"Nostre Français n'unt talent de fuïr": The Song of Roland and the Enculturation of a Warrior Class

Download 309.68 Kb.
Size309.68 Kb.
  1   2   3

John F. Benton

"Nostre Français n'unt talent de fuïr":
The Song of Roland and the Enculturation of a Warrior Class


uring the german siege of paris in december 1870, a learned and
patriotic medievalist, Gaston Paris, delivered a set of lectures at the
Collège de France on La Chanson de Roland et la nationalité'française.1 It
would now be timely for a specialist in contemporary history and literature
to prepare another study on the Song of Roland and modern nationalism,
particularly in the period of World War I. Influential historians have
blamed the newspapers and the popular press for inflaming public opinion
on the eve of the Great War.2 That "yellow journalism" helped to indoctri-
nate the masses who marched enthusiastically to war cannot be doubted,
but scholars and professors also played their part in the movement, and
while the press harangued the future foot soldiers, the academic elite was
addressing the officer class. Every poilu knew about Joan of Arc, but the
officers had also learned in their lycées of the valor of the heroes of
Roncevaux. While the greater part of this paper is devoted to the social and
political values conveyed by the Chanson in the Middle Ages, the use of
literature to buttress values can conveniently be illustrated by some refer-
ence to the Song in more modern times.

A dozen French translations of the Roland appeared between 1870 and

1914 and in 1880 it was assigned officially as a "texte classique à l'usage
des élèves de seconde."3 In 1900 a professor at the Lycée Henri IV told his
audience at the École Spéciale Militaire at Saint-Cyr, "La Chanson de
Roland est notre Iliade" and concluded, "Elle n'est pas seulement un sujet
d'étude pour nos esprits: c'est une des sources vives où nous devons
retremper nos âmes."4 During the summer of 1918 a professor at the École
Normale of Fontenay-aux-Roses urged future teachers, when reading to
future Rolands and Olivers from the Chanson, to show "le lien toujours
vivace qui joint au passé le présent" and the ideals inherited from "nos
aïeux du Moyen Age," including "ardent amour de la patrie," "culte
souverain de l'honneur," and "crainte de forfaire et d'être honni."5

Joseph Bédier's love of medieval France and contempt for German cul-

ture, expressed in the great Légendes e'piques he first published between
1908 and 1913, was shared widely by and with his countrymen.6 Five days
after Germany declared war on France, a friend of Charles Péguy, editorial-
izing in a Parisian Catholic daily, offered his readers two inspirational (if


238 Olifant/Vol 6, Nos. 3 & 4/Spring & Summer 1979

technically incongruous) quotations, "Finis Germaniae" and the invoca-

tion of the Carolingian Salic (and therefore, Germanic) Law, "Vive le Christ
qui aime les Francs."7 Battle strategies for the war were created by military
theorists like Ardent du Picq and the enthusiastic Colonel Grandmaison,
rather than medievalists and their sympathizers, but all shared a common
sense of national heritage and spirit. The Field Regulations of November
1913 declared: "L'armée française, revenue à ses traditions, n'admet plus,
dans la conduite des opérations, d'autre loi que l'offensive," and stated
formally, "Les batailles sont surtout des luttes morales."8 On the battle-
ground of morale, the Chanson de Roland could serve as a weapon.

As preparations for war against Germany developed, the Song leaped the

Channel and two translations appeared in England in 1907, just three years
after the Anglo-French Entente. Scott Moncreiff began his own translation
as a "solace" in the summer of 1918, and John Masefield in 1918
introduced each chapter of his apologetic Gallipoli with a hortatory
passage from the Song of Roland.9 The precise links connecting literature,
ideals and actions are most uncertain, for it is so difficult to distinguish the
determinants of behavior from their after-the-fact justification. Neverthe-
less, it seems clear to me that the Chanson played a part in the mentality of
the Great War. The use of the Roland in modern times, either for inspiration
or for solace, may be considered a part of the process of enculturation, a
process which is probably harder to understand but easier to recognize in
earlier or non-Western cultures than in our own.

"Enculturation" is a term recently created by anthropologists as an

alternative to "socialization" to distinguish different aspects of the educa-
tional process and its relationship to cultural change. Specialists differ
over the distinctions between these terms and their specific meaning, and I
will not insist on a matter of definition here. But since "socialization" may
lead one to think of children learning (however well) to listen respectfully
to their elders or of a page being taught not to pick his teeth in public, I have
preferred to use "enculturation" here, defined as "the process of acquiring
a world view."10

Great poetry both gives pleasure and teaches. The epic transmits infor-

mation about the heroic past, and in either its oral form or in successive
reworked texts, this information can change with circumstances. In non-
literate societies, we are told by two field anthropologists, "what continues
to be social relevance is stored in the memory while the rest is usually
forgotten," and in proof of this point they note how the Tiv people of

Benton/The Song of Roland and the Enculturation of a Warrior Class 239

Nigeria changed their "traditional" genealogies over forty years. As a

second example they cite the case of the state of Gonja in northern Ghana,
which was divided into seven divisional chiefdoms in the early part of this
century, at which time the local myths indicated that the founder of the
state, Jakpa, had seven sons; sixty years later two of the divisional chief-
doms had disappeared and in the collective memory of the people Jakpa
was said to have had only five sons. 11We should expect analogous changes
in an evolving story like that of Roland.

