JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 79
a
“Pagans are wrong and Christians
are right”: Alterity, Gender, and
Nation in the Chanson de Roland
Sharon Kinoshita
University of California, Santa Cruz
Santa Cruz, California
To be a medievalist, Bernard Cerquiglini has written, is to take a stand on
the Chanson de Roland.1 In the song’s long critical history, this has meant
taking sides on questions such as Roland’s heroism or his démesure, or on the
poem’s composition by a poet of genius or a singer of tales. Two things, at
least, seemed beyond debate: first, that the poem casts the Saracens as a
fierce and intractable Other, as epitomized in Roland’s unforgettable rallying cry, “Paien unt tort e crestiens unt dreit” [Pagans are wrong and Christians are right] (1015); and second, that women have little place in this stark
celebration of military valor, as evidenced in the scant thirty lines devoted to
the death of Roland’s fiancée Aude.2
Recent metahistories of our discipline have uncovered how strongly
the canonization of the Roland as the preeminent French epic is linked to
the historical exigencies of the late nineteenth century. Prized as a precocious assertion of French national sentiment, it performed important cultural work as the Third Republic strove to overcome the humiliation of the
Franco-Prussian War. At this foundational moment in the history of medieval French studies, epic was thus defined as the perfect expression of a feudal, Christian, nationalist ethos in which women had no part. In this essay
I revisit this gendered construction of difference to suggest that the representational regime of the Roland is far less secure than it initially appears. In
part one, I suggest that the poem is haunted by a crisis of nondifferentiation
strongly at odds with the monological fixity usually ascribed to it. Disengaging its representation of the pagans from modern racialist and Orientalist paradigms reveals how strongly it exemplifies a concept of alterity
markedly different from our own. The stark simplicity of Roland’s pronouncement “Paien unt tort e crestiens unt dreit” conceals the instability
unsettling each side of the confessional divide. The dualism it implies is
Journal of Medieval and Early Modern Studies 31:1, Winter 2001.
Copyright © by Duke University Press / 2001 / $2.00.
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 80
belied first by the parallelism that constructs the pagans as mirror images of
the Christians, and second by the possibility of their conversion, which gestures toward the collapse of all difference. On the Christian side, the unity
of Charlemagne’s empire is disrupted by the historical layering linked to
the process Robert Bartlett has memorably called the “europeanization of
Europe” that resulted in, among other things, the emergence of “Frank” as a
new collective identity.
In part two, I suggest that the Roland manages this instability
through its strategic deployment of gender. Far from being marginal to this
feudal Christian epic, Aude and Bramimonde are central to its construction
of difference. Unsettled by the indeterminacy of the boundary between
pagans and Christians, the Roland reinstates difference through the stock
epic figure of the Saracen queen. As one of only two women in the text,
Bramimonde is situated at the intersection of the medieval problematics of
gender and culture. As the poem’s lone Saracen woman, her brazenness
secures the boundary between pagans and Christians even as it signals her
availability for conversion. Part three turns to the Roland ’s conclusion.
Brought from Saragossa to Aix, the heart of Charlemagne’s empire, the
Saracen queen is contrasted with Aude, the text’s only other woman. Taciturn where Bramimonde is vociferous, intransigent where the pagan queen
is mutable, Aude exemplifies the fixity of a feudal-Christian order heretofore threatened by instability and fragmentation. Her death and Bramimonde’s conversion are not merely codas to the drama at Roncevaux; they
instantiate and guarantee the truth of Roland’s reassuringly simple pronouncement, “Paien unt tort e crestiens unt dreit.”
