, although no specific text connects this spoliation of
church lands with Peppin's rise to power, it is not unreasonable
to suggest that he took over church lands and then gave them to
his followers or would-be followers in order to secure and main-
tain their support.15
During the last decades of the seventh century Peppin gradu-
ally secured his position in Austrasia. He made his son Drogo
duke of Champagne and his other son, Grimoald, mayor of the
13
L.H.F., ch. 46: ". . . hii duces in odium versi contra Ebroinum, exer-
citum plurimum Austrasiorum eommotum, contra Theudericum regem et
Ebroinum aciem dirigunt. Contra quos Theudericus et Ebroinus cum hoste
occumint. . . ." Ewig, "Teilreiche," pp. 123-135.
" L.H.F., ch. 47; Fred, con't, ch. 4.
M
St. Boniface, ep., 50: ", . . de aecclesiastica religione, quae iam longo
tempore, id est non minus quam sexaginta vel septuaginta annos, calcata et
dissipata fuit. . . ." The historical worth of this letter has been attacked,
but the despoliation of church lands by monarchs and mayors before the
accession of Charles Martel is indicated throughout the period. See the evi-
dence presented by E. Lesne, La Propriete ecclesiastique en France (Paris,
1910), II, 2-6. See also Goffart, Le Mans, pp. 6ff, and the review of
Goffart's work by Wallace-Hadrill, Speculum, XLIII (1968), 721-722.
Bachrach, "Charles Martel," pp. 68-69. L.H.F., ch. 48; Fred, con't., ch. 5.
98
palace to King Childebert, Peppin went to war against the Fri-

sian Duke Radbod, defeated him, drove him from the fort at
Duurstede, and returned home with a great deal of booty. Not
only did Peppin increase his wealth by such expeditions, but he
also strengthened his ability to keep the loyalty of his followers.18
Some details are known about one of Peppin's supporters, the
domesticus Dodo. This man was well off economically and had in
his service a band of private retainers (pueri) whom he provided
with mail coats, helmets, shields, lances, swords, and bows and
arrows. During the era of the rois faineants it was the armed
followers (socii, pueri, sodales, and satellites) of magnates like
Dodo who formed the fighting forces in Austrasia and probably
in Neustria. To be a successful mayor of the palace it was neces-
sary to secure control of the forces these magnates commanded,
and Peppin was most efficient at accomplishing this.17
Peppin's house suffered severe setbacks, however, when Drogo
died of a fever and Grimoald was murdered. Grimoald's young
son Theudoald was made mayor in his father's place. When Pep-
pin died shortly thereafter, in 714, Theudoald was supported by
his grandfather's leudes, men like Dodo, who provided much of
the armed force to sustain him as mayor. But Theudoald proved
to be a poor leader and the magnates soon transferred their al-
legiance to a certain Ragamfred and made him mayor. Ragam-
fred strengthened his position by allying with Radbod.18
As mayor of Austrasia, Ragamfred was primarily opposed by
Peppin's bastard son Charles, who was later to obtain the sobri-
quet Martel. When Charles first began his struggle for power,
he had only a small following and was almost destroyed. In 715
Ragamfred and Radbod led their warbands against Charles and
13
L.H.F., chs. 48, 49: Fred, con't., ch. 6.
" V. Lamberti, ch. 11: "In diebus illis erat Dodo domesticus iam dicti
principes Pippeni, proprius con sanguinius eorum qui interfecti fuerant, et
erant ei possessiones multae et in obsequio eius pueri multi." Also ch. 13:
"Et erat multitude copiosa virorum pugnatorum ad bellandum, et erat
induit lurices et cassidis, clipeis et lanceis gladiisque precincti et sagittis cum
pharetris."
u
L.H.F., chs. 48, 49; Fred, con't., chs. 6, 8: "Franci . . . contra Theu-
doaldum et leudis Pippino quondam atque Grimoaldo inierunt certamen.
.. ." Also L.H.F., ch. 51.
99
MEROVINGIAN MILITAKY ORGANIZATION
forced him to retreat. Shortly afterward, Ragamfred plundered

the area around Cologne and obtained a large amount of trea-
sure. While the mayor's force was returning home, Charles at-
tacked it at Ambleve near Malmedy and inflicted heavy losses.
Although this seems to have been in the nature of a hit-and-run
raid, its success probably raised the morale of Charles's follow-
ers.19
By March of 717, Charles had gained sufficient strength to
attack Ragamfred who, having been defeated near Cambrai, fled
south with his puppet king, Chilperic. Charles pursued them to
Paris, but then abruptly reversed direction and sped to Cologne
where he obtained his father's treasure and established Chlotar
IV as king in Neustria.20 There can be little doubt that by ob-
taining control of his father's treasure Charles's situation was
benefited considerably. In the next year Charles defeated Rad-
bod, who had routed him three years earlier. Apparently it had
taken Charles all of three years, a victory over Ragamfred, and
the acquisition of his father's treasure to secure the support
necessary to defeat Radbod.21
Viewing Charles's rapid rise with alarm, Ragamfred sought
the aid of Duke Eudo of Aquitaine. After the death of Dago-
bert I, Eudo's family had successfully established itself as
the ruler of Aquitaine in much the same manner that other
magnates had assumed control in Neustria, Austrasia, Thuringia,
and Burgundy. Ragamfred obtained Eudo's aid with gifts and
extravagant promises. Eudo led a levy of Gascons north against
19
L.H.F., ch. 52: "Eo nempe tempore denuo exercitum movent, usque
ipsum flavium Mosam contra Carlum dirigunt; ex alia parte Frigiones cum
Radbode duce consurgunt. Carlus quoque super ipsos Frigiones inruit, ibi-
que maximum dispendium de sodalibus suis perpessus est, atque per fugam
delapsus, abscessit." Fred, con't., ch. 9: "Contra quern praedictus vir Carlus
cum exercitu suo consurgens certamen invicem inierunt. . . ." Note the use
of sodales and exercitus as synonyms.
20
L.H.F., ch. 53: "Eo itidem tempore predictus vir Carlus, exercitu com-
moto, iterum contra Chilpericum vel Ragamfredo consurgens." Fred, con't.,
ch. 10; Paul, Hist., VI, 42: "Nam cum in custodia teneretur, divino nutu
ereptus aufugit, ac primum contra Raginfridum cum paucis bis terque cer-
tamen iniit novissimeque eum aput Vinciacum magno certamme superavit."
a
V. Willihrodi, ch. 13: "Qui multas gentes sceptris adiecit Francorum,
inter quas etiam cum triumph! gloria Fresiam, devicto Rabbodo, paterno
superaddidit imperio." See Wallace-Hadrill, Fredegar, p. 88, n. 2,
100
Charles, who, in turn, chased the Aquitanians south of the Loire.

Charles then besieged Ragamfred at Angers, devastated the sur-
rounding area, and returned home with a great quantity of
booty.22
During the next few years Charles defeated the Saxons and the
Alamans, and subjugated the Bavarians. In 725 he led his fol-
lowers in an amazing series of actions: first he crossed the Rhine
and overwhelmed the Alamans, then he went south and sub-
dued the Bavarians, and before the year was over he crossed the
Loire and routed Duke Eudo once again.23
As Charles was consolidating his position with conquests and
acquisitions of booty, Eudo was having to deal not only with
him, but with frequent Muslim raids which devastated his lands.
Eudo's supporters garrisoned the fortified cities and fortresses of
Aquitaine, but in 721 the Arab leader al-Samh besieged and took
Narbonne and then laid siege to Toulouse. Eudo, however, with
a force of Aquitanian and Prankish followers — the latter per-
haps descendants of the magnates exiled to Aquitaine by Ebroin
several decades earlier — defeated al-Samh's army and drove it
from Gaul. The Arab leader was slain, but his successor, 'Anbasa
ibn Suhaim al-Kalbi, was more successful; he captured, or more
probably besieged, Carcassonne and Nimes. He raided as far
north as Autun and perhaps even as far as Luxeuil. While this
Arab force concentrated in Burgundy, another group, command-
ed by Abd al-Rahman, moved into Aquitaine and in the summer
of 732 went against Bordeaux. At the confluence of the Dor-
dogne and the Garonne, Eudo met Abd al-Rahman's army and
was soundly defeated. As the duke of Aquitaine fled north to
seek the aid of Charles, the Arabs also moved northward, sacking
the Church of St. Hilary at Poitiers and then taking the road for
Tours.24
23
L.H.F., ch. 53; Fred, con't., chs. 10, 11.
23
Fred, con't., chs. 11, 12, 13.
a
Chron. Moissiac, p. 290; Fred., con't, chs. 42, 43, 44. E. Levi-Proven-
cal, Histoire de I'espagne musulmane (Paris, 1950), I, 58; M. Baudot, "Lo-
calisation et datation de la premiere victoire reinportee par Charles Martel
centre les Musulmans," Memoires et documents pubUees par la Soci-
ete de I'Ecole des Chartes, XII (1955), 93-105, is very probably mistaken
in setting 733 as the date for this battle. Michel Rouche, "Les Aquitans
101
In October Charles met the Arab army somewhere between
Poitiers and Tours and decisively defeated it. Though few details
concerning the battle of Poitiers can be ascertained, it has re-
ceived a great deal of attention, probably because it has symbol-
ized for many the turning back of the Muslims and the saving
of Christendom.25 Only two texts — somewhat contradictory in
nature — provide information of apparent historical value about
the battle. According to the continuator of Isidore's chronicle,
". . . the people of the North seemed like a wall, enduring un-
moveable, and like a firm glacial mass they remained together,
cutting down the Arabs with their swords."28
Fredegar's continuator asserts that ". . . Charles boldly pre-
pared his troops and aggressively charged the enemy. With
Christ's help he overthrew their tents. He continued the battle
slaughtering and destroying, and when their king Abd al-Rahrflan
was killed, he destroyed them, crushing the army. He fought and
conquered; victorious over his enemies he triumphed."27
The Isidorian continuator establishes a literary conceit by re-
ferring to the "people of the North" (gentes septentrionales) and
by describing their formation metaphorically as a "firm glacial
mass" (zonis rigoris glacialiter). Despite these affectations the
author's main thrust seems to emphasize the stolidness of
Charles's followers; permaneo (to endure, hold out, stay to the
end) and maneo (to stay or remain) are the verbs used to charac-
ont-ils trahi avant la bataille de Poitiers?" Le Moy0n Age, 74 (1968), 25-
26, presents a very convincing argument based largely on the Arab sources
for the traditional date of 732. See also Bachrach, "Charles Martel," pp.
72-73, and Donald Bullough, "Europae Pater: Charlemagne and His
Achievement in the Light of Recent Scholarship," EHR, LXXV (1970),
84-89.
25
Jean-Henri Roy and Jean Devoisse, La Bataille de Poitiers (Paris,
1966), 334-335, provide an extensive list of works on the battle of Poitiers.
Îsidori continuatio Hispana, chs. 104-105, pp. 361-362: ". . . gentes
septentrionales in hicto oculi ut paries inmobiles permanentes sicut et zoraa
rigoris glacialiter manent adstricti, Arabes gladio enecant." This chronicler
also has been called Isidorus Pacensis and the Anonymous of Cordoba.
"Fred, con't, ch. 13: "Contra quos Carlus princeps audacter aciem in-
struit, super eosque belligerator inruit. Christo auxiliante tentoria eorum
subvertit, ad proelium stragem conterendam accurrit interfectoque rege
eorum Abdirama prostravit, exercitum proterens, dimicavit atque devicit;
sicque victor de hostibus triumphavit."
102
Rots Faineants AND MAYORS OF THE PALACE
terize the Northerners' posture in face of the enemy. By dwelling

on this theme of stolidness, especially in his use of verbs, and
by avoiding the use of sto (to stand) the Isidorian chronicler
suggests a comparison with those who fled instead of enduring
or persevering in the face of the enemy. Throughout the Mero-
vingian era both the Visigoths and the Gascons had a well-
established reputation for turning their back and fleeing in face
of the enemy rather than remaining bravely on the field of battle.
It seems reasonable that behind his metaphors the Isidorian
chronicler is commenting not upon the Northerners' tactics, but
upon their bravery and courage.28
The problems caused by the ambiguity of the Isidorian's meta-
phorical style are probably of less consequence than those result-
ing from uncritical acceptance of his remarks. In short, Isidore's
continuator is not reliable. He presents legend as fact and makes
obvious errors indicating his lack of familiarity with affairs north
of the Loire, as well as with affairs in Aquitaine. For example,
he calls Eudo "duke of the Franks" (dux Francorum nomine Eu-
do), when in truth he held no such title. The chronicler also re-
fers to Charles's followers as Europenses. The use of this term
supports those who contend that the work was not a late eighth-
century composition but a creation of the later Middle Ages. If
indeed the entire chronicle is not later medieval then it was
surely interpolated during the later Middle Ages.29
28
For the Visigoths, Gregory, Hift., II, 37. For the Gascons, Fred., IV,
78: "Wascones deinter moncium rupes aegressi, ad bellum properant. Cum-
que priliare cepissint, ut eorum mus est terga vertentes, . . ." See also
Fred, con't, chs, 44; 47: ". . . set statim solito more omnes Wascones
terga vertunt. . . ." See Bachrach, "Feigned Retreat," pp. 265-266.
^Levison, Deutschlands Geschichtsquellen, I, 91, indicates the unreli-
ability of this text as does F. Codera, "Manusa y el duque Eudon," Estu-
dios criticos de historia drabe-espanola (Saragossa, 1903), 140-169, in vol.
VII of the Coleccion de Zos estudios drakes, Levi-Provencal, Hist, de Tesp.
mus., I, 60-61, is less critical of the Isidorian continuator than is Codera. The
entire text, however, is in need of detailed restudy. Concerning "Europenses,"
D. Hay, Europe: The Emergence of an Idea (Edinburgh, 1957), p. 25,
accepts the eighth-century date, post-754, for the use of the term. But
throughout his work, Hay adduces no other examples of the term being
used in either the early or the high Middle Ages. The term does not take
hold until the late Middle Ages, if indeed that early. For a long time
scholars considered the text of the Isidorian continuator to be a sixteenth-
103
In contrast to the continuator of Isidore's chronicle, the eon-
tinuator of Fredegar's chronicle seems to be the most trustworthy
source for the career of Charles Martel. Count Childebrand,
Charles's half-brother, and probably a participant in the battle
of Poitiers, sponsored and oversaw the writing of the section con-
cerning this particular battle.80 It would seem unlikely that the
count would have permitted Fredegar's continuator to describe
the battle inaccurately — that is, if Charles's followers had stood
in a phalanx the count would not have allowed the chronicler to
write that they charged the enemy or vice versa. Thus it seems
reasonable to accept this version of the story.
Throughout the rest of his life, Charles campaigned vigorously.
The year after his victory at Poitiers he invaded Burgundy with
his magnates and their followers, and subjugated it to his authori-
ty. He then gave his leudes lands in the Lyonnais so that they
could maintain his interests in Burgundy. At about the same time
or perhaps a little earlier, Charles gained control of the Orleanais,
confiscated the holdings of his enemies in the area, and granted
some of this wealth to his satellites.31
In 734 Charles led a naval expedition against the Frisians who
had rebelled against him. His fleet penetrated between the
islands of Westergo and Ostergo, he encamped on the banks of
the Boom, defeated the Frisians, killed their leader Bubo, and
returned to Francia with an abundance of spoils.82 While Charles
was in Frisia, Eudo died. Charles therefore gathered the mag-
nates of his kingdom together with their followers; they crossed
the Loire and traveled to Bordeaux and then to Blaye. He pro-
century forgery (Levison, p. 91); Roy and Devoisse, Poitiers, pp. 293-294,
seem to accept a late date: "Extrait de 1'anonyine de Cordoue (vers 1376-
1437)." The Chron. Moissiac, p. 290, calls Eudo "princeps Aquitaniae."
80
Fred, con't., eh. 34: "Usque mine inluster vir Childebrandus comes
avunculus praedicto rege Pippino hanc historian! vel gesta Francorum dili-
gentissime scribere procuravit," This was written about 752 and antedates
the earliest possible date for the Isidorian continuator by at least two years.
Some scholars argue for two battles or encounters in the Poitiers campaign.
It is of course possible but unlikely that Isidore's continuator described one
engagement and Fredegar's continuator another.
31
Fred, con't., ch. 14. Bachrach, "Charles Martel," pp. 66-68. Bullough,
"Europae Pater," pp. 84-89. Cf. White, Med. Tech., p. 12.
32
Fred, con't., ch. 17.
104
Rote Faineants AND MAYORS OF THE PALACE
ceeded to occupy Eudo's lands, including the cities and fortresses