Information such as that just mentioned about genealogy and lineage is

useful to a society in understanding its past or present political organiza-
tion, but the transmission of techniques seems to play a very minor part in
epic poetry. A young warrior would never learn how to fight in battle
formation from hearing the Song of Roland.12 He would not even learn how
to use his sword. The "fragment" from The Hague, dated about the year
1000, describes an "epic stroke" which splits the middle of the opponent's
head and body and even cleaves the spine of his horse.13 This overhand
stroke is used repeatedly in the Chanson de Roland (except that in the
Baligant episode Charlemagne splits only the emir's head), the Bayeux
tapestry depicts the beginning swing of such a stroke, and its consequences
can be seen in numerous medieval illustrations.14 Now if one reflects on
this stroke, it is better suited to legendary heroes than to real-life survivors.
If a warrior raises his arm like a tennis player about to serve, he exposes the
vulnerable area of the armpit, loses the ability to parry all but a similar
stroke, and gains nothing from the forward movement of the horse. Accord-
ing to those of my students who have fought with heavy swords on foot (I
have not), a sweeping side-stroke is more powerful than an overhead
smash, because it can be delivered with the torque of the whole body. And
for a mounted warrior, a thrust is preferable to a cut, and the "epic stroke" is
particularly dangerous, because if the opponent veers, the stroke would
then descend on the head of the rider's own horse. In short, the Chanson de
Roland is not a manual of practical use for either a medieval warrior or a
modern historian. It teaches not skills but values or morale. This is surely
what the author of a thirteenth-century sermon had in mind when he wrote
of the use of the deeds of Charles, Roland, and Oliver "to give spirit to the

The complex problems of dating, of different chronological "layers" in

the Oxford text, and of different versions of the Chanson which were
produced and written after "Turoldus" had completed his epic, are
troublesome issues for any commentator. At this point I should say that I

240 Olifant/Vol. 6, Nos. 3 & 4/Spring & Summer 1979

find compelling the major arguments of "traditionalism" that the Oxford
text does contain much material which entered the collective or poetic
memory in earlier centuries, such as the "epic stroke" or the episode of
Charlemagne, like Joshua, making the sun stand still outside Zaragoza as
reported in the Annales Anianenses.16 Because it deals with traditional
material, the poem of "Turoldus" is notoriously hard to date precisely. The
reference to Saracen battle drums in the poem helps to place its composi-
tion after 1086,17 but beyond that point controversy rages. Some paleo-
graphers have dated the Oxford manuscript as late as ca. 1170, allowing
other critics to place its composition in the 1150's. Others consider that the
manuscript could have been written as early as 1125 and not later than
1150, thus ruling out a mid-twelfth-century composition and placing the
text either shortly before the First Crusade or within a generation or so
after. 18 For the purposes of this paper, dealing as it does with enculturative
values, the precision of dating is not of great importance, so long as one
accepts the principle that some portions of the epic entered the Chanson, in
either oral or written form, well before the twelfth century, but that in a
fluid tradition early material which does not have "social relevance,"
which goes against the cultural values of a later time, will tend to drop from
sight. What is certain is that even if it were first written in the late eleventh
century, the Chanson de Roland as we have it is culturally a twelfth-
century poem, copied in the twelfth century, cited by twelfth-century
authors, and popular enough to give birth to translations and other ver-
sions in the twelfth and thirteenth centuries. Moreover, a host of critical
studies shows that the poem of "Turoldus" is no mere pastiche of frag-
ments from different epochs, but was intended by its author to have a
unified form and meaning.19

The remainder of this paper is concerned with some of the ways in which

the Chanson might have inspired or justified the behavior of its twelfth-
century audience. Among the enculturative values which the Song of
Roland displays, the most prominent is the glorification of warfare, a "just
war," a "Holy War," to be sure, but warfare all the same. I doubt if many, if
indeed any, bellatores of the period when the Chanson was still a living
epic needed to be reassured that warfare was a proper and honorable
occupation. But when one looks outside the Chanson and the values of the
warrior class, one can see that they are in contradiction to the lingering
remnants of traditional Christian pacificism, a theme of great importance
in many of the Fathers and clearly expressed by the quotation in Sulpicius
Severus's popular vita of Saint Martin of Tours, "I am a knight (miles) of
Christ, I am not permitted to fight."20 Whether or not a Song of Roland was