I
The critical investment in the Chanson de Roland as an epic struggle between
Christians and Saracens dates from the Franco-Prussian War.3 In December
1870, the third month of the German siege of Paris, the celebrated medievalist Gaston Paris delivered a lecture at the Collège de France entitled “La
Chanson de Roland et la nationalité française.” At this moment of national
humiliation, the poem’s incantatory evocation of “sweet France” (“douce
France”) and its glorification of heroism and sacrifice provided historical
roots for the French sense of identity.4 The immediate target of this nationalistic fervor was of course the emergent German state. Not coincidentally,
the 1870s were also the decade of colonial expansion in North Africa. In
1870 France formally annexed Algeria, which had been under French mili80 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 81
tary rule since the conquest of Algiers in 1830.5 In 1871 tribal lands were
seized and redistributed to European settlers, including refugees from Alsace
and Lorraine, recently lost to the new German Empire.6 The resulting colonial regime served the interests of this new settler community, establishing an
educational system “designed to submerge the Arab-Islamic identity.”7 Under
the Third Republic, French writers and ideologues were formulating a
“coherent, systematic, and relatively original doctrine of . . . colonial imperialism” that emphasized the mission civilisatrice and vaunted the improvements in the material and moral order of North Africa since its conquest.8
Meanwhile, following Gaston Paris’s speech, the Chanson de Roland had
been institutionalized as France’s national epic, incorporated into the agrégation in 1877 and the standard secondary school curriculum in 1880.9 As the
French found compensation for their loss of Alsace-Lorraine in Muslim
Algeria, the Roland was becoming a lieu de mémoire, a memory site for the
accumulation of national unity, solidarity, patriotism, and cultural heritage.10
In Orientalism, Edward Said describes how the discourse of European expansion rationalized and naturalized the colonial venture by assuming a radical difference between westerners and the people they were
colonizing. “Europe” or “the West” was constructed as modern, rational,
vigorous, and enlightened, in opposition to an Oriental “Other” represented as static, decadent, and despotic.11 Yet even at the height of the colonial era, as has recently been argued, dichotomies between colonizer and
colonized “took hard work to sustain.” The “otherness of colonized persons
was neither inherent nor stable; his or her difference had to be defined and
maintained.” If this was the case in an age when colonial bureaucracies provided the means and scientific racism lent the justification for Europe’s
domination of non-Western cultures, it is all the more true in the Middle
Ages, when racial thinking, with its claims to biological essentialism, was
not yet “an organizing principle and a powerful rhetorical theme.”12
In the Middle Ages, Bartlett has argued, the privileged form of difference was not race but custom: peoples were distinguished not by biological essence but by language, law, “dress, domestic rituals, dietary habits,
hair-styles and a host of other habitual practices.” The key feature of this
cultural concept of difference is that, unlike race, such attributes could be
changed “not only from one generation to the next, but even within an individual lifetime.”13 For Bartlett, the history of Europe between the mid-tenth
and mid-fourteenth centuries is the history of just such transformations.
Reconceptualizing “Europe” not as a geographical entity given in advance
but as a culture of shared practices overlapping with but not identical to
Kinoshita / Alterity, Gender, Nation in the Roland 81
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 82
Latin Christendom, he describes its development in the heartland of the
former Carolingian Empire and its subsequent spread through conquest,
colonization, and acculturation.
This is not to say that the Franks did not notice physiognomy or
skin color, and indeed at first glance many of the Roland ’s gent paienur
seem to belong to a long tradition of demonizing the foreign other. With
skin as hard as iron and bodies covered by porcine bristles, peoples like the
Occians and the Micenes recall the monstrous figures—the Blemmyae, the
Cynocephali, the Sciopods—that, in the account of Pliny and his followers,
populate far-flung lands.14 Sporting a huge forehead and a half-foot gap
between his eyes, Marsile’s own brother Falsaron borders on the monstrous.
Though some of these images are easy to dismiss as products of medieval
ignorance, prejudice, and superstition, others are susceptible to being gathered up by later discourses of European racism. Especially troubling are certain connotations associated with blackness and Africa. Ethiopia, ruled by
Marsile’s uncle Marganice, is a cursed land (“tere maldite” [1916 ]) populated by a black race with large noses and wide ears (“La neire gent en ad en
sa baillie; / Granz unt les nés e lees les oreilles” [1917 –18]). Such baleful
images are reinforced by proper names that impressionistically imply “a
European fear of dark skin color, associated with evil and social worthlessness.” In Munigre, whose name encodes blackness, “the sun never shines;
wheat does not grow; rain never falls; and all rocks are black. It is a land of
demons” [Soleill n’i luist ne blet n’i poet pas creistre, / Pluie n’i chet, rusee
n’i adeiset, / Piere n’i ad que tute ne seit neire: / Dient alquanz que diables i
meignent] (980 – 83). Other names, like Malquiant, “an African from
Africa, son of Malcud” [D’Affrike . . . un Affrican . . . Malquiant, le filz al
rei Malcud] (1593 – 94 ) are built from prefixes “connoting evil and error.”