of southwestern Gaul.33
During the next three years Charles subdued the Burgundian
magnates of the Lyonnais once again and placed the area as far
south as Marseilles under the leadership of his own counts. This
southern expansion brought Charles into contact with Maurontus
and his supporters (socii) who had taken the fortified city of
Avignon and devastated the surrounding area. Charles sent
Childebrand and several counts and dukes with their followers
against Maurontus. They laid siege to Avignon, and when
Charles arrived, the besiegers made their assault; battering rams
were brought into action and rope ladders were used to scale
the walls. The city was taken and its garrison was destroyed.
Charles then moved his men against Narbonne; siege was laid to
the city and engines for its destruction were brought into place.
Charles did not complete the siege of Narbonne because he was
distracted by the Arabs, led by lussef ibn Abd ar Rahman, who
were encamped not far from Narbonne. Charles turned his atten-
tion to them and his forces besieged their camp. When the Arab
governor Ukba ibn al-Hadjdjadj learned of Charles's activity in
the south, he advanced against him with a large army. On the
swampy banks of the Berre in the valley of Corbieres the Arabs
were defeated; Charles's followers, taking to boats, cut down
those survivors who tried to escape across the river. After obtain-
ing a rich booty from the conquered Arabs, Charles devastated
the area around Nimes, Agde, and Beziers. While Charles was
fighting the Arabs, the Saxons revolted; he therefore hurried
north and decisively crushed them. In 737 Charles also sent
Childebrand with many dukes and counts into Provence. The
Avignon area was once again subjugated and Maurontus, who
seems to have returned to the city at the time when Charles was
in the west, fled to an island fortress in the Mediterranean.84
In the span of a quarter century Charles Martel accomplished
a feat not unlike that of Clovis more than two centuries earlier.
"Ibid., ch. 15; cf. Fred., IV, 56.
" Fred, con't, chs. 18, 19, 20, 21; Chron. Moissiac, p. 292. See Levi-
Provengal, Hist, de I'esp. mus., I, 63.
105
Beginning with a small personal following, Charles increased

his power by winning military victories and acquiring wealth in
land and treasure which he used to buy the support of the mag-
nates and their armed retainers.
In approximately 740 Charles divided control of the lands un-
der his sway between his sons Carloman and Peppin. Carloman
was given Austrasia, Swabia, and Thuringia, and Peppin received
Burgundy, Neustria, and Provence. The same year Peppin and
his uncle Childebrand with a following of magnates and their
followers entered Burgundy and once again asserted Carolingian
control.35
When Charles died in October 741, a crisis arose comparable
to those which had accompanied the deaths of Merovingian
kings a century or two earlier. His daughter Childtrudis married
the Bavarian Duke Odilo, her brothers' enemy. Duke Chunoald
of Aquitaine, the late Eudo's son, revolted against Carloman
and Peppin, as did the Alamans, and shortly after this Odilo rose
in rebellion.36 Peppin and Carloman moved their forces (exerci-
tus) at once against Chunoald. After defeating a levy of Romani
which opposed them at Orleans, the brothers moved south to
Bourges, burning the surrounding area. They continued their
pursuit of Chunoald, devastating the land through which they
traveled. They successfully besieged the enemy fortress of
Loches, captured the garrison (custodes) there, and breached the
walls making them indefensible. Carloman and Peppin then di-
vided the booty they had taken and recrossed the Loire with
many prisoners. In the autumn of the same year, soon after re-
turning from Aquitaine, the brothers set out with their men
(exercitus) again — this time against the Alamans. They crossed
the Rhine and moved south, camping on the banks of the
Danube. When the Alamans were confronted by Peppin and
Carloman's army, they capitulated; recognizing their subject
status, they gave gifts and hostages to the Carolingian mayors.37
85
Fred, con't., chs. 23, 24: ". . . Pippinus dux, commoto exercito, cum
. . . Childebrando duce et multitudine primatum et agminum satellitum
plurimorum Burgundia dirigunt. . . ."
36
Ibid., chs. 25,26.
97
Ibid., ch. 25.
106
Rois Faineants AND MAYOHS OF THE PALACE
In the winter of 743 the brothers found it necessary to move

against Odilo. To supplement their Prankish followers, the
brothers called up a general levy to deal with the Bavarians. For
a fortnight the forces of the mayors and the Bavarian duke faced
each other on the banks of the Lech at Apfeldorf near Epfach.
Finally one night the Carolingian army made its way through
the marshy terrain surrounding Odilo's position and advanced
from a direction where they were least expected since there was
no causeway. The attack succeeded and Odilo barely escaped
with a small group of his men. Both sides, however, suffered
severe losses.88
Fredegar's continuator indicates that the brothers were com-
pelled to call up a general levy for the Bavarian campaign. This
was the first time in more than a century that such a force had
been called up in Austrasia, and probably in Neustria as well. To
enhance their position, and perhaps to legitimize their summon-
ing such a force, the Carolingian rulers filled the vacant king-
ship with the Merovingian Childeric III. The brothers' previous
successes in Aquitaine and against the Alamans may also have
strengthened their position. Nevertheless, the Bavarian revolt
seems to have been critical and the pious Carloman who had
been working for church reform found it essential to retain
secularized church lands in 743. He justified his action on the
grounds that he required the revenue to support his military cam-
paigns in those days when warfare was raging or threatening on
all sides.39
The year following the defeat of Odilo, Carloman's forces
(exercitus) invaded the Saxon borderland and the inhabitants
submitted without a fight. At the same time the Alaman Duke
Theudebald revolted. Peppin, commanding a force of his picked
men ("cum virtute exercitus sui"), journeyed into the Swabian
38
Ibid., ch. 26: "Conpulsi sunt generalem cum Francis in Bagoaria ad-
moveri exercito venientesque super fluvium qui dicitur Lech. . . ."
80
Ibid., MGH Cap., I, no. 11, 2: "propter inmentia bella et pcrsecu-
tiones ceterarum gentium quae in circuitu nostro sunt . . . in adiutorum
exercitus nostri." The Pope accepted Carloman's actions and recognized
that the mayor's military activities also benefited the church. (MGH epist.,
Hi, no. 324.)
107
Alps, drove Theudebald from his rocky retreat, and once more
subjected the duchy. The next year the brothers invaded Gas-
cony; the inhabitants, after first provoking the mayors, suc-
cumbed without a battle. The Alamans, however, again rebelled
and Carloman's army (exercitus) crushed them and he executed
many.40
In 747 Carloman retired to a monastery and Peppin ruled
alone. The Saxons rebelled against their new ruler, breaking the
oath they had sworn earlier to Carloman. Peppin's followers (ex-
ercitus), with the help of the Wends and the Frisians, invaded
Saxon territory; the Saxons submitted without a fight and prom-
ised to pay an annual tribute of 500 cows. The Bavarians revolted
in 749 and Peppin moved against them with a large force. The
rebels fled across the Inn and Peppin established his camp on
the banks of the river, preparing to attack the enemy from the
river craft which he ordered to be made ready for that purpose.
Seeing the imminence of the peril presented by Peppin's naval
preparations, the Bavarians decided to yield without a fight.41
Throughout the next two years Peppin ruled in peace and his
Merovingian puppet, Childeric III, reigned from his oxcart. First
with Carloman, and later alone, Peppin had subjugated the
former Merovingian kingdoms (except Aquitaine) to his control.
In 751 this Carolingian mayor, the true ruler of Gaul, became
Rex Francorum, replacing the Merovingian who merely reigned.
Thus the Merovingian dynasty came to an end and with it the
epoch of the mayors and the rots faineants.*2
The military in this era (638-751), as described in the numer-
ous campaigns treated above, cannot be characterized with a
simple generalization. In 638 Dagobert had left his sons a very
complex military organization. The Roman heritage was pre-
served in the milites, fortress garrisons, and centenae, and to
some degree in naval organization on the major rivers, and in
defensive organization on the old Roman road system. The
* Fred, con't., chs. 27, 28, 29.
tt
lW.,chs. 30, 31,32.
"Ibid., ch. 33: ". . . commoto exercito cum magno agmine appara-
tu. . . ."
108
other important groups functioning during the era of the last

ruling Merovingians (593-638) were local levies, regional levies,
general levies, standing forces (scara), and the personal follow-
ings of the magnates and the kings.
For a short while after Dagobert's death no great changes
seem to have taken place in the military of Merovingian Gaul,
with the exception that authority passed from the monarch to
the mayors of the palace and other important magnates. In 639
a regional levy of Austrasians from beyond the Rhine, the local
levies of the Saintois and Mainz (probably first organized by
Dagobert I or his father), and the followings of the Austrasian
magnates were called to service. These forces were used against
the rebellious Duke Radulf of Thuringia who summoned his re-
gional levy and his supporters among the magnates with their
followers. In Burgundy a few years later, the patrician Willebad
called up a general levy of his patriciate, the remains of the Bur-
gundian standing army which was traditionally led by the patri-
cian, and the magnates who supported him, both lay and ecclesi-
astical, with their followers. This force was opposed by the
mayors of Neustria and Burgundy with their personal retainers
and those of the magnates who supported them.
The Willebad affair is only one illustration of the importance
of the personal retainers during this period. For it can be said
that from the 640s on the main forces operating under the com-
mand of the mayors and other magnates were the personal armed
followings. Ebroin regained his place as mayor of the palace in
Neustria with the help of his followers (socii). He was later killed
by a Frankish magnate named Ermanfred and his socii. King
Dagobert II made an effort to reestablish royal power with the
help of a personal armed following of socii supplied to him by
Bishop Wilfrith. The antrustiones who once served the kings in
the centenae and at the court (cum armis suis) were probably still
recruited, but now used their arms to fight in the service of the
mayors of the palace who controlled the fisc and provided for
their support. On occasion some magnates would become power-
ful enough to organize a group of warbands. Thus Peppin II and
Duke Martin gathered a large force of Austrasians composed of
109
sodales and socii to attack Ebroin. Peppin also used loyal mag-
nates with their followers, or perhaps his own personal retainers,
to garrison strongholds under his control.
The forces led by Charles Martel and Peppin III are described
with some consistency by Fredegar's continuator. On four occa-
sions Charles's army is simply called an exercitus.43 One of these
forces which the continuator calls an exercitus is described by the
author of the Liber Historiae Francorum, upon whom the former
relies, as being formed of Charles's sodales.** At one point Frede-
gar's continuator depicts Charles's army as an agmen multitude
and at another time as a hostis Francorum.45
Fredegar's continuator describes eight of the forces led by the
mayors Carloman and Peppin III before the latter became king.
All but one of these are described simply as exercitus.46 The re-
maining body is called an exercitus generalis cum Francis or a
general army including Franks.47 Thus eleven of the fourteen
forces led by the mayors during the last thirty-five years of the
Merovingian era concerning which there is a clear description are
denoted by the term exercitus. Both Fredegar and his continuator
consistently use the term exercitus to mean a personal armed
following during the period of the last two Carolingian mayors.
When Fredegar's continuator wants to indicate some other kind
of force (such as a general levy), he uses a term like exercitus
generalis. Some additional light on the nature of the term exer-
citus as utilized by Fredegar's continuator may be obtained from
a description of Peppin's forces in 740: "Peppinus . . . commoto
exercito, cum avunculo suo Childebrando duce et multitudine
primatem et agminum satellitum plurimorum. . . . "48
The personal armed followings of the magnates seem to have
dominated the military in Austrasia, and this was probably the
case in Neustria and perhaps also in Burgundy. In Aquitaine, the
most Romanized part of Gaul, a more complex military structure
43
Ibid., chs. 9, 10, 13, 18.
44
L.H.F., 52.
45
Fred, con't, chs. 12, 19.
48
Ibid., chs. 24, 25, 27, 29, 31, 32.
" Ibid., ch. 26.
48
Ibid., ch. 24.
110
survived: a general levy of Gascons (hostis Vascanorum] func-

tioned, as did local levies such as the Romani at Orleans. The for-
tified cities and many fortresses of Aquitaine were provided with
garrisons.49
Though there seems to have been some simplification in mili-
tary organization during the era of the mayors, tactics remained
quite similar to those of earlier times. Siege operations, so com-
mon throughout the Merovingian era, continued, especially in
the more Romanized parts of Gaul. In 737 when Charles Martel
invaded southern Gaul he laid siege to Avignon, attacked the
walls with battering rams, used rope ladders to scale the heights,
and took the fortress. Dagobert II attacked fortified positions
as did Ghislemar who took the fortress at Namur belonging to
Peppin II, who, in turn, besieged the fortified city of Duurstede.
Naval operations also remained a part of Merovingian military
capability. Charles led a sea invasion of the Frisian islands in 734,
and in 737 his forces used river craft to cut off a Muslim retreat
on the river Berre. Peppin prepared naval craft to go against the
Bavarians in 749. Mounted operations also do not seem to have
been neglected by the mayors during this last period of Merovin-
gian history. Ebroin's pursuit of Leudesius and Theuderic IV was
surely carried out on horseback. Charles Martel's hit-and-run
raids in 715, his campaign in 717, and his threefold campaign of
725 in which he fought the Alamans across the Rhine, subdued
the Bavarians in southern Germany, and crossed the Loire to raid
in Aquitaine were most likely carried out at least in part by
mounted forces. Actual evidence for cavalry charges is limited.
Charles's forces which fought on the swampy banks of the Berre
would have had little chance to fight on horseback; Peppin's
troops who in 743 picked their way through the marshy terrain
at Apfeldorf where there was no causeway could not possibly
have been mounted; and it is unlikely that Peppin's followers who
drove the Alaman Duke Theudebald from the Swabian Alps kept
to their horses. It is more difficult to find clear-cut instances of
mounted activity from 639 to 751 than it was in earlier periods of
49
Ibid., chs. 15, 20, 25, 46.
Ill
Merovingian history. Chaubedo's attack in 643 seems to have
been more a personal effort than a mass charge, and not com-
parable to Theuderic I's charge across the Thuringian plains in
531 or Landri's at Soissons in 593 or Dagobert's against the Sax-
ons in 623. Under the mayors of the palace there seem to have
been no mounted shock attacks after the battle of Poitiers;
though at the battle itself Charles Martel's followers who charged
the Muslims, overran their tents, and pursued them hotly may
well have done so on horseback and not on foot.50
Thus, although the Merovingian kings gradually lost control
of Gaul to the mayors and the magnates, their military organiza-
tion, which had become more diverse through the centuries, but
had retained much of the Roman past, continued to play a sig-
nificant and influential role in this last century of Merovingian
history.
M
112
CHAPTER VI
Conclusion
LTHOUGH previous studies dealing with Merovingian military
Aorganization have been limited to a small selection of the
available evidence, the methods and conclusions of the more
influential scholarly works are worth discussing if only to indi-
cate how they might have benefited from a use of the entire
corpus of evidence.
In his Institutions militaires de la France which appeared in
1863, Edgard Boutaric devoted some nineteen pages to the mili-
tary of "la premiere race."1 He argued that the Franks as an
ethnic group in the Merovingian kingdoms were only a minority
of the population and had to rely heavily upon their Gallo-
Roman subjects for both the conquest and the defense of Gaul.2
Nevertheless, when discussing the armament and organization of
the Merovingian armies, Boutaric seems to ignore his earlier re-
marks and the evidence on which they are based and allows the
Byzantine historian Agathias's characterization of the Franks as
an ethnic group to serve as a description of the rank and file of
the Merovingian armies.3 Boutaric does argue, however, that the
mounted bodyguards of the kings and the armed followers of
1
Paris, 1863, pp. 50-69.
2
Boutaric, Inst. milit., pp. 50-54. Fustel de Coulanges, Hist, des inst.,
IV, 293ff, generally agrees with this conclusion.
3
Boutaric, Inst. milit., pp. 64-65.
113
the magnates formed a class of professional soldiers and played
an important role in the Merovingian military.4 He further main-
tains that imperial influence was felt especially in the organiza-
tion of siege warfare.5 Perhaps the most crucial weaknesses of
Boutaric's work are the inadequate length of his study, which
does not allow full exploitation of the sources, and his neglect
of the military during the last century of Merovingian history.
Boutaric's balanced though limited account of the Merovingian
military has won few adherents among the most influential his-
torians in the field. From the many articles, essays, and chapters
of general works which have appeared since 1863, an image of
the Merovingian military emerges of poorly armed and ill-disci-
plined levies of infantrymen performing military service in re-
sponse to a primal Teutonic duty owed by all freemen. Roman
military institutions are generally regarded as having ceased to
exist, and the numerically preponderant and tactically decisive
elements of the Merovingian military from 481 until at least 732
are the farmer-soldiers of the pagi fighting under their counts.
With the general acceptance of this view, scholars have devoted
their efforts to ascertaining when and under what circumstances
these levies of Prankish footmen, supported on occasion by Gallo-
Romans, were replaced by heavily armed knights who, serving as
feudal vassals, are alleged to have dominated the battlefields of
Europe during the Middle Ages.6
Heinrich Brunner's 1887 article, "Der Reiterdienst und die An-
fange des Lehnwesens," embodies what is today the dominant
picture of the Merovingian military.7 Relying heavily upon the
writings of Procopius and Agathias, Brunner unequivocally
*Ibid., pp. 64-67. Fustel de Coulanges, Hist, des inst., IV, 289, argues
the opposite: "L'armee composee de vrais soldats c'est desorganisee et dis-
paru."
5
Boutaric, Inst. milit., p. 66.
8
Claudio Sanchez-Albornoz, En torno a las Ortgenes del Feudalism*)
(Mendoza, 1942), III, 1-27, and Lynn T. White, Jr., Medieval Technology
and Social Change (Oxford, 1962), pp. 5, 6, 7, 137, and 138.
7
ZRG, VIII, 1-38, reprinted in Brunner, Forschungen zur Geschichte des
deutschen und franzdsischen Rechts (Stuttgart, 1894), pp. 39-74. All cita-
tions here are to the former. See also Brunner, Deutsche Rechtsgeschichte,
2nd ed. (Munich, 1928), II, 269ff. Cf. Georg Waitz, Deutsche Verfassungs-
geschichte (Berlin, 1882), II, passim.
114
CONCLUSION
equates their descriptions of the tribal military customs of the