Benton/The Song of Roland and the Enculturation of a Warrior Class 241

chanted by a Taillefer to inspire the Norman army before the battle of
Hastings, the twelfth-century authors William of Malmesbury and Wace
considered the Chanson appropriate for such an occasion.21 While the
warriors of William the Conqueror or of the succeeding period can scarcely
have been expected to have puzzled over Tertullian and Basil, and presum-
ably gave no thought to the fact that Saint Martin deserted from the Roman
army, not all bishops were warriors like the poetic Turpin or the historical
Odo of Bayeux, who as members of the clergy were forbidden by canon law
to bear arms.22

Attitudes toward warfare varied, of course, and I am not sure that the

Gregorian Reform itself marks a great turning point on the issue of Chris-
tian pacifism. In the early eleventh century, Bishop Hubert of Angers was
excommunicated for fighting at his king's command, while Bishop Wazo
of Liège did lead troops in battle, but conscientiously did so unarmed.23
When critics, with their own ideas about the nature of Christianity (medie-
val or modern), call the Chanson a devoutly Christian epic, a Vita or Passio
sancti Rolandi, we need to remember that Archbishop Turpin provided an
uncanonical model for any clergy who heard the poem.

In what I have just said about the positive value placed on warfare I noted

that the war commemorated in Roland, unlike those in the more common
epics of revolt such as Raoul de Cambrai, was a "Holy War" against the
infidel. Practically all medievalists agree that there is a relationship be-
tween the Chanson and the development of the idea of crusade, the
Entstehung des Kreuzzugsgedankens, though the chronology of that rela-
tionship has been hotly disputed. As Carl Erdmann has put it, "Some say
that 'the Chanson de Roland would be impossible without the First Cru-
sade,' while others maintain that 'the crusade would be incomprehensible
without the Chanson de Roland.'"25 Erdmann did not attempt to date the
poem, except to state that "the Chanson cannot antedate the time of Alex-
ander II."26 The theme of a bellum domini fits both the actual historical
circumstances of the invasion of Spain and the period after the renewed
religious expansionism of Christian Europe in the 1060's, that is, about the
time of the writing of the Nota Emilianense, and so is of little help in dating
either the Oxford Roland or its predecessors. While I feel that the Chanson
accords well with a military-religious ethos common in the late eleventh
and early twelfth centuries, what strikes me most is how difficult it is to
find clearly demonstrable echoes of the crusade to recover Jersualem in a
text which took its present written form well after the First Crusade.27
Equally, it is surprising how seldom the name of Roland or reference to the

242 Olifant/Vol. 6, Nos. 3 & 4/Spring & Summer 1979

Chanson appears in the extensive literature written in support of the
crusading movement.28 Even in the early fourteenth century, when Pierre
de la Palu turned to a literary source for his treatment of Charlemagne in his
Liber Bellorum Domini, he used the story of Amis and Amiloun.29 The
Chanson de Roland must have been immensely popular in the twelfth and
later centuries (as manuscripts, translations, and onomastics all attest), its
ethos does support the militant expansion of Christianity, but the vigor of
scholarly debate over precise dating suggests that it could have been
composed before 1095, and the relative silence of crusading sources and
propaganda with respect to Roland or the epic Charlemagne indicates that
contemporaries could have understood, described, and advocated
crusades quite well if the Chanson de Roland had never existed.

Besides the positive value placed on fighting itself and on militant

Christianity, a third obvious but easily misunderstood cultural value
transmitted by the Chanson is loyalty to king and country.30 This point too
has been amply treated in our literature on the Chanson, and in this brief
discussion I wish particularly to express a caveat about an apparent
"French" or "Capetian" nationalism. The "patriotic" loyalties which I
think the Oxford Roland exemplified for its listeners were loyalty to the
warrior's highest recognized ruler, the ruler who led one's army into battle,
a loyalty which we may call "feudal," whether its object was an emperor,
king, duke, or count, and willingness to fight and die for one's homeland,
for Tere Majur, the familiar theme of pro patria mori, whatever that home-
land or pays actually was. In addition, the Chanson also stresses imperial
authority in terms which seem particularly appropriate to the Anglo-Nor-
man "empire," in part because William the Conqueror ruled to some
degree in a Carolingian mode.31 It may at first seem perverse not to empha-
size the importance of dulce France and not to argue that the author of the
Oxford Roland was trying to inculcate loyalty to the king of France (any
king, from Henry I to Louis VII, depending upon the date assigned to the
text). But let us consider the historical origins and context of the poem a bit
more fully.