It is not surprising, therefore, that Paula Gilbert Lewis observes in the
Roland that its representations of Africans have much in common with the
“naive but dangerous” discourse of modern racism.15
At the same time, it would be misleading to take this representation
of black Africans as exemplary of the medieval attitude toward the other.16
In the Roland, the primary categories of difference are Saracen and pagan,
forms of which occur over thirty and one hundred times, respectively.17 But
though phrases like “la gent paienur” (1019) may seem to hint at a sense of
biological difference, the referent of such terms is notoriously unstable. In
most cases they correspond to the Muslims — Turks, Arabs, and Berbers —
whom eleventh- and twelfth-century crusaders were battling in the Middle
East, Spain, and southern Italy. Occasionally, however, they refer to popula82 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 83
tions like the Saxons, Slavs, and Hungarians— peoples on the northern and
eastern fringes of Latin Europe—Christianized after the time of the historical Charlemagne but before the composition of extant chansons de geste. In
the Roland, pagandom encompasses a vast array of physical or ethnic types:
the bristle-backed Micenes, the seductively handsome Margariz of Seville,
Marsile’s son Jurfaleu the Blond (1904), and his nephew, who bears the distinctly Germanic name Aëlroth (1188). What these disparate figures have
in common is their allegiance to Baligant and to the unholy trinity of
Mahomet, Tervagant, and Apollin.
The Saracens, then, should be understood not as a race but as a culture, in Appadurai’s sense of situated difference: group identities are constituted by the conscious mobilization of certain attributes “to articulate the
boundary of difference”—attributes then naturalized as “essential” to group
identity.18 Once conceptualized in this way, the most striking thing about
the Roland ’s gent paienur is that they are virtually indistinguishable from
the Franks. Beyond their exotic names and their occasionally frightful attributes, the pagans speak the same language as the Christians. Whether
exchanging ambassadors or haranguing each other in formulaic displays of
bravado, the two sides have no need of interpreters.19 Each camp, as critics
inevitably note, is in fact a mirror image of the other. Both are orderly feudal hierarchies. Like opposing kings on a chessboard, Charlemagne and
Marsile/Baligant convoke their barons to council, array their troops, harangue
their adversaries, and carry personal standards into battle.20 This will toward
parallelism accounts for the poem’s most notorious “inaccuracy”—the Saracen “trinity” of Mahumet, Apollin, and Tervagant. Both sides imagine their
gods as feudal overlords, expected to dispense favors and miracles commensurate with the devotion they are offered. Identical to the Franks in language
and custom, the pagans arguably differ from the Franks in religion and
nothing more. This play of similiarity-in-difference is responsible for oddly
contradictory descriptions like the following:
Reis Corsalis, il est de l’altre part.
Barbarins est e mult de males arz.
Cil ad parlet a lei de bon vassal:
Pur tut l’or Deu ne volt estre cuard. (885 – 88)
[King Corsablis is there too.
He is a Berber, expert in evil arts.
He spoke like a good vassal:
For all God’s gold he would not be a coward.]
Kinoshita / Alterity, Gender, Nation in the Roland 83
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 84
Enabled by the Roland ’s paratactic logic (each verse functioning as an
autonomous unit of thought, the connections between them made not by
the poet but the hearer),21 this passage both condemns Corsablis’s evil ways
and praises him as a worthy vassal — high praise in a poem which resolves
the opposition between its two main heroes in an affirmation of their common vassalhood: “Roland is brave and Oliver is wise; both are marvelous
vassals” [Rollant est proz e Olivers e sage; / Ambedui sunt d’un grant vasselage] (1093–94).
“Pagans are wrong and Christians are right.” The difference is foundational: Marsile is predestined to come to a bad end because he “serves
Mahumet and prays to Apollin” [Mahumet sert e Apollin recleimet] (8). Yet
the insistence on the parallelism between Saracen and Frank produces two
opposite effects. The first is to introduce the possibility of conversion. If
chivalry is one of the defining practices of Frankish Christendom, then a
warrior like Margariz of Seville cedes little to his Christian counterparts:
“No pagan is such a good knight” [N’i ad païen de tel chevalerie] (960).