Franks as an ethnic group with the military customs of the Mero-
vingian armies.8 Brunner drew his account of a horde of half-naked
infantrymen armed with axes and barbed spears who performed
military service because they were freemen from these two By-
zantine writers, neither of whom had ever seen a Merovingian
army, much less a purely Prankish fighting force. Brunner cites
as supporting evidence for his view an Anglo-Saxon poem which
has been dated variously from the eighth to the tenth century.
According to Brunner, this poem provides accurate evidence for
an event that took place in Gaul about 515. In developing his
argument Brunner neglects Gregory of Tours's account of this
episode in 515 which does not support the thrust of the consider-
ably later English text. To round out his study of Merovingian
military tactics, Brunner cites a passage from Gregory of Tours
describing Clovis splitting open the head of a warrior with an
ax. From these four accounts — two Byzantine, one Anglo-Saxon,
and one Merovingian — none of which deals with the period
after 560, Brunner concludes that Merovingian military tactics
and armament from 481 to 732 were synonymous with the tribal
practices of the Franks in all significant aspects. The 732 date
for the terminus of this form of organization and tactics is pro-
vided for Brunner by the description of the battle of Poitiers
given by the continuator of Isidore's chronicle in which Charles
Martel's armies are alleged to have fought on foot. To help sup-
port his argument that a great change was wrought in the mili-
tary after this battle, Brunner cites a provision of the Ripuarian
Prankish law which lists horses, arms, and armor in a manner
which might lead to the conclusion that the national weapons of
the Franks had been changed not only in fact but also in law
during the eighth century.9 Brunner apparently was unaware of
8
Brunner, "Reiterdienst," pp. 2-3.
9
Ibid., pp. 3-4, and 13-16. Lex Rib., 40, 11: "Equum . . . sanum pro
septem solid, tribuat. Spatam cum scoigilo pro septem solid, tribuat. . . .
Brunia bona pro duodecim solid, tribuat. Helmo condericto pro sex solid,
tribuat. scuto cum lancia pro duos solid, tribuat." It is doubtful that Brunner
is correct in assuming that this list of values for the above can be viewed
as a list of the national arms of the Ripuarian Franks.
115
an early seventh-century compilation of this law code providing

the same list upon which the revolutionary thesis rests.10 Brun-
ner's failure to use all available historical evidence and his mis-
use of the Ripuarian code undermine the validity of his argument.
The weaknesses of Brunner's theory, at least in part, were
grasped by some of the scholars who have contended that there
was a gradual development of mounted warfare under the Mero-
vingian kings. Perhaps the most articulate spokesman for this
position is Charles Oman whose History of the Art of War in the
Middle Ages appeared in 1898.n Although Oman accepts the
concept, formulated by Brunner and based upon Byzantine
sources, of the Merovingian army as a levy of half-naked Franks,
he holds this view to be valid only through the sixth and part
of the seventh century at which time he sees the gradual devel-
opment of a cavalry force. In defending Brunner's account of the
Merovingian military for the era of Clovis and his sons, Oman
argues that the Franks had changed little from the days of Taci-
tus and "bore a great resemblance to their Sigambrian or Chama-
vian ancestors. . . ,"12 Apparently Oman believed the Sicam-
brians had changed their customs radically during the period
between Caesar's description of them as a mounted people and
the early second century A.D., when Tacitus was silent about their
tactics.13
Though Oman admits the Romans or Gallo-Romans served in
the armies of the Merovingian kings, he maintains "it is quite
clear that the conquerors did not adopt the arms of the con-
quered, and that the survival of Roman garb and weapons among
the Gauls disappeared in the sixth century." He remarks further
that "they [the Franks] seemed to have borrowed nothing from
their Roman predecessors."14 Oman does not indicate the nature
10
R. Buchner, "Die Rechtsquellen," Wattenbach-Levison Deutschlands
Geschichtsquellen im Mittelalter (Weimar, 1953), pp. 23ff.
11
London, 1898. A second and expanded edition in two volumes appeared
in 1923. All citations here are to the 1923 edition. Oman published an
earlier essay entitled The Art of War in the Middle Ages (Oxford, 1885).
™ Art of War, I, 51-52.
13
Caesar, E.G., VI, 35, quoted in ch. I, n. 36.
™ Art of War, I, 53; cf. 54.
116
CONCLUSION
of the evidence which would clarify this conclusion and ignores

the evidence to the contrary.
Oman's primary purpose is not to demonstrate the Prankish
nature of the Merovingian armies or the lack of imperial influ-
ence, but rather to show how a group of primitive Franks be-
came the heavily armed knights of European chivalry. To demon-
strate the gradual nature of this change, Oman begins by citing
an instance of Franks fighting on horseback in 539 and provides
a host of such examples up through the first third of the seventh
century.10 For some unknown reason he neglects Gregory's refer-
ence to a Frankish cavalry charge in 531, as well as accounts from
the last century of Merovingian history, except for the battle of
Poitiers, at which, he grudgingly admits, Charles Martel's troops
fought on foot. He does contend that such a tactic was very un-
usual by then.18
The most comprehensive effort to illustrate the gradual devel-
opment and expansion of the use of heavily armed horsemen in
Gaul from approximately 500 to 900 is Hans von Mangoldt-
Gaudlitz's Die Reiterei in den germanischen tmd frankischen
Heeren bis zum Ausgang der deutschen Karolinger which ap-
peared in 1922. According to von Mangoldt-Gaudlitz, the fighting
forces of early medieval Europe were Germanic and they evolved
from the various tribes which conquered the Roman empire. He
devotes his first chapter to the German military from its initial
appearance in history to the end of Roman rule in Gaul, arguing
that some groups customarily fought on horseback whereas others
were generally infantry-oriented.17 In order for the Franks, or
presumably for any group of infantry people, to develop an out-
15
10
Ifcfe£,I,54ff.
Ibid., I, 58. Gustav Roloff, "Die Umwandlung des frankischen Heers
von Chlodoweg bis Karl den Grossen," Neue Jahrhucher fur das klassische Al-
tertum, IX (1902), 389-399, is also of gradualist persuasion. On pp. 390-
391, n. 1, Roloff contends that Brunner's interpretation of the evidence for
the battle of Poitiers is wrong. He also argues against Brunner's interpreta-
tion of the accounts given by Procopius and Agathias. In doing so, Roloff
attacks by implication Oman's acceptance of the Byzantine authors, as well
as his treatment of Isidore's continuator on Poitiers.
" Berlin, 1922, pp. 3-12. "Die zeit von dem ersten Auftreten der Germa-
nen in der Geschichte bis zu dem Untergang der Rbmerherrschaft in Gal-
lien."
117
standing cavalry two conditions, asserts von Mangoldt-Gaudlitz,
were necessary: the economic means to support large numbers of
horsemen, and an enemy which could be combated effectively
only on horseback. Basing his interpretation on these two pre-
requisites, he concludes that as the Franks came to dominate
more territory they obtained the means to support a greater num-
ber of horsemen; at the same time the Franks found it ncessary
to fight against the Visigoths, Avars, Lombards, and Muslims, all
of whom, he maintains, fought on horseback and against whom
the Franks could be decisive only as cavalrymen. Having thus
defined the problem and the framework for its solution, von
Mangoldt-Gaudlitz tries to demonstrate the gradual development
of horsemen among the Franks by listing instances in which
fighting forces of the Merovingian kings and magnates are noted
as appearing in battle or some paramilitary operation on horse-
back.18 He then makes a similar catalog for the Carolingian
period. This statistical exercise is completed by counting the
instances of mounted Merovingian and Carolingian units men-
tioned in the sources selected. He concludes that since the refer-
ence to horsemen gradually increased throughout the period
(500-900) in rough proportion to both Prankish expansion and
Prankish wars against mounted enemies, then the number of
horsemen increased also.19
The diverse nature and varying quantity and quality of the
source materials of the early Middle Ages invalidate von Man-
goldt-Gaudlitz's statistical method which is based upon the as-
sumption that the surviving notices of equestrian activity in the
limited number of selected sources are a representatively bal-
anced sample of the whole. These criticisms of von Mangoldt-
Gaudlitz's statistical method may not, however, convince the less
mathematically oriented reader of the work's weaknesses. For
those who would defend von Mangoldt-Gaudlitz's work it should
be noted that he devotes only thirteen pages to the written
sources of the Merovingian period, disregards much of the evi-
18
Von Mangoldt-Gaudlitz, Reiterei, pp. 13-20, and 49.
™lbid., pp. 21-25, and 25-49. The archaeological evidence is used in a
similar manner, pp. 85, 67-73.
118
CONCLUSION
deuce, ignores the last century of the era, and seems to think
that all elements in the Merovingian armies were German. In
fairness to the author it should be pointed out that his work is not
an attempt to present a balanced account of Merovingian mili-
tary organization, but simply an effort to destroy Brunner's
thesis of military revolution. In doing so von Mangoldt-Gaudlitz
agrees with Brunner that there were only Germanic fighting men,
but to contradict Brunner's contention that the Merovingian
Franks were preponderantly footmen, he introduces fifteen texts
indicating that Merovingian forces served on horseback.
Eugen von Frauenholz, Hans Delbriick, and Ferdinand Lot
have tried to demonstrate the gradual development of cavalry
in Gaul, and deserve some mention if only because their works
are frequently cited as authoritative. Von Frauenholz states, with-
out evidence, that the Franks had an important body of horse-
men which continued to develop throughout the Merovingian
era. In support of this position he argues that horses were avail-
able to the Franks and since mounted combat was the most
effective way to fight, it would have been absurd for the Farnks
not to have done so.20 This technical determinism is invalid in
terms of method, and surely not a substitute for evidence.21
In his remarks on the Merovingian military Delbriick does little
more than von Frauenholz to inspire the confidence of his read-
ers. He makes relatively few references to the Merovingian armed
forces, reproduces a few passages from the sources which he occa-
sionally introduces with a cryptic sentence or two, but on the
whole makes no effort to indicate their relevance. Nor does he
endeavor to evaluate the sources critically or to integrate the
material he quotes into his narrative. He concludes, nevertheless,
that the gradual development of horsemen among the Franks was
complete by Charlemagne's reign, at which time vassal armies
composed of horsemen were used exclusively.22
20
Das Heerwesen der germanischen Friizett (Munich, 1935), p. 61.
=1
P. Sawyer and R. Hilton, "Technical Determinism: The Stirrup and the
Plough," Past and Present, XXIV (1963), 90ff, provide some trenchant com-
ment on this method.
22
Geschichte der Kriegskunst in Rahmen der poUtischen Geschichte, 2nd
ed. (Berlin, 1921), II, 427-431, 453-459, 475-476. The first edition ap-
peared in 1907.
119
Ferdinand Lot's study is still the most scholarly attempt to
show the gradual development of the use of horsemen in Gaul.
Lot's primary concern is numbers and in this area he is reason-
able and draws some perceptive conclusions.28 Lot admits that
the Merovingian armies were composed of many peoples, but
allows his section on the "organisation, tactique, et armament des
Francs" as an ethnic group described in the Byzantine sources
to stand for the whole of the Merovingian military. Lot assumes
gradual development to be the nature of historical change during
the Middle Ages and rejects Brunner's thesis out of hand, despite
his own failure to deal with the sources on the last century of
the Merovingian military.24
In reaction to Delbriick's extreme claims that mounted vassals
exclusively formed the Carolingian army, W. Erban and H. Fehr
not only argue that the Merovingian army was composed essen-
tially of free Prankish warriors fighting on foot in the period of
Clovis and his sons (as most scholars who argued for gradual
change believed), but follow Brunner in contending that this was
the case in the era of Charles Martel as well. Yet where Brunner
saw a radical change from an army with a preponderance of
infantry to one with a preponderance of cavalry, Erban, Fehr,
and Karl Rubel maintained that the Carolingian armies were
neither in law nor in fact numerically superior in horsemen, but
predominantly infantry levies of free Franks. By concentrating on
legal sources and interpreting them with an image of Germanic
society rooted in the Mark theory, these authors create a model
of the military which owes more to nineteenth-century romanti-
cism than to early medieval evidence.25
a
L'art m&ltmre et les armies au mot/en age (Paris, 1946), I, 74-79.
"•Ibid., pp. 80-90. See also F. Lot, C. Pfister, and F. L. Ganshof, Les
destinees de Fempire en Occident de 395 A 888, 2nd ed. (Paris, 1940), pp.
665ff, and especially pp. 687-688. White, Med. Tech., p. 137, attacks Lot's
a priori assumptions about the nature of historical change, only to substitute
another
25
set of a priori assumptions.
W. Erban, "Zur Geschichte des karolingischen Kriegswesens," HZ, CI
(1908), 321-336; Fehr, "Das Waffenrecht der Bauern im Mittelalter," ZHG,
XXXV (1914), llSff; Rubel, "Frankisches und spatromisches Kriegswesen,"
Banner Jahrbiicher, CXIV (1906), 136-142. Brunner, "Reiterdienst," pp. 3-
4, and 7, "Der Zeitpunct, in welchem die Reiterei bei den Franken das
Ubergewicht iiber die Fusstruppen erlangte. . . ." He argues that by Char-
120
CONCLUSION
Alfons Dopsch, whose work is an unremitting effort to show

continuity between the late empire and the Middle Ages, recog-
nized in Brunner's arguments a dangerous threat to his own his-
torical position. As Brunner's most tenacious critic, Dopsch de-
voted about six pages of his two-volume Wirtschaftliche und
Soziale Grundlagen der Europaischen Kulturentwicklung aus der
Zeit von Caesar bis auf Karl den Grossen to the Merovingian
military, arguing that the Germanic tribes, and especially the
Franks, had mounted troops in substantial numbers and of fine
quality. To prove this, Dopsch cites a number of texts ignored by
Brunner, and even some ignored by the gradualists, which indi-
cate the existence of forces of the Merovingian kings serving on
horseback.26 Dopsch also uses the researches of Erban, Fehr, and
Rubel to contend that levies of freemen fighting on foot con-
tinued to be of importance in the Carolingian era. Thus, in his
opinion, the military in the Merovingian era was predominantly
infantry, though horsemen were of considerable significance, and
he sees the situation largely unchanged during the Carolingian
period.27
In response to more than a half-century of criticism directed
against Brunner, C. von Schwerin and H. Voltini came to his
support, but in doing so fundamentally altered his original
lemagne's reign a substantial part of the army was composed of horsemen.
In describing this process he writes: ". . . dass er ins achten Jahrhundert
hinaufreiche, weil bereits die Heere Karls des Grossen zum guten Theile
aus Reiterei bestanden hatten" (p. 7).
The Erban-Fehr interpretation is criticized in an important article by
Heinrich Dannenbauer, "Die Freien im karolingische Heer," Aus Verfas-
sungs- und Landesgeschichte: Festschrift filr T. Mayer (Lindau, 1954), I,
49-64
28
Vienna, 1920, II, 292-297.
27
Dopsch, Grundlagen der europaischen Kulturentwicklung, pp. 297-299.
H. A. Cronne, "The Origins of Feudalism," History, XXIV (1939), 257,
finds Dopsch's arguments "not a little bewildering, and to some extent
mutually destructive." It seems rather that Cronne misunderstands Dopsch.
White, Med. Tech., p. 5, is also bewildered, and believes that when Cronne
made the remark above, he was referring to "military historians who deny
that the second quarter of the eighth century witnessed any decisive
change in methods of fighting." Dopsch, on the contrary, seems to be argu-
ing that these positions are not mutually destructive and can be brought
into harmony.
121
thesis.28 Brunner's most recent defender, Lynn White, Jr., suc-