Whatever the means of transmission, the Song of Roland had to have its

origin in the battle, the twelve-hundredth anniversary of which was
commemorated in 1978. And when that event was celebrated in song, the
ruler to whom highest honor was given had to be Charlemagne, and his
warriors necessarily Franks. Charlemagne, the unifier of Christian Europe,
became in legend and belief a universal hero, a Christian Alexander, but
once the Capetians had overthrown the Carolingians, it took a long time for

Benton/The Song of Roland and the Enculturation of a Warrior Class 243

the Capetian monarchy to develop a special affinity for him. His canoniza-

tion in 1165 was, after all, the work of an anti-pope opposed by the French
monarchy and acting at the instigation of the German emperor. A prophecy
created at Saint-Valéry-sur-Somme about 1040 predicted that the Cape-
tians would hold the throne of France for seven generations, when it would
return to a descendant of Charlemagne. Early in the twelfth century Hugh
of Fleury and Sigebert of Gembloux both stressed the usurpation of the
Capetians in their influential histories. Only at the end of the twelfth
century did the reditus ad stirpem Caroli become a literary theme centering
on Philip Augustus.32

One of the literary merits of the Roland is that Charlemagne is a noble

figure throughout, ready to become the saint placed in Heaven by Dante,
and is not insulted as he is in many other chansons, though his compara-
tive impotence at the trial of Ganelon may reflect this tradition.33 It has
been suggested that the Chanson de Roland was commissioned by Suger or
one of his successors to strengthen the prestige of the Capetian monarchy,
but this idea, even if it does not fall on chronological grounds, finds
precious little support in the writings of Suger himself. In his student days
a distinguished French medievalist set out to write a diplôme on "L'idée de
Charlemagne dans la pensée de Suger," and abandoned the project when
he found, as can be seen by anyone who consults the index of the Oeuvres
completes de Suger, that the great propagandist of Saint-Denis scarcely
mentioned Charlemagne and when he did treated him as simply one more
king who had the good sense to make donations to the abbey. The argu-
ments of Professor Hans-Erich Keller, which derive a good deal of rhetor-
ical force from repetition and the cumulation of philological detail, all too
often must depend on the concept of mystification. "Mystification" can be
useful for a writer of religious allegory or a humorous author, like Geoffrey
of Monmouth, who had to deal with split allegiances in a turbulent politi-
cal setting, but it is a senseless technique for a propagandist trying to
strengthen royal power, when direct writing would be so much more

The reputation of Charlemagne, as far as we can tell, never died. As the

Nota Emilianense suggests, a cantar de Rodlane about Charlemagne and
his peers may have existed in Castile in the eleventh century.35 Scholars
have placed the origins of the Chanson de Roland in areas of France as
widely separated as the Midi and Brittany.36 The existence of songs about
Charlemagne in his native Germanic tongue has been postulated, though
textual evidence for an early Song of Roland in German is as lacking as

244 Olifant/Vol. 6, Nos. 3 & 4/Spring & Summer 1979

evidence for a Chanson in Francien dialect. For an historian unable to form
an independent judgment about the philological discussions, the argu-
ments about the "national origin" of the Chanson seem analogous to those
about the origin of Honorius Augustodunensis, who was born in one place
and traveled a lot.37 What is clear from translations and manuscripts is that
after the beginning of the twelfth century the Chanson de Roland could
find a welcome home in France, England, Germany, Norway, Italy, Spain,
and eventually Brazil.

In short, one did not have to be French or a subject of the Capetian

monarchy to enjoy and be inspired by the Song of Roland. The Anglo-
Normanisms of the Oxford Roland, combined with other evidence, includ-
ing the name Turoldus, have suggested to some a Norman or Anglo-Nor-
man origin.38 A French critic has recently written that if the author of the
Oxford Roland was Norman or of Norman origin, he did not reveal "de
véritables partis pris normands," and that "son idéal s'élève pourtant
au-dessus de tous les particularismes, et apparaît déjà vraiment
national."39 I would prefer either to consider the ideal of the Chanson
supranational or to say that the term nationalism has little or no meaning in
the period we are considering. In The Battle of Maldon Byrhtnoth, who in
dying for his king and homeland exhibits much the same fierce pride and
loyalty as Roland, has been said to show "un haut sentiment national."40
Perhaps so, but it is probably to the England of Byrhtnoth's king, Ethelred
II, that we owe our single medieval manuscript of Beowulf, a manuscript
which shows that the glory of the Danish court stayed alive in an Anglo-
Saxon kingdom which had suffered grievously from Danish attacks.41 If we
move forward to the twelfth century we find that the French welcomed
Arthur and the "matter of Britain" without regard for his national origins.
In the medieval world we are considering, loyalty and love of homeland
were admired and "nationalism" was not understood in the sense that it is
today. With equal justice it could be said that if the author of the text from
which the Oxford Roland was copied was French, he did not reveal "any

Share with your friends:
  1   2   3

The database is protected by copyright ©essaydocs.org 2020
send message

    Main page