Since to be Christian is inseparable from being Charlemagne’s vassal, a
chivalrous pagan would need only swear loyalty to the Frankish emperor to
become a valorous Christian. After all, religion is, quintessentially, one of
those features susceptible to transformation in an individual’s lifetime. This
possibility is made all but explicit in the description of another of Marsile’s
lieutenants:
Uns amurafles i ad de Balaguez;
Cors ad mult gent e le vis fier e cler;
Puis que il est sur sun cheval muntet,
Mult se fait fiers de ses armes porter;
De vasselage est il ben alosez;
Fust chrestiens, asez oüst barnet. (894–99, my emphasis)
[There is an emir from Balaguer.
He has a noble appearance, his face is proud and bright;
When he is mounted on his horse,
He bears his arms proudly.
He is well known for his vassalage;
Were he a Christian, he would be a worthy baron.]22
Imperceptibly, the demarcation between Christian and pagan begins to
seem less intransigent, more insecure. Roland’s cry, we should remember, is
part of his battlefield speech on the duties of a good vassal:
84 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 85
Pur sun seignor deit hom susfrir destreiz
E endurer e granz chalz et granz freiz,
Sin deit hom perdre e del quir e del peil.
Or guart chascuns que granz colps i empleit,
Que malvaise cançun de nus chantet ne seit!
Paien unt tort e chrestiens unt dreit. (1010 –15)
[For his lord one should suffer hardships
And bear both great heat and great cold,
And one should lose both hide and hair.
Now let each man take care to strike great blows:
Let no bad song be sung about us!
Pagans are wrong and Christians are right!]
But if it is the readiness to suffer destreiz for one’s lord that defines the perfect vassal, then stout pagans like Margariz of Seville and the emir of Balaguer are ideal candidates for conversion. Equal to their Frankish counterparts in all but religion, they need only accept Christianity to become
exemplary barons.23
Such potential, however, is never realized. Though in later chansons
de geste it is not unusual for praiseworthy pagans to convert, in the Roland
they are fated to die at Roncevaux with the rest of their companions. With
the single exception of Bramimonde, Charlemagne shows little interest in
winning the hearts and minds of the enemy. Indeed, conversion is viewed
with suspicion: the pagan scheme to trick the Franks into retreating hinges
on Marsile’s false promise to embrace Christianity, as scripted by his advisor
Blancandrin: “[Tell him] you will follow him [to Aix] at Michaelmas and
receive Christian law; you will be his vassal in honour and in all your goods”
[Vos le sivrez a la feste seint Michel, / Si recevrez la lei de chrestiens, / Serez
ses hom par honur e par ben] (37 – 39). Marsile adopts the plan, without
ever seriously considering the possibility of conversion. It is true that at the
height of their battle Charles exhorts Baligant to convert, but it is an offer
the emir summarily dismisses:
“Receif la lei que Deus nos apresentet,
Chrestïentét, e puis t’amerai sempres;
Puis serf e crei le rei omnipotente!”
Dist Baligant: “Malvais sermun cumences!”
Puis vunt ferir des espees qu’unt ceintes.  (3597 – 3601)
Kinoshita / Alterity, Gender, Nation in the Roland 85
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 86
[“Receive the law that God holds out to us,
the Christian faith: I’ll love you from that moment;
believe in Him, serve the Almighty King!”
Said Baligant: “You preach a bad sermon.”
And then they strike with the swords at their sides.]
In besieging Saragossa, Charlemagne seeks less to win new converts to his
faith than to extend the territorial limits of Latin Christendom.
On the one hand, the poem’s lack of interest in conversion reflects
historical reality: early crusaders showed little interest in proselytizing; not
until the thirteenth century, with the emergence of the mendicant orders,
did the Latin church make concerted attempts at missions.24 More fundamentally, it points to an impending crisis of nondifferentiation. If the possibility of conversion is held open, then any sense of identity which depends
on the opposition between self and other is intrinsically unstable. As the
parallelism between Christians and pagans suggests, the designation Christian turns out to be less stable than it initially appears. This uncertainty, I
think, derives from the experience of the crusades in the late eleventh and
early twelfth centuries. The westerners who answered Urban II’s call for an
armed pilgrimage to Jerusalem set out in the certainty that they were going
to liberate the Holy Land and protect their fellow Christians. Once in the
East, however, they met with much more than they had bargained for.