cinctly summarizes the new version of the Brunner thesis on the
Merovingian military. White and Brunner's other adherents ad-
mit that in regard to numbers, levies of free Frankish footmen
were preponderant in both the Merovingian and Carolingian
period. Here the arguments of Erban, Fehr, and Rubel seem to
prevail and Brunner's defenders discard any notion of a revolu-
tion in the conversion of the numerically superior infantry force
in the Merovingian era into a numerically superior cavalry force
by Charles Martel or his successors. Because those who have tried
to justify Brunner's interpretation accept as true the idea that the
"Merovingians fought to some extent on horseback, . . ." they
seem to have succumbed to Dopsch's position — the infantry
remained predominant throughout both the Merovingian and
Carolingian eras, yet both periods witnessed the use of calvary
forces as well. Brunner's defenders would not accept such a
conclusion, however, for they contend that although infantry
levies of ill-armed and ill-disciplined Franks were the "tactically
decisive" arm in the Merovingian military, Charles Martel revolu-
tionized warfare by making the cavalry the decisive military
arm.29 This picture continues to prevail in the historiography at
present, but neither Brunner nor his adherents ever examined the
military campaigns of Charles Martel and Peppin III in detail.
Such an investigation demonstrates conclusively that cavalry was
not the tactically decisive element of the Merovingian military
under the last two mayors of the palace.30
A modern study deserving special notice is Jean-Pierre Bod-
mer's Der Krieger der Meroioingerzeit, which may be character-
ized in German as a work of "historische Anthropologie," and
may be translated into American jargon as "historical sociology."
Bodmer's aim is to describe the world of the Merovingian warrior
88
Von Schwerin in his edition of Brunner's Deutsche Rechtsgeschichte
(Munich, 1928), II, 277, n. 30, and 279, n. 33, and Voltini, "Prekarie und
Beneficium," Vierteljahrschrift fur Sozial und Wirtschaftsgeschichte, XVI
(1923), 293-305.
29
White, Med. Tech., pp. 3, 5, 6, 11,13, 137, 138.
80
Bullough, "Europae Pater" pp. 84-90, and Bachrach, "Charles Martel,"
pp. 49-75, for a critique of the Brunner thesis and its modern defenders.
122
CONCLUSION
in his role as a fighting man within the framework of social insti-
tutions. Though this effort is noteworthy for its attempt to relate
modern sociological techniques to early medieval life, Bodmer
bases his work on the assumptions set forth in the historical
studies discussed above. This may help to explain why Bodmer
limits his study to the period from 481 to 639 and omits more
than a century of the Merovingian era. Bodmer is in the unenvi-
able position of writing a work of historical sociology in which
his sociological investigations and insights, no matter how fruit-
ful, are severely handicapped by the inadequate historical studies
upon which he has been forced to rely.31
As has been seen, the most influential works on the Mero-
vingian military have been distortingly selective in their use of
evidence and chronologically limited in their choice of subject
matter. These previous historical interpretations have been de-
termined by two significant trends. The first concentrates upon
the Franks as an ethnic group and equates their military insti-
tutions with those of the entire Merovingian military establish-
ment. For the most part scholars have been satisfied with this
oversimplification of the Merovingian military organization in
which half-naked levies of poorly armed and ill-disciplined
Frankish infantrymen, depicted as the numerically preponderant
and tactically decisive military element, performed military ser-
vice because of some putative Teutonic duty.32 The second trend
81
Zurich, 1957, pp. 9-14. Bodmer discusses the value of previous works
and admits that the Merovingian period has been badly treated: "In den
meisten dieser Werker spielt die Merowingerzeit die Rolle einen Stiefkin-
des." Nevertheless, Bodmer relies heavily on at least three of these works,
and says of Delbriick's study: "Delbriicks Geschichte der Kriegskunst
enthalt neben der leidigen Polemik doch so viel Gutes, dass sich di Bentit-
zung dieses Werkes stets empfiehlt."
M
A recent historical conference produced two massive volumes on mili-
tary matters in the early Middle Ages; Ordinamenti Militarl in Occidente
nell 'Alto Medioevo (Settimane di Studio del Centro Italiano di Studi sutt
'Alto Medioevo, XV [Spoleto, 1968]). Of the many studies which appear in
these volumes only one, Joachim Werner, "Bewaffnung und Waffenbeigabe
in der Merowingerzeit," I, 95-108, deals with military organization in
Merovingian Gaul in even a limited manner. This essay is devoted essen-
tially to Frankish weapons. In their expansion on Schmidt's work, Zolner
(with J. Werner), Geschichte der Franken, pp. 150-167, provide no con-
ceptual innovation.
123
focuses upon the means and circumstances by which these primi-

tive warriors were developed into a heavily armed feudal cavalry
that is alleged to have dominated the battlefields of medieval
Europe and to have formed the basis of the nobility.33
A study of all the significant texts relating to Merovingian
military organization refutes these widely accepted interpreta-
tions, and from such an examination a far different picture
emerges. When Clovis died in 511, he left his sons a military
establishment which owed much to the empire. Fortified cities
and castra were garrisoned by erstwhile imperial military per-
sonnel (milites and laeti) and their descendants. Gallo-Roman
magnates, both lay and ecclesiastical, were the sworn followers
(leudes) of the Merovingian monarchs; they were a formidable
force in local affairs and their loyalty was essential to a success-
ful reign. Throughout the lands of the royal fisc — much of which
had been the imperial fisc and upon which Roman military
colonists had been settled — Clovis organized centenae or garri-
son settlements of Franks and others who, as sworn members of
the king's trustis, served in military and paramilitary actions. The
king also kept a large group of armed men at his side, usually
called pueri in the sources, who were also probably members of
his trustis.
Clevis's sons changed little of what they inherited. The mag-
nates, both Gallo-Romans and Franks and their supporters,
played a large role in military operations during their reigns, as
did garrisons in fortified positions. The heterogeneous ethnic
composition of the Merovingian military, including Alamans,
Alans, Taifals, and Saxons, as well as Gallo-Romans and Franks,
was expanded by Clovis's sons with the inclusion of the Burgun-
dians and their military establishment. This acquisition brought
the Merovingian kings not only fortresses, fortified cities, garri-
sons, magnates and their followers, but a standing army, probably
of imperial origin, which served under the patrician of Burgundy
who was usually of Gallo-Roman origin.
Under Clovis's grandsons a great innovation was introduced
into Merovingian military organization — the local levy. Through
88
White, Med. Tech., pp. 1-38.
124
CONCLUSION
these local levies, which were established only in the cities of

Neustria and Aquitaine during this period, Clovis's grandsons
were able to mobilize much larger numbers of fighting men on a
regular basis than had been previously possible. The wars of the
latter part of the sixth century also saw the occasional use of
general levies which included pauperes and inferiores.
The last of the ruling Merovingians built upon the work of
their predecessors. Local levies were developed in less Roman-
ized parts of Gaul like Mainz and the Saintois. A territorial levy,
similar to the territorial levy of Champagne which had been de-
veloped during the era of Clovis's grandsons, was organized in
Austrasia during this period. Gregory of Tours, whose historical
account provides the bulk of the evidence for the reigns of Clovis,
his sons, and his grandsons, ceased writing in the last decade of
the sixth century at which time the writings of Fredegar and his
continuator become the basic source for the Merovingian military.
Despite some differences in terminology brought about by the
change in authors, the continuity in the Merovingian military
through the reign of Dagobert I remains clear.
The era of the rois faineants and the mayors of the palace
marks some noteworthy changes in Merovingian military organi-
zation. In Austrasia especially, the various types of levies fade out
of existence before the middle of the seventh century and military
force is left in the hands of the magnates and their armed follow-
ers. The military in Neustria and Burgundy experienced the same
kind of change, but in Aquitaine, the most Romanized part of
Gaul, more complex forms of military organization including
garrisons, levies, and the magnates with their followings survive.
Throughout the last century of Merovingian history, the mag-
nates and their armed supporters became the primary military
forces in most of Gaul. This is not to say that the number of
magnates or even the number of their armed followers increased
for no evidence concerning numbers is to be found. The magnates
and their retainers grew more important militarily during this
period because other segments of the military in Merovingian
Gaul ceased to exist.
125
Despite the increasing significance of the magnates' followings,

the mayors of the palace continued the same techniques used by
the Merovingian kings in acquiring the support of these groups.
Gifts in movable wealth and estates were provided by mayors like
Aega and Peppin III as they had been by kings like Clovis, Theu-
deric I, or Dagobert I. Charles Martel, who is alleged to have
introduced revolutionary changes, did no more and no less in his
efforts to acquire and maintain the support of the magnates and
their followers than had his predecessors.34
Although previous historical interpretations maintain that the
Merovingian forces were poorly armed with only axes and barbed
spears, it can be justly concluded that the heterogeneous ele-
ments of the Merovingian military used various kinds of wea-
pons. For example, the magnates in Merovingian Gaul provided
their followers with mail coats, helmets, shields, lances, swords,
bows and arrows, and horses. These retainers resembled in arma-
ment the followers of the Gallo-Roman potentiates who had
flourished under the empire. The remnants of the imperial mili-
tary who were absorbed into the Merovingian military continued
to use their old armor and weapons, and among Rome's erstwhile
allies integrated into the Merovingian military, the Alan cavalry
so dominated the military of Armorica that their influence was
recognizable there at least into the tenth century and probably
into the twelfth century.
The local levies of the Neustrian and Aquitanian cities seem
to have been well armed with, among other things, defensive
armor and mounts. Because the elements of lesser social and
economic status (pauperes and inferiors?) which were included
in the general levies seem to have been poorly equipped and
fought with anything which might come to hand, the general
levies were infrequently used and of little military value. The
armament of the trans-Rhenish peoples, who were at times called
upon for military aid, was little affected by conditions in the
West, and they were generally less well armed than members of
the same ethnic groups who had fallen under more direct imperial
84
126
CONCLUSION
influence. Among the rank and file of the Saxon, Prankish, or

Wendish hordes raised beyond the Rhine, body armor was gen-
erally lacking; the sword, spear, and ax served as offensive wea-
pons, and horsemen were generally few in number.
The tactics of the Merovingian armies varied from situation

to situation and were dictated, in part, by what may perhaps be
characterized as Merovingian strategy. The fundamental prin-
ciple of Merovingian strategy within Gaul was the acquisition
and defense of as many fortified cities and castra as possible.
Outside of Gaul — south of the Alps and Pyrenees — Merovingian
strategy generally seems to have been limited to raiding for
booty and acquiring tribute. These were also the military goals,
at least in part, of Merovingian operations east of the Rhine, al-
though the main intention here was to extend actual political
power beyond the Rhine through military activity.
A primary aim of the Merovingian kings and later of the
mayors was the control of wealth, which explains in part their
interest in maintaining possession of the cities throughout Gaul.
For they were the organizational centers of much wealth: lay
and ecclesiastical administrations were based in the cities; taxes
and tolls were collected there; and markets and merchants flour-
ished. The cities were walled and defensible and with their sur-
rounding area they formed a miniature state to which the local
inhabitants even had a particularistic loyalty. The Merovingian
kings placed garrisons in these cities; these forces differed or-
ganizationally from the local levies and from the general popu-
lace and served under commanders different from those who
led the local levies.
Warfare throughout the Merovingian era, especially in the
more Romanized parts of Gaul — Aquitaine, Neustria, and Bur-
gundy— focused on these walled cities. From Clevis's siege of
Paris in the 480s to Charles Martel's siege of Avignon in 737, the
investment of fortified positions, whether walled cities or garri-
soned castra, dominated Merovingian warfare. The Roman origin
of this city-oriented aspect of Merovingian strategy and the tac-
tics of siege warfare which necessarily evolved from it cannot
127
be too strongly emphasized. The use of siege engines of various

types and the techniques for cutting off provisions and communi-
cations, as well as the means of controlling the population in
areas surrounding besieged positions and of supplying the be-
sieging forces, though organized to varying degrees and applied
with widely differing results, mark the Roman nature of Mero-
vingian warfare both tactically and strategically.
Siege warfare was not the only tactic of the Merovingian mili-
tary which owed a debt to the empire. The organization of naval
forces, such as that which defeated the Danes in 515, and the
utilization of river craft for military and paramilitary maneuvers
of the Loire, Rhine, and Rhone suggest imperial rather than
Frankish influence. In many battles of the Merovingian era which
did not involve siege or water operations, tactics seem to have
been flexible and dictated more by the demands of a particular
situation than by tribal custom. If the tactics of the Merovingian
military may be characterized, perhaps the only suitable term
would be flexible. Siege warfare, mounted attacks, naval opera-
tions, and infantry movements were all employed by Merovingian
armed forces. Mounted troops, which seem to have been com-
paratively numerous (even important elements of the local levies
were probably mounted), were not committed to fighting on
horseback and if the situation were appropriate they dismounted
and fought on foot.
The Merovingian military was formed from many peoples and

groups, its tactics were flexible, and its strategy was variable.
Perhaps the most important conclusion that can be drawn from
the present study, especially in light of previous scholarship, is
that the Merovingian military was greatly influenced by the
Roman empire and its institutions, and it owed comparatively
little to the Franks, who were only a minority of the population
and a small part of the fighting forces. As with many aspects of
Merovingian life, the military organization recalls Romania and
not Germania.
128
Appendix
APPENDIX
The Byzantines' View