Almost immediately, their relations with the Byzantine Empire deteriorated
into a mix of suspicion, resentment, and outright hostility.25 The reasons
were political: alarmed at the zealous masses who had responded to his
appeal for military assistance, Alexios I showed no compunction in undermining their campaign. But they were also cultural: in the eastern Mediterranean, the crusaders were often less struck by the faith they shared with
Byzantine Greeks or Armenians than by their cultural differences. In some
ways, the westerners found their eastern coreligionists as strange as the Muslim enemy they had come to fight.
The crusades thus made evident the failure of Christianity alone
to secure a strong common identity. In his chronicle of the First Crusade,
Fulcher of Chartres attempts to construct just such an imagined community:
Who ever heard of such a mixture of languages in one army,
since there were French, Flemings, Frisians, Gauls, Allogroges,
Lotharingians, Allemani, Bavarians, Normans, English, Scots,
86 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 87
Aquitainians, Italians, Dacians, Apulians, Iberians, Bretons,
Greeks, and Armenians? If any Breton or Teuton wished to
question me, I could neither understand nor answer. But we who
were diverse in languages, nevertheless seemed to be brothers in
the love of God and very close to being of one mind. ( XIII,
4 –5)26
In Fulcher’s utopian vision, the linguistic and cultural divisions among these
diverse nations were easily overcome by their common religion. Reality,
however, fell short of this ideal. Within a few years of their conquest of
Jerusalem, some crusade leaders were openly negotiating with their Seljuk
“enemies” against their Byzantine “allies,” while the Greeks remained targets
of western resentment throughout the twelfth century.
On the other hand, even as the crusaders’ experience fractured the
myth of a unified Christian culture, it produced the dim awareness of a
commonality among “Latin” Christians. Once in the eastern Mediterranean,
feudal nobles who at home would have thought of themselves as Normans
or Angevins, Flemings or Champenois, gradually became conscious of a
shared culture that distinguished them not only from Muslim Turks and
Arabs but from a spectrum of nonwestern Christians as well. Needing a
term to express this newly perceived corporate identity, the westerners began
referring to themselves as Latin.27 Soon, however, another designation
emerged: Frank. The name itself was originally an ethnic term, and the crusaders themselves first used it in a narrow sense to mean “men from northern France.” It was the Greeks and Muslims who began using it as a general
term for all westerners, but the westerners soon found it convenient to
adopt this broader usage for themselves. Though by and large synonymous
with “Latin Christian,” it was not identical to it. In the twelfth and thirteenth centuries, the designation Frank “implied modernity and power.” It
came to refer to men from the “core” of the former Carolingian Empire, the
agents of expansion of a cluster of institutions and practices — a culture
that Robert Bartlett identifies as “European.” The term, Bartlett writes,
became shorthand for “aggressive westerner” far from home.28
At the turn of the twelfth century, this collective identity was only
just beginning to emerge. In the Chanson de Roland, the word Franc and its
derivatives appear over one hundred times. As a designation for Charlemagne’s forces, it functions in specular relation to the term gent paienur.29 But
its actual use falls somewhere between the narrow sense of “men from northern France” and the broad sense of “aggressive western Christian.” While all
Kinoshita / Alterity, Gender, Nation in the Roland 87
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 88
the men of Charlemagne’s army are haunted by the memory and the promise
of “douce France,” other moments reveal the instability of their imagined
community. In the catalogue of troops preceding the poem’s final battle,
Franks constitute the first, second, and tenth of Charlemagne’s divisions. The
remaining batallions, however, are composed of “nations” that retain their
distinctive identities: Bavarians, Alemanni, Normans, Bretons, Poitevins and
Auvergnats, Flemings, Lorrainers and Burgundians (laisses 218 –25).30 The
differences between this list and Fulcher’s are striking: the Greeks are
conspicuously absent (even though Roland boasts of having conquered
Constantinople) and the Armenians— eastern Christians with whom the
crusaders contracted political and marital alliances—are excised from Christendom altogether. Like the Slavs and Hungarians, peoples converted to
Christianity around the beginning of the eleventh century, two centuries
after the time of the historical Charlemagne and one century before the composition of the Oxford Roland, the Armenians are represented as part of Baligant’s army. These omissions, registering westerners’ uncertainty when confronted by cultural rather than religious difference, reveals the contingency of
Frankish-Christian identity even as the poem attempts to construct it.
These slippages in the representational economy of the Chanson de
Roland become more pronounced if we recall Roland’s dying speech at the
end of the battle of Roncevaux. Remembering all the battles he has won
with his sword Durendal, he enumerates all the peoples he has defeated:
Jo l’en cunquis e Anjou e Bretaigne,
Si l’en cunquis e Peitou e le Maine;
Jo l’en cunquis Normendie la franche,
Si l’en cunquis Provence e Equitaigne
E Lumbardie e trestute Romaine;
Jo l’en cunquis Baiver e tute Flandres
E Burguigne e trestute Puillanie,
Costentinnoble, dunt il out la fiance,
E en Saisonie fait il ço qu’il demandet;
Jo l’en cunquis e Escose e . . .
E Engletere, que il teneit sa cambre. (2322 – 32)
[With this (sword) I conquered Anjou and Brittany for him,
With it I also conquered Poitou and Le Maine for him,
With it I conquered Normandy the free for him,
With it I also conquered Provence and Aquitaine for him,
And Lombardy, and all Romagna,
88 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 89
With it I conquered Bavaria and all Flanders for him,
And Burgundy and all Poland,
Constantinople, which pledged him loyalty,
And Saxony, where he does as he wills;
With it I conquered Scotland and . . . for him,
And England, his chamber, his own domain.]
The very nations now composing Charlemagne’s army are themselves
revealed to be recent conquests. In this light, the catalogue of troops cited
above does not so much describe Frankish unity as perform it, assimilating
former enemy tribes like the Bavarians and the Burgundians into the imagined community of “douce France.” Peter Haidu has argued that Roland
must die to assure the transition from feudalism to the monarchical state:
the great barons, each loyal to Charlemagne but otherwise prone to quarreling among themselves, are supplanted by new men like Rabel and Guineman, subjects of a new kind of national state.31 But his re-collection of the
historical layering underlying the nascent Frankish state suggests that Roland
must die so that the memory of the violence that has gone into the formation of Charlemagne’s empire may die with him.32 Charlemagne’s annihilation of the Saracens to avenge his nephew’s death is the crucible in which a
collective Frankish identity is forged out of the various “nations” comprising
his army—subjects of a “douce France” symbolically (if anachronistically)
demarcated from Marsile’s “clere Espaigne” by the high peaks and dark valleys of the Pyrenees.33
In this light, the battle of Roncevaux matters less for the pagan
defeat than for the creation of a Frankish-Christian imagined community.
This would explain the Roland ’s lack of interest in conversion: Saracens
must remain Saracens in order that the Christians can become Franks.
Ganelon’s treason bespeaks the fragility of a polity bound only by its members’ individual feudal allegiance to Charlemagne. When Marsile tenders his
false promise to embrace “Christian law” [la lei de chrestiens] (39), he
threatens to disrupt the precarious binarism through which Charlemagne’s
empire is defined. In Roland’s “Paien unt tort e crestiens unt dreit,” a militant assertion of Carolingian superiority echoes with anxiety over a concealed crisis of differentiation. Constructed as parallel in all but religion, the
Franks and pagans differ significantly only in one other respect: the role
accorded their women. The instauration of difference begun by Roland’s
death and Charlemagne’s revenge can culminate only in Aix, with the death
of Aude and the conversion of Bramimonde.