of the Franks
T HE FRANKS, though a minority of the population in Gaul, did

play a part in the Merovingian armies. For generations, how-
ever, historians have accepted the descriptions of Prankish tactics
and armament given by the Byzantine writers Procopius and Aga-
thias as the basis for understanding the entire Merovingian mili-
tary establishment.1 A careful study of these texts suggests they
should be reevaluated.2 Toward the middle of the sixth century
Procopius wrote concerning the Franks:
At this time [539] the Franks, hearing that both the Goths
and Romans had suffered severely by the war . . . for-
getting for the moment their oaths and treaties . . . (for
this nation [lOvo-s] in matters of trust is the most treach-
erous in the world), they straightway gathered to the num-
ber of one hundred thousand under the leadership of Theu-
debert, and marched into Italy: they had a small body of
cavalry about their leader, and these were the only ones
armed with spears [Sdpara], while all the rest were foot
1
Bachrach, "Procopius, Agathias, and the Frankish Military," pp. 435-
43a2
The only specific criticism of these texts and their defenders appears in
a footnote by Gustav Roloff, "Die Umwandlung des frankischen Heers,"
p. 392, n. 1. Walter Goffart, "Byzantine Policy," p. 77, n. 12, maintains that
Agathias 's descriptive chapters on the Franks contain outrageous errors. In
this he follows E. Cougny, Extraits des auteurs grecs concernant la g6o-
graphie et I'histoire des Gaules, V (Societe de ITiistoire de France [Paris,
1866], 415, n. 1), Unfortunately, neither author comments on Agathias's
description of the Franks' military tactics and armament. Cf. A. Cameron,
"How Did the Merovingian Kings Wear Their Hair?" RBPH, XLIII (1965),
1208, and "Agathias on the Early Merovingians," pp. 130-132, 139-140.
131
soldiers having neither bows nor spears [Sopara], but each
man carried a sword and shield and one axe. Now the iron
head of this weapon was thick and exceedingly sharp on
both sides, while the wooden handle was very short. And
they are accustomed always to throw these axes at one
signal in the first charge and thus to shatter the shields of
the enemy and kill the men.3
Some two decades after Procopius gave this account, Agathias
wrote:
The military equipment of this people [20vos] is very simple.
. . . They do not know the use of the coat of mail or
greaves and the majority leave the head uncovered, only a
few wear the helmet. They have their chests bare and backs
naked to the loins, they cover their thighs with either
leather or linen. They do not serve on horseback except in
very rare cases. Fighting on foot is both habitual and a
national [Trarptov] custom and they are proficient in this. At
the hip they wear a sword and on the left side their shield
is attached. They have neither bows nor slings, no missile
weapons except the double edged axe and the angon
[ayywvas] which they use most often. The angons are spears
[Sopara] which are neither very short nor very long; they
can be used, if necessary, for throwing like a javelin, and
also, in hand to hand combat. The greater part of the angon
is covered with iron and very little wood is exposed. Above,
at the top of the spear, on each side from the socket itself
where the staff is fixed, some points are turned back, bent
like hooks and turned toward the handle. In battle, the
Frank throws the angon, and if it hits an enemy the spear
is caught in the man and neither the wounded man nor
anyone else can draw it out. The barbs hold inside the
flesh causing great pain and in this way a man whose
wound may not be in a vital spot still dies. If the angon
strikes a shield, it is fixed there, hanging down with the
butt on the ground. The angon cannot be pulled out be-
cause the barbs have penetrated the shield, nor can it be
cut off with a sword because the wood of the shaft is cov-
ered with iron. When the Frank sees the situation, he quick-
ly puts his foot on the butt of the spear, pulling down and
the man holding it falls, the head and chest are left un-
protected. The unprotected warrior is then killed either by
a stroke of the axe or a thrust with another spear [86pan].
Such is the equipment of the Frankish warriors. . . .4
8
Procopius, H.W., VI, xxv, 3ff.
4
Agathias, Hist., II, 5.
132
APPENDIX
Although it is clear that Agathias is giving a general descrip-
tion of Prankish armament and tactics, it is less obvious in Pro-
copius's case. Procopius begins by characterizing the army that
Theudebert led into Italy in 539, but he soon makes his account
a general one with a portrayal of the Prankish ax and tactics
"always" («c() used by the Franks. It is not unfair to argue that
Procopius, as well as Agathias, intended to describe Prankish
military customs in general rather than the armament and tactics
of a particular army in a particular campaign. Though Procopius,
as a member of Belisarius's staff in Italy, may have had the
opportunity to obtain information from men who had faced the
Franks in battle in 539, there is no evidence to support this. By
providing his readers with a grossly exaggerated figure for the
size of Theudebert's army, Procopius casts doubt upon the accu-
racy of his remarks in this passage.
Procopius asserts that the Prankish infantry, which he contends
made up the vast majority of Theudebert's army, do not carry a
spear ($6pv), while Agathias argues that the ango, which he
classes as a spear (So'pu), is the Prankish foot soldier's primary
weapon and spends almost two-thirds of his description demon-
strating how the weapon is used and why it is so effective. Aga-
thias's account may even indicate that at least some Prankish
footmen carried more than one spear: the ango mentioned above
and a spear (86pv) used for thrusting. In yet another context
Aagthias portrays Prankish infantry using a javelin (a/covi-iov).5
Although Procopius does not mention the ango, he maintains that
the throwing ax is the Prankish infantryman's key weapon, and
his description of its construction and use dominates his account
of Prankish tactics and armaments. Agathias also classifies the
Franks' ax as a missile weapon, but he pictures it as a hand
weapon as well.
A study of the archaeological evidence is of little help in clari-
fying the use of weapons since Prankish graves yield not only
angos and axes, but a variety of spearheads.6 Sidonius Apol-
linaris, writing during the second half of the fifth century, states
that the Franks used the ax and also the ango (lancea uncata). Is
it possible that Sidonius's pedantic letter to his friend Domnicius,
a military buff who was "fond of looking at arms and armed
5
Ibid., 1,15.
8
Salin, Civ. merov., I, 356, 377, 385, 389, for angos; III, 23ff, for axes;
and 13ff, for a variety of spears. Werner, "Bewaffnung und Waffenbei-
gabe," pi. V, shows that the distribution of ongo-finds are east of the Rhine
which strongly suggests that the Franks whom Agathias describes were not
recruited from Merovingian lands but from trans-Rhenish areas.
133
men," served as a basis for Procopius's and Agathias's narratives

written about a century later? Although Sidonius offers an ac-
count of Frankish warriors in battle, he does not confirm the tac-
tical formations described by Procopius and Agathias. Sidonius
demonstrates that the Frankish footmen used not only the
barbed spear, but a spear (hasta) which does not seem to have
been barbed.7 The importance of the spear to the Frankish foot-
man, as indicated by Libanius during the previous century, is
of course denied by Procopius.8
Gregory of Tours, who was much closer in proximity to the
Franks and much closer in time to Procopius and Agathias than
was Sidonius, describes many battles in which Franks took part,
but he does not confirm the tactics described by either Pro-
copius or Agathias. Gregory pictures the Franks using their axes
on five separate occasions. In only one case, however, is the ax
characterized as being thrown; in the other instances it is used as
a hand weapon. Gregory does not deal with the strong points of
the ango as Agathias does — in fact he does not even mention it,
although he often writes about spears being used. The writ-
ten sources and the archaeological evidence indicate that the
short sword (scramasax) was a significant Frankish weapon, yet
neither Procopius nor Agathias seems to know of its existence.9
Procopius and Agathias do concur on several points. They both
maintain that the Franks did not use the bow. Gregory of Tours,
however, who probably knew more about the Franks than did
his Byzantine contemporaries, disagrees with them on this point,
quoting Sulpicius Alexander to the effect that the Franks had
used the bow since at least the fourth century. The archaeological
evidence confirms Gregory's view, for numerous arrowheads
have been found in Frankish graves.10 Procopius and Agathias
also agree that horesmen were unimportant among the Franks.
According to Procopius, the Franks had a very few horsemen
gathered about their leader, and Agathias claims that the Franks
served on horseback only on rare occasions. Both these descrip-
7
Sidonius Apollinaris, Ep., 4, 20, and Pan., 5, 247ff.
8
Libanius, oratio, lix, 275.
8
Gregory, Hist., IX, 35, for the ax being thrown; two instances in II, 27;
VII, 14, VIII, 19, for the ax being used as a hand weapon. For spears of
all kinds, IV, 30, V, 32, IX, 10, III, 14, V, 25, VII, 29; and for the
scramasax, IV, 51. For the importance of the scramasax see H. Althoff,
Waltharii Poesis (Leipzig, 1905), pp. 376-377, and Salin, Civ. merov., Ill,
45ff.
10
Gregory, Hist., V, 20; V, 48; X, 16; and for the quotation from Sulpici-
us, II, 9. For arrowheads in Frankish graves, Salin, Civ. merov., I, 356, and
II, 87, 239, 241.
134
APPENDIX
tions, however, are relative judgments; what may be a small
number to one observer may well be considered a large number
by another, whereas what might be a rare occasion to one might
seem reasonably frequent to another. The observer's norm or
basis for calculation determines such judgments and the norm
of the classical world was one horseman to every ten footmen.11
For the Byzantine army of the mid-sixth century, field forces
seem to have been about fifty percent infantry and fifty percent
cavalry. It should also be noted that for contemporaries, the
Byzantine cavalry was the core of the army, and the infantry
played an insignificant role.12 Seen in light of their norm, Pro-
copius and Agathias may well have thought that horsemen were
few among the Franks or rarely used by them. In lieu of any
more definite data (if Procopius and Agathias are to be trusted
on this point at all) the sole conclusion that can be drawn is that
horsemen were of less importance to the Franks as an ethnic
group than they were to the Byzantines. There is, however, sub-
stantial information about Frankish military customs from which
a much different interpretation emerges. Agathias himself writes
of a small battle near Rimini in 554, in which the Byzantines
fought against a force of some 2000 Franks, about half of whom
were horsemen.13 Perhaps this was one of those rare occasions
when, according to Agathias, the Franks fought on horseback.
In 531 Theuderic I and Chlotar I led an expedition of what
seems to have been purely Frankish warriors into Thuringia and
sought out the main force of the enemy. Shortly before the two
armies were to engage, the Thuringians managed to maneuver
the Franks into what they hoped would be a disadvantageous
position. They dug pits in a field and covered them with sod so
that the approaching Frankish horsemen would be unaware of
the trap. When the Franks charged across the field to meet the
Thuringians, their horses were tripped up by the ditches so that
they were slowed down and almost had to turn back.14
The elaborate preparations made by the Thuringians suggest
that they had had some previous experience with Frankish caval-
ry; because they expected the Franks to be on horseback, they
had to devise plans to combat their mounted tactics. The Thur-
ingians may well have learned the capabilities of the Frankish
11
J. K. Anderson, Ancient Greek Horsemanship (Berkeley, 1961), p. 141.
12
Bury, Later Roman Empire, II, 77-78, 85, 132, 136ff. Robert Grosse,
Romische Militarsgeschichte vom Gallienus bis zum Beginn der byzantin-
ischen Themenverfassung (Berlin, 1928), pp. 283ff, and 313ff.
13
Agathias, Hist., I, 21.
14
Gregory, Hist., Ill, 7.
135
horsemen some fifteen years earlier when Theuderic intervened
quite successfully in one of their civil wars.15 Gregory of Tours
relates the story of the charge of Frankish horsemen in detail
without a word of surprise concerning their equestrian tactics.
Gregory, of course, lived in Gaul and was probably accustomed
to seeing Frankish horsemen. The Thuringians certainly were not
surprised by the Franks' use of horsemen in 531; indeed they had
prepared for it. It is difficult to say how Procopius and Agathias
would have reacted had they witnessed this battle.
Horsemen also seem to have been an important element of
Clovis's army. Before his campaign against the Visigoths in 507,
Clovis issued specific orders to his troops concerning the taking
of fodder and water for their mounts. Although these commands
may well have been meant primarily for his Armorican allies
(among whom were a significant number of Alan horsemen), it
was apparently an unimportant Frank — not a noble or royal
bodyguard — who was executed for violating the orders.16 To
support further the conclusion that cavalrymen played a signifi-
cant role in Clovis's forces, it should be recalled that Clovis and
his followers were identified by at least one comparatively well-
informed contemporary as descendants of the Sicambri who were
well known for their equestrian prowess.17 Clovis's father, Chil-
deric, was so equestrian-oriented that he had the head of his
warhorse entombed with him at Tournai.18 The Franks of the
fifth and even the fourth centuries had gained such a reputation
for their horsemanship that they were recruited by the empire
for cavalry regiments. Of the four primarily Frankish units listed
in the Notitia Dignitatum in approximately 425, all were cavalry
regiments.19
Although Frankish equestrian prowess seems to have been
ignored in the East, it was widely recognized in the West. Ra-
banus Maurus quotes a common Frankish proverb to the effect
that one must begin at puberty if one is to become an effective
horseman.20 Einhard asserts that the chase and equestrian exer-
cises were Frankish national customs.21 What some scholars con-
15
18
Ibid., Ill, 4.
17
Ibid., 11, 37.
Caesar, B.C., VI, 35.
18
Cf. Wallace-Hadrill, Long-Haired Kings, p. 162.
19
See ch. I, n. 33.
20
Rabanus Maurus, De procinctu Romanae miliciae, p. 3: "Unde et vul-
garicum proverbium ac nostris familiare est quod dicitur: in pube posse fieri
equitem, majoris vero aetatis aut vix aut numquam."
a
Einhard, V. Karoli, ch. 22: "Exercebatur adsidue equitando ac venando:
136
APPENDIX
sider the national arms of the Ripuarian Franks are listed in the
first extant copy of their law code (about 630), which suggests
that the Ripuarians were also horsemen.22 Equestrian activity
among the Salians seems to have equaled that of the Ripuarians.
Chlotar II, for example, was as devoted to the chase as Charle-
magne.23 This was also true of Chilperic I, who was killed while
dismounting from his horse after returning from the hunt, and
of his son Merovech who rode out to the chase despite great
dangers.24
An abundance of horses, which is a prerequisite for effective
cavalry organization, is indicated clearly by the frequent legisla-
tion regarding them even in the earliest compilation of the Lex
Salica (about 510).25 Horses appear so frequently in the Salian
Prankish laws that at least one noted medieval economic historian
has been led to contend that the Franks had a great many horses
and thus could use them for plowing when everyone else used
oxen.28
The significance of horses is further illustrated by the elaborate
organization of horse-breeding estates under the Merovingian
kings. These estates were under the central control of the con-
stable (comes stabuli) who was an important member of the
royal household. Each estate was administered by a bailiff who
in turn supervised numerous wranglers and grooms. Because of
their jobs these men seem to have enjoyed a preferred position
in society which others of comparable legal or social status, but
of different occupational status, did not have.27
quod illi gentilicium erat, quia vix villa in terns natio invenitur quae in hac
arte Francis possit aequari."
"Lex Rib., 40 (36), 11.
21
Fred., IV, 42: ". . . Chlotharius . . . venacionem feramm nimium as-
siduae utens. . . ."
24
Gregory, Hist., V, 14: "Ait file praesto putans esse interfectores, ait ad
Merovechum; "Ut quid hie quasi signes et timidi resedemus et ut hebetis
circa basilicam occulimur? Verdant enim equi nostri, et acceptis accipitribus,
cum canibus exerceamur venatione spectaculisque patulis iocundemur.'" VI,
46: "Quadam vero die regressus de venatione iam sub obscura nocte, dum
de equo susceperitur et unam manu super scapulam pueri reteniret, ad-
veniens quidam eum cultro percutit sub ascellam iteratoque ictu ventrem
eius perforat; statimque profluente cupia sanguinis tarn per os quam per
aditum vulneris, iniquum fudit spiritum."
* Lex Sal., 9, 1; 10, 1; 23; 27, 3-5; 37, 15; 38, 1 and 3, 6 [7, 8]; 47, 1;
65,1.
* Charles Edmond Perain, "The Evolution of Agricultural Technique,"
The Cambridge Economic History, ed. M. Postan et al., 2nd ed. (Cam-
bridge, 1966), I, 142.
27
Fustel de Coulanges, Hist, des inst., IV, 148-149. Gregory, Hist., Ill,
15; V, 39; VIII, 40; IX, 38; and Gregory, G.M., I, 29.
137
If Procopius's and Agathias's remarks concerning the horse-
men of the Franks are understood as meaning that they had a
number of horsemen so inconsequential as to be hardly worth
notice, then the Byzantine writers and their interpreters are sure-
ly mistaken. If, on the other hand, their comments are taken to
imply that horsemen were of less importance to the Franks than
to the Byzantines, then they are probably correct. The thrust
of the evidence, however, seems to indicate that the Franks had
a relatively significant body of horsemen within the Merovingian
military establishment.
The utilization of Procopius's and Agathias's judgments con-
cerning Prankish armament and tactics would seem to require
more caution than has been usual heretofore. If one were to insist
that Procopius's description was of a specific force in a particular
campaign, the danger of generalizing about the Frankish military
from it is manifest. If, however, Procopius's account is a general
one, as argued here, then its defenders must realize that he and
Agathias contradict each other as decisively as both are contra-
dicted by sources closer to events and presumably better in-
formed.
Though both Byzantine authors list weapons that were surely
used by the Franks, the tactics they characterize are otherwise
unconfirmed. It must be made very clear that Procopius's and
Agathias's narratives are only of limited value for the study of
the Franks as an ethnic group and of no value if used as a
description of the heterogeneous Merovingian military.
138
Bibliography
Bibliography
Primary Sources
Agathias Myrinaei Historiarum Libri Quinque, ed. R. Keydell. Berlin, 1967.
Beowulf, 3rd ed., ed. F. Klaeber. Boston, 1951.
Bon. ep.: S. Bonifatii et Lulli espistolae, ed. M. Tangl. MGH Ep., vol. I.
Hannover, 1916.
Caesar, B. G.: C. lull Caesaris Commentarii rerum in Gestarum, ed.
T. Holmes. Oxford, 1914.
Cap. lex Sal. add.: Pactus Legis Salicae, ed. K. Eckhardt. MGH LL,, vol. IV,
pt. 1. Hannover, 1962.
Chron. Caesaraug., s.a.: Chronicorum Caesaraugustanorum reliquiae a.
CCCLV-DLXXXI, ed. T. Mommsen. MGH AA., vol. XI, reprint. Berlin,
1961.
Chron. Moissiac: Chronicon Moissiacensi, ed. G. Pertz. MGH SS., vol. II.
Hannover, 1829.
Einhard, V. Karoli: £iginhard, Vie de Charlemagne, ed. and trans.
L. Halphen. Paris, 1923.
Form. Andec.: Formulae Merowingici et Karolini aevi, ed. K. Zeumer. MGH
LL., vol. V. Hannover, 1886.
Fred.: Chronicarum quae dicuntur Fredegarii scholastici libri IV, ed.
B. Krusch. MGH SSRM., vol. II. Hannover, 1888.
Fred, con't.: Chronicarum quae dicunter Fredegarii scholastici libri IV cum
Continuationibus, ed. B. Krusch. MGH SSRM., vol. II. Hannover, 1888.
Gesta Dagoberti, ed. B. Krusch. MGH SSRM., vol. II. Hannover, 1888.
Gesta Francorum, Patrologia Latinae, ed. J. P. Migne, vol. CXXXIX. Paris,
1880.
Gregory, G.M.: Gregorii episcopi Turonensis, Liber in Gloria Martyrum,
ed. B. Krusch. MGH SSRM., vol. I. Hannover, 1885.
, Hist.: Gregorii episcopi Turonensis, Libri Historiarum, ed. B.
Krusch and W. Levison. MGH SSRM., vol. I, pt. 1. Hannover, 1951.
_, Virt. S. Mart.: Gregorii episcopi Turonensis, Liber I-IV de
virtutibus sancti Martini episcopi, ed. B. Krusch. MGH SSRM., vol. I.
Hannover, 1885.
_, V.P.: Liber vitae patrum, ed. B. Krusch. MGH SSRM., vol. I.
Hannover, 1885.
_, Liber de passione et virtutibus Santi Juliani Martyris, ed.
B. Krusch. MGH SSRM., vol. I. Hannover, 1885.
Isid., Hist. Goth.: Isidori lunioris episcopi Hispalensis, historia Gothorum,
141
Wandakmtm, Sueborum, ad. a DCXXW, ed. T. Mommsen. MGH AA.,
vol. XI, reprint, Berlin, 1961.
Isidori continuatio Hispana, ed. T. Mommsen. MGH AA., vol. XI, reprint.
Berlin, 1961.
John Biclar: Johannis abbatis BicTarensis chronica a DLXVII-DXC, ed.
T. Mommsen. MGH AA., vol. XI, reprint. Berlin, 1961.
Jordanes, Getica: Jordanis, De origine actibus Getarum, ed. T. Mommsen.
MGH AA., vol. V, pt. 1, reprint. Berlin, 1961.
L.B.: Leges Burgundionum, ed. L. De Salis. MGH LL., vol. II. Hannover,
1892.
Lex Rib.: Lex Ribuaria, ed. R. Buchner. MGH LL., vol. III. Hannover,
1954.
Lex Rom. Burg.: Lex Romano Burgundionum, ed. L. De Salis. MGH LL.,
vol. II. Hannover, 1892.
Lex Sal: Foetus Legis Salicae, ed. K. Eckhardt. MGH LL., vol. IV, pt. 1.
Hannover, 1962.
L.H.F.: Liber Historiae Francorum, ed. B. Krusch. MGH SSRM., vol. II.
Hannover, 1888.
Libanius, oratio; Libanii Opera, ed. R. Foerster, vol. IV. Leipzig, 1908.
Marculf, form.: Formulae Merowingici et Karolini aevi, ed. K. Zeumer.
MGH LL., vol. V. Hannover, 1886.
MGH Cap.: Capitularia Regnum Francorum, ed. A. Boretius. MGH LL.,
vol. II. Hannover, 1883.
Mir. Martini Abbatis: Miracula Martini Abbatis Vertavensis, ed. B. Knisch.
MGH SSRM., vol. III. Hannover, 1896.
Notitia Dignitatum utriusque imperil, ed. O. Seeck. Berlin, 1876.
N. Vol.: Valentinian III, Novellae, Codex Theodosianus, ed. T. Mommsen,
vol. II. Berlin, 1905.
Pardessus: Diplomata Chartae, Epistolae, Leges, 2 vols., ed. J. M. Pardessus.
Reprint. Darmstadt, 1969.
Pass. Leod.: Passio Leudegarii I, ed. B. Krusch. MGH SSRM., vol. V.
Hannover, 1916.
Paul, Hist.: Pauli Diaconi Historia Langobardontm, ed. L. Bethmann and
G. Waitz. MGH SSRL., Hannover, 1878.
Procopius, H.W.: History of the Wars, 5 vols., ed. and trans. H. Dewing.
London, 1914-1928.
, H.W.: Procopius Caesaraensis, Opera, ed. J. Haury and
G. Wirth, vol. II. Leipzig, 1963.
Rabanus Maxims, "De procinctu Romanae miliciae," ed. E. Diimmler,
Zeitschrift fur deutschen Altertum, XV (1872), 443-451.
Sid. Apol.: Sidonii Apollinaris, Epistulae et Carmina, ed. C. Luetjohann.
MGH AA., vol. VIII. Berlin, 1887.
Tacitus, Germania, ed. R. Robinson. Middletown, Conn., 1935.
V. Caes.: Vita Caesarii epfecopi Arlatensis, ed. B. Krusch. MGH SSRM.,
Hannover, 1896.
V. Columbani, ed. B. Krusch. MGH SSRG., Hannover, 1905.
V. Dalmat.: Vita Dalmatii episcopi Ruteni, ed. B. Krusch. MGH SSRM.,
vol. III. Hannover, 1896.
V. DanieUs: Vita S. Danielis Stylitae, ed. H. Delehaye, Analecta Eollandi-
ana, XXXII (1913), 121-214.
V. Desid.: Vita S. Desiderii episcopi Viennensis, ed. M. Bouquet, vol. III.
Paris, 1869.
142
BIBLIOGRAPHY
V. Epiphann, ed. F. Vogel. UGH AA., vol. VII. Berlin, 1883,
V. Eptadii, ed. B. Krusch. UGH SSRM., vol. III. Hannover, 1896.
V. Eucherii, episcopi Aurelianensis, ed. W. Levison. MGH SSRM., vol. VII.
Hannover, 1919.
V. Galact.: De S. Galactario Episcopo Lascurrensi, Act. sanct., vol.
XXVII. Paris, Rome, July 1868.
V. Genov.: Vita Genovefae mrginis Parisiensis, ed. B. Krusch. MGH SSRM.,
vol. III. Hannover, 1896.
V. Landiberti, ed. B. Krusch. MGH SSRM., vol. IV. Hannover, 1913.
V. Maximini, ed. B. Krusch. MGH SSRM., vol. III. Hannover, 1896.
V. Remigii, ed. B. Krusch. MGH SSRM., vol. Ill, Hannover, 1896.
V. Valent.: Vita S. Valenti Presbt/teri, ed. M. Bouquet, vol. III. Paris, 1869.
V. Wilfrithi: The Life of Bishop Wilfrid by Eddius Stephanus, ed. and
trans. B. Colgrave. Cambridge, 1927.
V. Willibrodi, ed. W. Levison. MGH SSfiM., vol. VII. Hannover, 1919.
Secondary Works
Althoff, H. Waltham Poesis. Leipzig, 1905.
Anderson, J. K. Ancient Greek Horsemanship. Berkeley, 1961.
Bachrach, Bernard S. "The Alans in Gaul," Traditio, XXIII (1967),
476-489.
"Was There Feudalism in Byzantine Egypt?" Journal of the
American Research Center in Egypt, VI (1967), 163-166.
"A Note on Alites," Byzantinische Zeitschrift, LXI (1968), 35.
"Another Look at the Barbarian Settlement in Southern Gaul,"
Traditio, XXV (1969), 354-358.
"The Origin of Armorican Chivalry," Technology and Culture,
X (1969), 166-171.
"Charles Martel, Mounted Shock Combat, the Stirrup, and
Feudalism," Studies in Medieval and Renaissance History, VII (1970),
49-75.
"Procopius, Agathias, and the Prankish Military," Speculum,
XLV (1970), 435-441.
_. "Procopius and the Chronology of Clovis's Reign," Viator, I
(1970), 21-31.
"The Feigned Retreat at Hastings," Mediaeval Studies, XXXIII
(1971) ,264-267.
Baudot, M. "Localisation et datation de la premiere victoire remportee
par Charles Martel contre les Musulmans," Memoires et documents
publics par la Societe de I'tcole des Chartes, XII (1955), 93-105.
Bloch, Marc. Feudal Society. Trans. L. A. Manyon, London, 1961.
Bodmer, Jean Pierre. Der Krieger der Merowingerzeit und seiner Welt.
Zurich, 1957.
Boutaric, Edgard. Institutions militaires de la France avant les armies
permanentes, Paris, 1863.
Brunner, Heinrich. "Die Reiterdienst und die Anfange des Lehnwesens,"
ZRG, VIII (1887), 1-38.
Forschungen zur Geschichte des deutschen und franzosischen
Rechts. Stuttgart, 1894.
Deutsche Rechtsgeschichte. 2nd ed. Munich, 1928.
Buchner, R. "Die Rechtsquellen." In Wattenbach-Levison Deutschlands
Geschichtsquellen in Mittelalter. Weimar, 1953.
143
Bullough, Donald. "Europae Pater: Charlemagne and His Achievement in
the Light of Recent Scholarship," EHR, LXXV (1970), 84-89.
Bury, J. B. "The Notitia Dignitatum," }RS, X (1920), 131-154.
. History of the Later Roman Empire. 2 vols. Dover ed. New York,
1958.
Butler, R. M. "Late Roman Town Walls in Gaul," The Archaeological
Journal, CXVI (1959), 25-50.
Cam, Helen. Local Government in Francia and England. London, 1912.
Cameron, Averil. "How Did the Merovingian Kings Wear Their Hair?"
RBPH, XLIII (1985), 1203-1216.
"Agathias on the Early Merovingians," Annali della Scuola
Normale Superiore di Pisa, XXXVII (1968), 95-140.
Claude, Dietrich. "Untersuchen zum fruhfrankischen Comitat," ZflG,
LXXXI (1964), 1-79.
Codera, F. "Manusa y el duque Eudon," Estudios criticos de historia drabe-
espanola-. In Collection de los estudios drabes, vol. VII, pp. 140-169.
Saragossa, 1903.
Collingwood, R. G., and J. N. L. Myres. Roman Britain, 2nd ed. Oxford,
1937.
Cougny, E. Extraits des auteurs grecs concernant la geographic et fhistoire
des Gaules, vol. V. Societe de 1'histoire de France, Paris, 1866.
Cronne, A. H. "The Origins of Feudalism," History, XXIV (1939), 251-259.
Dalton, O. M. The History of the Franks by Gregory of Tours. 2 vols. Ox-
ford, 1927.
Dannenbauer, H. "Die Freien im karolingische Heer." In Aus Verfassungs-
und Landesgeschichte: Festschrift fur T. Mayer, vol. I, pp. 49-64. Lindau,
1954.
Dauzat, A., and C, Rostaing. Dictionnaire etymologique des noms de Ifeux
en France. Paris, 1963.
Delbriick, Hans. Numbers in History. London, 1913.
Geschichte der Kriegskunst in Rahmen der politischen Geschichte.
6 vols. 2nd ed. Berlin, 1920-1932.
Deloche, Maximin. La trustis et I'antrustion royal sous les deux premieres
races, Paris, 1873.
Dill, Samuel. Roman Society in Gaul in the Merovingian Age. London, 1926.
Dippe, Oskar. Gefolgschaft und Huldigung im Reiche der Merotoinger.
Wandsbeck, 1889.
Doehaerd, R. "La richesse des Merovingiens." In Studi in onore di Gino
Luzzatto, vol. I, pp. 30-46. Milan, 1949,
Dopsch, Alfons. Wirtschaftliche und soziale Grundlagen der europdischen
Kulturentivicklung, 2 vols. 2nd ed. Vienna, 1924.
"Die Leudes und das Lehnwesen," Ml&G, XLI (1926), 35-43.
"Beneficialwesen und Feudalitat," MIOG, XLVI (1932), 1-36.
Dupraz, Louis. Contribution a 1'histoire du Regnum Francorum. Fribourg,
1948.
Erban, W. "Zur Geschichte des karolingischen Kriegswesens," HZ, CI (1908),
321-336.
Ewig, Eugen. "Die frankischen Teilungen und Teilreiche (511-613),"
Akademie der Wissenschaften und der Literatur, IX (1952), 651-715.
"Die frankischen Teilreiche im 7 Jahrhundert (613-714),"
Trierer Zeitschrift, XXII (1953), 85-144.
Fehr, H. "Das Waffenrecht der Bauern im Mittelalter," ZRG, XXXV
(1914), 111-211.
144
BIBLIOGRAPHY
Fischer, K. The Burgundian Code. Philadelphia, 1949.
Frauenholz, Eugen von. Das Heerwesen der germanischen Friihzeit. Mu-
nich, 1935.
Freeman, E. A. Western Europe in the Eighth Century and Onward. Lon-
don, 1904.
Fuller, J. F. C. The Decisive Battles of the Western World. 3 vols. London,
1954-1956.
Fustel de Coulanges, N. D. Histoire des institutions politiques de I'ancienne
France. 6 vols. Paris, 1888-1914.
Ganshof, F. L. "Note sur le sens de 'Ligeris' au titre XLVII de la loi salique
en dans le 'Querolus.'" In Historical Essays in Honor of James Tait, pp.
111-120. Manchester, 1933.
"Note sur les engines de 1'union du benefice avec vasalite." In
Etudes d'histoire dediee a la memoire de Henri Pirenne, pp. 173-189.
Brussels, 1937.
Feudalism. Trans. P. Grierson. London, 1952.
"A propos de la cavalrie dans les armees de Charlemagne," CRAI
(1952) ,531-537.
"L'Origine des rapports feodo-vassaliques," I problemi della civ-
ilita carolingia: Settimane di studio del Centra Italiano di Studi sull' Alto
Medioeveo, I (1954), 27-53.
Goffart, Walter. The Le Mans Forgeries: A Chapter from the History of
Church Property in the 9th Century. Cambridge, Mass., 1966.
Grenier, Albert. Manuel d'archeologie Gallo-Romain. 5 vols. Paris, 1931-
1960.
Grohler, H. Vber I]rsprung und Bedeutung der franzosischen Ortsnamen. 2
vols. Heidelberg, 1913-1935.
Grosse, Robert. Rdmische Militarsgeschichte vom GaUienus bis zum Beginn
der byzantinischen Themenverfassung. Berlin, 1928.
Guilhiermoz, Paul. Essai sur I'origine de la noblesse en France au Moyen
Age. Paris, 1902.
Hay, Denis. Europe: The Emergence of an Idea. Edinburgh, 1957.
Hodgkin, Thomas. Italy and Her Invaders. 8 vols. London, 1880-1899.
Hollister, C. W. Anglo-Saxon Military Institutions. Oxford, 1962.
Hoyt, R. S. Europe in the Middle Ages. 2nd ed. New York, 1966.
Jones, A. H. M. The Later Roman Empire. 2 vols. Norman, Okla., 1964.
Jullian, C. Histoire de la Gaule Romcdne, vol. VIII. Paris, 1926.
Kroell, Maurice, "fitude sur 1'institution des Lites en franc." In Etudes d'his-
toire juridiques offertes a Paul Frederic Girard, vol. II, pp. 125-208. Paris,
1913.
Kurth, G. Histoire poetique des Merovingiens. Paris, 1893.
Clovis. 2 vols. 2nd ed. Paris, 1901.
Etudes franques. 2 vols. Paris, 1919.
Latouche, Robert. The Birth of the Western Economy. London, 1961.
Lecointre, J. L. "La Bataille de Poitiers entre Charles Martel et les Sarasins:
1'Histoire et la legende; origine de celle-ci," Bulletin de la Societe des
Antiquaires de I'Ouest, 3rd ser., VII (1924), 632-642.
Leighton, Albert. "Early Medieval Transport." Ph.D. dissertation, Berkeley,
1964.
Lesne, E. La Propriete ecclesiastique en France. 6 vols. Paris, 1910-1943.
Levi-Provencal, £. Histoire de I'espagne musulmane. 3 vols. Paris, 1950.
Levillain, L., and C. Samaran. "Sur le lieu et la date de la bataille de Poi-
tiers en 732," BEC, XCIX (1938). 243-267.
145
.. "Campus Martius," EEC, CVII (1947-1948), 62-68.
Lewis, A. R. Emerging Medieval Europe, A.D. 400-1000. New York, 1967.
Liebmann, R. Der Untergang des thuringischen Konigsreichs in den Jahren
531-5. Meiningen, 1911.
Lindenschmidt, Ludwig. Handbuch der deutsche Altertumskunde. Braun-
schweig, 1880-1890.
Longnon, A. Geographie de la Gaule au VI' siecle. Paris, 1878.
Les noms de lieu de la France. Paris, 1920.
Lot, Ferdinand. "Les migrations saxonnes," RH, CXIX (1915), 1-40.
L'lmpot fancier et la capitation personelle. Paris, 1928.
"La conquete du pays d'entre Seine-et-Loire par les Francs,"
RH, CLXV (1930), 241-253.
The End of the Ancient World and the Beginnings of the Middle
Ages. Trans. P. M. Mariette. New York, 1931.
"Origine et nature du benefice," Anuario de Historia del Derecho
Espanol, X (1933), 175-185.
"Les limites de la Sapaudia," Revue savoisienne, LXXVII
(1935), 146-156.
"La 'Notitia Dignitatum utriusque Imperii . . . ,'" RE A,
XXXVIII (1936), 285-338.
"La victoire sur les Alamans et la conversion de Clovis," RBPH,
XVII (1938), 63-69.
L'Art militaire et les armees au moyen age en Europe et dans
proche orient. 2 vols. Paris, 1946.
C. Pfister, and F. L. Ganshof. Les destinees de I empire en Occi-
dent de 395 a 888. 2nd ed. Paris, 1940.
MacMullen, Ramsey. Soldiers and Civilians in the Later Roman Empire.
Cambridge, Mass., 1963.
Mangoldt-Gaudlitz, Hans von. Die Reiterei in den germanischen und frank-
ischen Heeren bis zum Ausgang der deutschen Karolinger. Berlin, 1922.
Mercier, E. "La Bataille de Poitiers et les vraies causes du recul de Finva-
sion arabe," RH, VII (1878). 1-13.
Mercier, M., and A. Seguin. Charles Martel et la Bataille de Poitiers. Paris,
1944.
Mitteis, Heinrich. Lehnrecht und Staatsgewalt. Weimar, 1958.
Musset, Lucien. Les invasiones. 2nd ed. Paris, 1969.
Niemeyer, J. F. Mediae Latinitatis Lexicon Minus. Leiden, 1963.
Odegaard, Charles. Vassi and Fideles in the Carolingian Empire. Cambridge,
Mass., 1945.
Oman, Charles. The Art of War in the Middle Ages. Oxford, 1885.
The Art of War in the Middle Ages. Oxford, 1898.
History of the Art of War in the Middle Ages. 2 vols. London,
1923.
O'Rahilly, Thomas F. Early Irish History and Mythology. Dublin, 1946.
Paribeni, R. "Le dimore dei potentiores ne basso impero," Mitteilungen des
deutschen archdologischen Instituts, LV (1940), 131-148.
Peucker, Eduard von. Das deutsche Kriegswesen. 2 vols. Berlin, 1860-1864.
Pirenne, H. "Le fisc royal de Tournai." In Melanges d'histoire du moyen age
offerts a M. Ferdinand Lot. Paris, 1925.
Planitz, Hans. "Die Scharmannen von Priim." In Festschrift fur Heinrich
Lehmann, pp. 55-69. Berlin, 1937.
Postan, M., ed. Cambridge Economic History, vol. I. 2nd ed. Cambridge,
1966.
146
BIBLIOGRAPHY
Roloff, Gustav. "Die Umwandlung des frankischen Heers von Chlodowig bis
Karl den Grossen," Neue Jahrbiicher fur das klassische Altertum, IX
(1902), 389-399.
Roth, Paul. Geschichte des Beneficialwesens von dltesten Zeiten bis ins
zehnte Jahrhundert. Erlangen, 1850.
Feudalitdt und Unterthanverband, Weimar, 1863.
Rouche, Michel. "Les Aquitans ont-ils trahi avant la Bataille de Poitiers?"
MA, LXXIV( 1968), 5-26.
Roy, Jean-Henri, and Jean Devoisse. La Bataille de Poitiers. Paris, 1966.
Riibel, Karl. "Frankisches und spatromisches Kriegswesen," Banner Jahr-
biicher, CXIV (1906), 136-142.
Russell, Jeffrey. Medieval Civilization. New York, 1968.
Salin, Edouard. La Civilisation m4rovingienne. 4 vols. Paris, 1949-1959.
Sanchez-Albornoz, Claudio. En torno a las Origenes del Feudalismo. 3 vols.
Mendoza, 1942.
Sawyer, Peter. The Age of the Vikings. London, 1962.
, and R. Hilton. "Technical Determinism: The Strirrup and the
Plough," Past and Present, XXIV (1963), 90-100.
Schmidt, Ludwig. Geschichte der deutschen Stdmme bis zum ausgange der
vdlkerwanderung. 8 vols. Berlin, 1904-1908.
Seeck, Otto, "Das deutsche Gefolgswesen auf romischen Boden," ZRG,
XVII (1896), 97-119.
Sprandel, Rolf. "Dux et Comes in der Merowingerzeit," ZRG, LXXIV
(1957) ,41-84.
"Bemerkungen zum friihfrankischen Comitat," ZRG, LXXXII
(1965), 288-291.
Stein, Ernst. "Die Organisation der westromischen Grenzeverteidigung im V.
Jahrhundert und das Burgunderreich am Rhein," Bericht des deutsches
archdologisches Institut, Rom.-Germ. Komm., XVIII (1928), 92-114.
Eistoire du Bos-Empire. 2 vols. Paris, 1949-1959.
Stein, F. Adelsgrdber des achten Jahrhundert in Deutschland. 2 vols. Berlin,
1967.
Stein, S. "Der Romanus in des frankischen Rechtsquellen (Eine Antwort),"
Historische Vierteljahrschrift, XXXI (1936), 232-250.
Stephenson, Carl. "The Origin and Significance of Feudalism," AHR, XLVI
(1941), 788-812.
Stevens, C. E. Sidonius Apottinaris and His Age. Oxford, 1933.
Strayer, Joseph. "The Two Levels of Feudalism." In Life and Thought in
the Middle Ages. Ed. Robert Hoyt. Minneapolis, 1967.
Stroheker, K. F. Der senatorische Adel in spd'tantiken Gallien. Tubingen,
1946.
Taylor, Charles. "Note on the Origin of the Polyptych." In Melanges d'his-
toire offerts a Henri Pirenne, vol. II, pp. 475-482. Brussels, 1926.
Tessier, Georges, Le bapteme de Clovis. Paris, 1964.
Thompson, E. A. "The Settlement of Barbarians in Southern Gaul," JRS,
XLVI (1956), 65-75.
"The Barbarian Kingdoms in Gaul and Spain," Nottingham Me-
dieval Studies, VII (1963), 3-33.
The Early Germans. Oxford, 1965.
The Goths in Spain, Oxford, 1969.
Van de Vyver, Andre. "La victoire contre les Alamans et la conversion de
Clovis," RBPH, XV and XVI (1936 and 1937), 859-914 and 35-94.
147
"Clovis et la Politique mediterranienne." In Etudes d'histoire
dediee a la memoire de Henri Pirenne, pp. 367-388. Brussels, 1937.
"L'unique victoire centre les Alamans et la conversion de Clovis
en 506," RBPH, XVII (1939), 793-813.
"La chronologic du regne de Clovis d'apres la legende et d'apres
1'histoire," MA, LIII (1947), 177-196.
Veeck, W. Die Alamannen in Wurttemberg. 2 vols. Berlin, 1931.
Verbruggen, J. F. De Krijgskunst in West-Europa in den Middeleeuwen.
Brussels, 1954.
"L'Armee et la strategic de Charlemagne." In Karl der Grosse,
vol. I, pp. 420-436. Dusseldorf, 1965.
Vercauteren, F. "Comment s'est-on defendu au IXe siecle dans 1'empire franc
contre les invasiones normandes?" XXX* congres de la Federation ar-
cheologique et historique de Belgique, pp. 117-132. Brussels, 1935.
Verlinden, Charles. Les origines de la frontiere linguistique en Belgique et
la colonisation franque. Brussels, 1955.
Voltini, H. "Prekarie und Beneficium," Vierteljahrschrift fur Sozial und
Wirtschaftsgeschichte, XVI (1923), 293-305.
Waitz, Georg. Vber die Anfdnge der Vassalitat. Gottingen, 1856.
Deutsche Verfassungsgeschichte. 8 vols. Berlin, 1874-1885.
Wallace-Hadrill, J. M. The Fourth Book of the Chronicle of Fredegar. Lon-
don, 1960.
The Long-Haired Kings. London, 1962.
Werner, Joachim. "Zur Entstehung der Reihengraberzivilisation," Archaeo-
logia Geographia, I (1950), 23-32.
"Bewaffnung und Waffenbeigabe in der Merowingerzeit," Ordi-
namenti Militari in Occidente nell 'Alto Medioevo (Settimane di Studio
del Centro Italiano di Studi sutt 'Alto Medioevo, XV [Spoleto, 1968]), I,
95-108.
White, Lynn, Jr. Medieval Technology and Social Change. Oxford, 1962.
Wolfram, Herwig. Intituhitio I: Lateinische Konigs- und FUrstentitel bis zum
Ende des 8. Jahrhunderts, UlOG, Erganzungband XXI, 1967.
Zollner, Erich. Geschichte der Franken. Munich, 1970.
148
Index
Index
Abbelin, Count, 81 Aprunculus, Bishop of Langres, 7
Abd al-Rahman, 101-102 Aquitaine, 38, 66, 69, 83, 85
Abundantius, Duke, 86 Arabs, 105
Adalgisel, Duke, 93, 97 Arele, 77
Aega, mayor of the palace, 94, 126 Arian Christians, 7
Aegilen, patrician, 58, 60, 62 Aridius, Gallo-Roman magnate, 9
Aetius, 3 Arigisel, Prankish magnate, 22, 30,
Agathias, 26, 113, 114, 131: on 32
Prankish warfare, 132-136; eval- Aries, 12, 17, 37
uation of, 137-138 Armati, 32, 49, 52
Agen, 85. See also Levies Armorici, 3, 10, 136
Aighyna, Saxon magnate, later Duke Arnebert, Duke, 84-85, 87
of Gascony, 84-85, 87 Arnulfings, 97
Aire, 67 Arvernians, 21, 31. See also Levies
Aix, 40 Audovald, Duke, 60
Alamans, 3, 5-8, 10, 12, 27, 36, 81, Audovar, 37-38, 70
86, 88, 101, 106, 124. See also Aurelianus, Gallo-Roman magnate, 8,
Laeti and Levies 15
Alans, 3, 10, 12, 15, 17, 36, 124 Austrasia, 28, 35, 47, 48, 67, 74, 75,
Alaric, King of the Visigoths, 11 82, 83, 84, 87-88, 97
Albi, 11 Austrasians, 28, 31, 32, 33, 43, 81,
Alsace, 81 83, 86, 97
Amalaric, King of the Visigoths, 26 Austrovald, Count, 62, 63
Amalbert, 94 Autun, 67, 78
Amalgar, Duke, 85, 87, 95 Auvergne, 20, 28. See also Levies
Amantius, patrician, 38 Auxerre, 48, 78
Amid, 52, 55 Avars, 36, 37, 43, 118
Amiens, 5 Avenches, 81
Amo, Lombard Duke, 40 Avignon, 9, 10, 17, 37, 39, 40, 54,
'Anbasa ibn Suhaim al-Kalbi, 101 55, 57, 67, 105, 111, 127
Angers, 47, 63. See also Levies Avius, magnate, 57
Ango, 132-134 Axes, 132-133
Angouleme, 12, 17, 53, 57. See also
Levies Barontus, Duke, 87
Aninsula, monastery of, 47, 48 Basilius, magnate, 38
Antestius, magnate, 63-64 Baudulf, magnate, 78
Antrustiones, 13, 14, 32, 33, 71, 72, Bavarians, 101, 108
97, 109 Bayeux, 10. See also Levies
151
Beaucaire, 62 Carloman, mayor of the palace, 106,
Belgica II, 5 108
Bella civilia, 36, 37, 43, 60 Casilinum, battle of, 27
Bellinzona, 60-61 Cautinus, Bishop of Auvergne, 7, 30
Beppolen, 63, 67 Cavalo, 78
Berchar, mayor of the palace, 98 Cavalry. See Horsemen
Bernay, 51 Cedinus, Duke, 61
Berre, battle of, 105 Celsus, patrician, 37-41
Berthar, Count, 78, 82, 95 Centenae, 25, 32, 34, 46, 71-72, 97,
Berthefred, Prankish magnate, 55, 108, 109, 124
64,65 Chadoind, referendary, 87
Bertoald, mayor of the palace, 76-77 Chalons, 29, 67
Bertoald, Saxon Duke, 86 Champagne, 48, 55, 75. See also
Berulf, Duke, 50-53 Levies
Besancon, 78, 80 Chanao, Count, 29
Beziers, 22 Charegisel, chamberlain, 43
Bladast, Duke, 53, 57 Chararic, regulus, 13
Blois, 57, 66, 76. See also Levies Charibert I, King, 29, 36, 38
Boantus, magnate, 64 Charibert II, King, 85
Bobo, Duke, 93 Charlemagne, 119
Bodilo, Prankish magnate, 95 Charles Martel, 99-106, 110-112,
Bordeaux, 12, 41, 48, 57, 67. See 115, 120, 122, 126, 127
also Levies Chartres, 10, 84. See also Levies
Boso, Duke, 63, 64, 67, 71 Chastel-Marlhac, 20-21
Boso of fitampes, magnate, 85 Chatuarian Franks, 14
Bourges, 50, 51, 53. See also Levies Chaubedo, magnate, 95, 112
Bretons, 52, 63, 64, 68 Childebert I, King, 18, 20, 22, 25,
Brioude, Church of Saint Julien at, 26, 29, 44, 58
20 Childebert II, King, 46-49, 54, 55,
Britain, 78 57, 60-61, 64-66, 72, 74-75
Britons, 3 Childebrand, Count, 104, 105, 110
Brittany, 29 Childeric, regulus, 4, 5, 14, 136
Brives-la-Gaillarde, 57 Childeric, Saxon magnate, 57
Brodulf, magnate, 84-85 Childeric II, King, 95, 97
Brunhild, Queen, 46-48, 77, 83-84, Childeric III, King, 107, 108
89 Chilperic I, King, 31, 36, 72, 79:
Bubo, Frisian chief, 104 and bella civilia, 41-43; death of,
Buccelin, Alaman Duke, 27, 89 56-57, 60, 137; forces of, 45, 51-
52, 54, 88; and local levies, 52, 54,
Bucellarii, 16, 23, 44 56, 66; and magnates, 44, 47, 50,
Burgundian kingdom, 20, 23, 32, 33, 55; military activity of, 45-49, 53
39, 41, 47, 55, 60, 67, 71, 74, 75, Childtrudis, 106
76, 84, 85, 86, 87: standing army Chlochiliach, Danish King, 18
of, 23, 41, 47, 55, 71, 76 Chlodomer, King, 18, 32
Burgundians, 3, 6, 7 Chlodovech, 38, 39-41, 43, 45, 47
Chlotar I, King, 18-22, 25, 28, 30-
Caberet, 62 32, 36, 37, 80
Cabriers, 22 Chlotar II, King, 57, 67, 74-77, 82-
Caesar, Julius, 14, 116 86, 89-90, 137
Cahors, 42, 85 Chlotar IV, King, 100
Cambrai, 3 Chora, 78
Camels, 58 Chramn, 29, 30, 34
Carcassonne, 17, 61-63, 71, 101 Chramnelen, Duke, 87, 95
152
INDEX
Chramnulf, magnate, 84 Ermanfred, Prankish magnate, 98,
Chrodoald, magnate, 93 109
Chunibert, Bishop of Cologne, 92 Erminarius, magnate, 84-85
Chunoald, Duke of Aquitaine, 106 Escaritos, 81
Ciuciolo, count of the palace, 48 Eudo, Duke of Aquitaine, 100-106
Claudius, magnate, 64, 89 Eunomius, Count, 50
Claudius, Visigothic commander, 63 Europenses, 103
Clovis, King, 3, 16, 18, 19, 31, 36, Eutropius, Gallo-Roman magnate, 16
88, 105: conversion to Christian- Exercitus, 81-83, 87-88, 97, 106-
ity, 6-7, 9; expands kingdom, 4-6, 110
9-13; fighting forces of, 13-15,
17, 124, 136; relations with Fara, magnate, 93
church, 7-8, 11 Feigned retreat, 28
Clovis II, King, 92, 94-95 Fideles, 31, 97
Cologne, 3, 67, 82 Firminus, Count, 28, 37-39, 70
Columbanus, 77-81, 90 Flaochad, mayor of the palace, 94-
Comes stabidi, 137 95
Compiegne, 77 Franks, 7, 13, 17, 27, 36, 113, 115,
Convenae, 58-60 118, 122, 124, 127, 131, 138
Coutance, 10 Fredegund, Queen, 42, 46, 47, 50,
Custodes, 50, 51, 62, 71, 78, 80, 88, 57, 63, 74
89, 96, 106 Frisia, 104
Frisians, 104, 108
Dagobert I, King, 84-86, 90, 92, 94,
96, 108-109, 112 Galactorius, Bishop of Beam, 7
Dagobert II, King, 109, 111 Gallien, 49, 52
Danes, 18, 20, 21, 34 Gallo-Romans, 3, 7, 15, 17, 36, 124
Desiderius, Duke, 47, 53, 57, 62, 63, Gap, 40
71 Gararic, Duke, 57
Die, 40 Garrisons: Angouleme, 12; Beau-
Dijon, 9, 29 caire, 62; Besancon, 78; Bordeaux,
Dio, 22 12; Caberet, 62; Carcassonne, 61,
Dodo, 99 71; Chastel-Marlhac, 21; Langres,
Domnicius, Gallo-Roman magnate, 34; Loches, 106; Saintes, 12; Sois-
133 sons, 45; Toulouse, 12; Tours, 51;
Dormelles, battle near, 76 Vienne, 24; Vollore, 20-21. See
Dragolen, Duke, 49, 52 also Antrustiones, Centenae, Cus-
Drogo, 98 todes, Mttites
Gascony, 53, 76, 85, 87, 103, 108
Eberulf, magnate, 64 Geneva, 67
Ebrachar, Duke, 63 Germania, 128
Ebroin, mayor of the palace, 95-98, Ghislemar, magnate, 98, 111
101, 109-111 Godegisel, Duke, 42
Ecdicius, Gallo-Roman magnate, 16, Godegisel, magnate, 65
23 Godigisel, Burgundian King, &
Edict of Paris, 84 Godinus, magnate, 44, 72
Egidius, Bishop of Rheims, 55-56 Godinus, son of Warnachar, 84
Einhard, 136 Godomar, Burgundian King, 22
Embrun, 40 Goths, 61, 62, 63, 68
Ennodius, Duke, 67 Gregory of Tours, 6, 11, 12, 18, 25-
Erchinoald, mayor of the palace, 94- 29, 36, 50, 135
95 Grimoald, son of Peppin I, 92-94, 97
153
Grimoald, son of Peppin II, 98 Levies: Alaman, 27, 81, 86, 88; Bur-
Gundobad, Burgundian King, 9 gundian, 61, 86, 87; Gascon, 99;
Gundobald, Duke, 42, 43, 46 general, 25, 60-61, 71, 107, 126;
Gundovald, the Pretender, 54-62, north of Seine, 61; north of Saone,
68,72 61; west of Rhone, 61; territorial
Gundulf, Duke, 55 of Champagne, 47, 60, 66, 68, 76,
Guntram, King, 29, 36, 48, 50, 67, 83, 125; trans-Rhenish, 35, 43, 46,
68, 72, 89, 97: and bella civilia, 55, 73, 76, 82-S3, 88, 93, 126-127
37-38, 41-42, 60; death of, 74; local (city) levies: Agen, 53, 63,
fighting forces of, 60, 62, 64, 66; 66, 68, 71, 76; Angers, 47, 52-53,
internal struggles of, 54-55; mili- 66, 76; Angouleme, 53, 61, 66, 76;
tary activity of, 39, 43, 47, 49, Auvergne, 37, 38, 43, 54, 61, 66,
53-54, 58-59, 61, 63-64 67, 70, 71, 76; Bayeux, 52, 66, 76;
Guntram Boso, Duke, 42, 46, 52, 54, Blois, 57, 66, 76; Bordeaux, 53,
55,67 63, 66, 68, 71, 76; Bourges, 55,
57, 60-61, 66, 71, 76; Chartres,
Hermanfrid, Thuringian King, 20 66, 76; Chateaudun, 42-43, 46,
Herpin, Count, 81 52, 66, 76; Le Mans, 66, 76;
Herpo, Duke, 48 Mainz, 93, 109, 125; Nantes, 53,
Hilarius, Bishop of Aries, 7 66, 76; Orleans, 57, 66, 76, 111;
Horse-breeding estates, 31, 78 Perigueux, 53, 61, 63, 66, 68, 71,
Horsemen, 11, 14-15, 19-20, 24, 76; Poitiers, 52, 53, 57, 60, 66, 71,
26-28, 58-60, 71, 111-112, 117- 76; Saintes, 53, 61, 66, 76;
122, 124, 126, 134-137 Saintois, 93, 109, 125; Toulouse,
Hostis Francorum, 110 53, 62, 63, 66, 68, 71, 76, 86;
Hostis VascanoTum, 111 Tours, 42-45, 47, 51-53, 57, 60,
66, 68, 76; Velay, 54, 66-67, 76
Imnachar, magnate, 28 Lex Salica, 14, 16, 137
Infantry, 27, 112, 114-115, 120-125 Liber Historiae Francorum, 8, 74, 7£
135, 138 Limoges, 28, 42, 47, 51, 57
Inferiores, 64, 71, 125, 126 Loches, 106
Innowales, Count, 93 Lodeve, 62
Isidore's continuator, 102-103 Lombards, 39^1, 55, 60-61, 86, 11*
Italy, 26, 27, 55, 60-61, 79, 86 Lupus, Duke, 47, 55
lussef ibn Abd ar Rahman, 105 Luxeuil, monastery at, 77-78
Lyons, 67
Justinian, Byzantine Emperor, 33
Maastricht, 67
Laeti, 5, 10, 14, 21, 25, 32, 33, 62, Mainz, 67. See also Levies
70-73, 78-SO, 124 Manaulf, Burgundian magnate, 95
Laffaux, battle at, 75
Landri, mayor of the palace, 74, 76, Mark theory, 120
77, 112 Marseilles, 54
Langres, 34, 79, 82 Martin, Duke, 97, 109
Le Mans, 66, 76. See also Levies Maurontus, magnate, 105
Leo, Byzantine Emperor, 16, 44 Meaux, 79
Lescar, 67 Medard, 51
Leudast, Count, 48-52, 79 Melun, 15, 50, 53
Leudegisel, Duke, 58-62, 71 Merovech, son of Chilperic, 46-49,
Leudes, 13, 30-31, 72, 77, 84-85, 52, 72, 137
89-92, 99, 104, 124 Merovech, son of Chlotar II, 76
Leudesius, magnate, 95—96, 111 Metz, 67, 82
Leutharius, Alaman Duke, 27, 89 Milan, 60
154
INDEX
Milites, 25, 33-34, 41, 50-51, 62, Quintianus, Bishop of Rodez, 7
71-73, 78-80, 88-89, 108, 124
Milituniae, 79 Rabanus Maurus, 136
Minor populus, 55-56, 71 Radbod, Duke of Frisia, 99-100
Mummolus, patrician, 38—41, 44, 47, Radulf, Duke of Thuringia, 92-94,
54-55, 59, 67 109
Munderic, Prankish magnate, 20-21, Ragamfred, mayor of the palace, 99-
29 101
Muslims, 101-102, 118 Ragamund, magnate, 78
Ragnachar, 4, 9, 13, 90
Naix, 82 Ragnovald, Duke, 53
Nantes, 11, 64, 78-79. See also Rauching, Duke, 65, 72
Levies Reccared, King of the Visigoths, 61-
Narbonne, 101, 105 62
Narses, 27 Reguli, 3, 6, 8, 10, 13, 24
Naval forces, 4, 18-20, 34, 38, 89, Remigius, Bishop of Rheims, 7—8, 14
108, 111, 128 Rennes, 10
Rheims, 5, 29, 36, 55, 75
Neustria, 66, 69, 74, 77, 83, 85, 97 Riccar, 13
Nevers, 67, 78 Rignomer, 13
Nicetius, Duke, 62, 67 Rigunth, 57
Nimes, 61, 62, 101 Ripuarian Franks, 137
Nobiliores, 87 Ripuarian Law, 115, 137
Notitia Dignttatum, 14, 21 Roccolen, Count, 46
Rodan, Lombard Duke, 40
Odilo, Bavarian Duke, 106-107 Rodez, 11-12
Olo, Duke, 60 Roman soldiers, 3, 10
Orange, 62 Romani, 79, 106, 111
Orleans, 10, 77-78. See also Levies Romania, 128
Ostrogoths, 12, 131 Rouen, 46
Otto, Prankish magnate, 92
Sagittarius, Bishop of Gap, 60
Paris, 4-5, 17, 29, 51, 53, 75, 77-79, St. Boniface, 98
127 Saint-More, 5
Pauperes, 64, 71, 125-126 Saintes, 12, 47, 85. See also Levies
Peppin I, 92 Saintois, 93, 109, 125. See also Levies
Peppin II, 97-98, 109-110 Salmaise, 34
Peppin III, 101, 108-111, 122, 126 Samh, -al, 101
Perigueux, 53, 57, 85. See also Levies Samo, 86
Pisaurum, battle of, 27 Saragossa, 26, 86
Poitiers, 11, 28, 38-39, 42, 49-51, Sarmatians, 5, 12, 17, 34
57-58, 67, 112: battle of, 117; Satellites, 99, 104
Church of St. Hilary at, 49-50, Saxons, 15, 28-29, 32, 36, 41, 82-
101. See also Levies 87, 101, 105, 108, 124, 127: of
Potentiates, 126 Bayeux, 10, 15, 52, 63, 71; in
Praetextatus, Bishop of Rouen, 48 Italy, 39
Procopius, 26, 79, 113-114: on Scapthar, magnate, 28
Prankish warfare, 131-136; eval- Scara, 81, 87-88, 109
uation of, 137-138 Scarfa, 78
Protadius, mayor of the palace, 77 Seltz, 81
Pueri, 32, 38, 46, 50-52, 57, 64-65, Seniores, 87
72-73, 85, 97, 99, 124 Senlis, 5
Pussy, 79 Septimania, 12, 61, 63, 68, 71
155
Sermaise, 5 Thaumastus, Gallo-Roman magnate,
Sermiers, 5 23
Sermoise, 5 Theodore, Bishop of Marseilles, 54
Sicambri, 14, 116, 136 Theodoric the Great, 10
Sidonius Apollinaris, Gallo-Roman Therouanne, 48-49
magnate, 10, 23, 133 Theudebald, Alaman Duke, 107-108,
Siege engines, 9, 17, 59, 60 111
Siege of: Angouleme, 17; Aries, 12, Theudebald I, King, 26, 35, 88-89
17; Avignon, 9, 17, 55, 105, 111, Theudebert, son of Chilperic, 42, 45
127; Bellinzona, 60-61; Bourges, Theudebert I, King, 18-19, 22, 25-
53; Carcassonne, 17, 101; Chalons, 27, 30-35, 85, 88, 133
29; Chastel-Marlhac, 20; Cler- Theudebert II, King, 75-84, 90
mont, 37-38; Narbonne, 101, 105; Theuderic I, King, 11, 18, 21-22,
Nimes, 101; Paris, 4, 17, 129; 31-33, 44, 79, 85, 112, 136
Saragossa, 26; Toulouse, 101; Ver- Theuderic II, King, 75-84, 89-90
dun, 4, 17; Vitry, 21; Vollore, 20- Theuderic IV, King, 96, 111
21; Wogastisburg, 90 Theudoald, Count, 79
Siggo, magnate, 47 Theudoald, mayor of the palace, 99
Sighar, magnate, 38 Thuringians, 3, 5, 19, 31-32, 75, 82-
Sigibert, regulus, 12 83, 135
Sigibert I, King, 46, 47, 48, 50, 82: Tiberius, Byzantine Emperor, 54
and bella civilia, 36-39, 41-42; Titus, Gallo-Roman magnate, 16, 44
death of, 43, 45, 56; fighting Tolbiac, battle of, 6, 8
forces of, 43, 45, 88; problems Toul, 67: battle of, 82
with magnates, 44, 65, 72 Toulouse, 4, 12, 57, 61, 63, 85, 101.
Sigibert II, King, 92-93, 97 See also Levies
Sigila, magnate, 43 Tournai, 3-4, 14
Sigismir, Burgundian prince, 24 Tours, 38-39, 48-52, 67, 80: Church
Sigivald, Frankish magnate, 21, 30— of Saint Martin at, 64-68. See also
31,44 Levies
Sigulf, magnate, 42-43, 46, 48 Tribuni, 34, 51, 78-80
Sisenand, King of the Visigoths, 86 Tntstis, 13, 124
Socii, 24, 52, 55, 95, 97, 99, 105,
109-110 Ukba ibn al-Hadjdjadj, 105
Sodales, 97, 99, 110 Uro, 92
Soissons, 36, 45, 46-47, 74-75: arms Ursio, Frankish magnate, 55, 65
factory at, 4; kingdom of Romans Uzes, 67
at, 3, 5
Spain, 54, 57, 86 Valence, 40, 67
Spire, 67 Valentinus, Gallo-Roman magnate,
Sportulae praetoriani, 23 34
Stipendia paludatis, 23-24 Vassals, 120
Strasbourg, 67 Velay, 54, 66-67, 76
Sulpicius Alexander, 134 Venerandus, Duke, 86
Superiores, 74 Verdun, 4, 17, 67
Syagrius, 3-4, 8, 30 Verus, Bishop of Tours, 7
Syagrius, Gallo-Roman magnate, 30- Vienne, 9, 24
31 Viri fortes, 52
Syrivald, Frankish magnate, 30 Visigoths, 3, 7, 10-12, 16, 23, 103,
136
Tacitus, 116 Vitry, 21
Taifals, 12, 17, 29, 33, 36, 44, 124 Vollore, 20-21
Technical determinism, 119 Volusianus, Bishop of Tours, 7
156
INDEX
Vouille, battle of, 11 Willebad, patrician, 85, 87, 94-95,
109
Waldebert, 84 Wintrio, Duke, 60, 74-75, 89
Wandelbert, magnate, 95 Wiomad, 4
Waratto, mayor of the palace, 98 Wogastisburg, 86, 90
Warnachar, mayor of the palace, 83, Worms, 67
84 Wulfoald, Duke, 95, 97
Waroch, Breton count, 53
Wends, 87, 108, 127 Zaban, Lombard Duke, 40
Wilfrith, Bishop of York, 96, 109 Zulpich, 67: battle at, 82
157

Bernard Bachrach - Merovingian Military Organization

Uploaded by

Copyright:

Available Formats

Bernard Bachrach - Merovingian Military Organization

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bernard Bachrach - Merovingian Military Organization

Uploaded by

Copyright:

Available Formats

Merovingian

UNIVERSITY OF MINNESOTA PRESS

Library of Congress Catalog Card Number: 70-187164

T his monograph is a study of the military institutions which

A FTER the death of Aetius in 454, imperial power in Gaul

Merovingian military began with Clevis's warband and grew with

It is not known precisely at what date Clovis finally secured

Whether or not Clovis had made overtures to the bishops as a

vingian control, and Gregory of Tours somewhat optimistically

In the decade after his baptism, Clovis vastly increased the

equestrian nature of Prankish arms. The Roman empire had

The remains of the formal units of the Roman military also

extremely limited, it does seem reasonable to conclude that

Although troops other than Franks probably took part in

dia paludatis were probably former Roman soldiers who con-

to be cut down and barricades to be built; this made his position

Munderic's group, on the other hand, was not organized under

was supplemented by the use of Roman fortifications through-

was not as fortunate as he had been in 562. According to

to swim across the Rhone. A large number of Arvernians were

Mummolus was able to move west, driving Chlodovech from

kingdom, Mummolus barred their way and forced them to pay

tory, as well as further evidence of Roman influence on these

long after Sigulf, one of Sigibert's supporters, drove him out.

raised to the kingship of Chilperic's lands by the latter's erst-

under a duke and integrated into the Merovingian military

activity in Gaul. By early 576, Chilperic had taken Paris, cap-

Brunhild had escaped to Austrasia. By this time (it was late

to Childebert. In this crisis in allegiance there was a lack of

when he arrived at the town of Therouanne with a chosen group

Dragolen among them—had bands of private armed retainers

order his troops besieging Bourges to withdraw. For Guntram

Chilperic. The minor populus attacked Egidius, who barely es-

when Guntram learned of Gundovald's activities in the south, he

Bishop Sagittarius were killed. Some of the other magnates, how-

MILITARY COMPLEXITY: 561-593

In negative terms, it has been ascertained that the personnel of

alone a part of the levy of Auvergne. There are other instances

The Last of the

As a general stratagem for the battle, the queen suggested a

Baudulf, one of Theuderic's magnates, and was taken to Besan-

Upon reaching Nantes, Ragamund handed his prisoner over

The Merovingian interest in legislating for these people sug-

the pagus of Avenches is said to have included many counts

armies would seem to exclude Theudebert's local levies from

queen and her great-grandsons and handed them over to Chlotar

against Merovingian domination. When Dagobert realized that

sum regnum Austrasiorum, and universi leudes in Auster in exer-

The Rois Faineants

N OT LONG after the death of King Dagobert in 638, his son

general levy of the trans-Rhenish peoples living under Austrasian

In the struggle for power in the first few years following

magnates and their followers. Willebad, who realized he would

For example, we have seen that as early as 639 Sigibert II had

leaving. A Frank named Ermanfred and his socii then murdered

palace to King Childebert, Peppin went to war against the Fri-

forced him to retreat. Shortly afterward, Ragamfred plundered

Charles, who, in turn, chased the Aquitanians south of the Loire.

terize the Northerners' posture in face of the enemy. By dwelling

ceeded to occupy Eudo's lands, including the cities and fortresses