Kinoshita / Alterity, Gender, Nation in the Roland 89
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 90
II
Eleven years after his inaugural lecture “La Chanson de Roland et la nationalité française,” Gaston Paris published two articles that established “courtly
love” as the predominant theme of medieval French romance. Like his work
on the chanson de geste, his studies on the matière de Bretagne aimed at making medieval studies more rational and scientific — in pointed contrast to
the dilettantism of the previous generation, specifically of his father Paulin
Paris, his predecessor at the Collège de France. For the son, romance exemplified a self-conscious literary aesthetic in which courtly love — notably in
the works of Chrétien de Troyes — codified the subjection of sexual passion
to reason.34
Separated by little more than a decade, Paris’s two polemical interventions have to a surprising degree set the agenda for discussions of gender
and genre ever since. Accorded pride of place for its role in the emerging
ideology of nationalism, the Chanson de Roland came to determine what
counted as epic. Its ethos of valor and self-sacrifice on the battlefield, its
dualist distinction between Christian and pagans were taken as normative of
the genre. More subtlely, its “marginalization” of Aude and Bramimonde
seemed to establish the epic as a masculine genre, devoted to the enculturation of a warrior class.35 As one critic writes: “Women barely feature in this
masculine world. Aude and Bramimonde, the only two female characters in
the text, mediate male relations: Aude represents the bond between Oliver
and Roland and Oliver’s threat not to allow Roland to marry her (1719 –
21) signals the tension between them, whilst the conversion of Bramimonde symbolizes Charlemagne’s victory over Marsilie.”36 Conversely,
romance was fixed as the genre of the feudal knight’s encounter with
woman. Extricated from her place in the matrix of feudal society, the
courtly lady—exemplified by Guenevere in Chrétien de Troyes’s Le Chevalier de la Charrete — was mystified as the necessarily elusive object of male
chivalric desire. As long as these binary oppositions remained unexamined,
women’s role in epic and in the historical and ideological contexts which
produced it were bound to remain invisible.
Recently, feminist medievalists have done much to redress this
impression of the epic’s monological construction of gender, its “difficulty
in tolerating difference,” its “obsessional, but ultimately unsuccessful attempt
to repress and marginalize alterity.”37 In an important move, Sarah Kay has
challenged the critical tendency to dismiss female characters in epic as the
result of “romance influence.” Because the Chanson de Roland ’s homosocial
90 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 91
emphasis on heroism and masculine community has been taken as normative, critics have read the greater — sometimes much greater — degree of
female presence in the other ninety-nine or so chansons de geste as the product of generic “contamination.” By displacing this model of normativity
and deviance, Kay convincingly shows how integrally women figure in
the epic’s preoccupations of warfare, lineage, feudal loyalty, and crusade.38
This stunning critical reassessment, however, leaves the Roland untouched:
even while recoding its “preeminence” as “eccentricity,” Kay accepts the
received critical notion of its marginalization of women. She implies that
women’s importance in epic can only be brought to light when attention is
shifted away from the Roland toward the unjustly neglected array of lesserknown songs.
But Aude and Bramimonde, I argue, are central to this “masculine”
drama of Ganelon’s betrayal, Roland’s death, and Charlemagne’s revenge. In
a sense, the small role accorded them in comparison to female characters in
other chansons de geste concentrates and, paradoxically, magnifies their ideological significance. Given the crisis of nondifferentiation lurking behind
the bold declaration “Paien unt tort e crestiens unt dreit,” it is the opposition between the vociferously dissenting queen and the quiet Christian maid
that, finally, secures the difference between pagans and Franks. Bramimonde, the Saracen queen who embraces the Christian God “for love,” is
the site where alterity is both articulated and overcome. Ironically, her ideological importance is obscured when attention is focused on either race or
gender. She is marginalized as a subset (“woman”) in the larger category of
pagan, or as a subset (“Saracen”) of that already-marginalized category,
“women.” Located at the confluence of the two groups against which Frankish masculinity defines itself, she cannot be reduced to one or the other but
signifies in the text as a female pagan.
As the figure through whom the Roland genders its politics of difference, Bramimonde appears three times in the text, assuming greater political importance on each occasion. During Ganelon’s embassy to Marsile, she
takes part in the presentation of gifts concluding the treacherous pact. After
her husband’s vassals Valdabrun and Climorin offer Ganelon a sword, a helmet, and a ritual kiss, the Saracen queen steps forward to welcome the renegade Frank in the language of feudal love, her greeting —“I love you well,
lord,” [ Jo vos aim mult, sire] (635) — pointedly inverting the defiant challenge—“I have no love for you at all” [ Jo ne vus aim nïent] (306) — which
the traitor had hurled at Roland during the Franks’ council.39 Then the
queen gives Ganelon two precious necklaces for his wife. By this gesture she
Kinoshita / Alterity, Gender, Nation in the Roland 91
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 92
at once alludes to the sorority of Frankish women into which she will be
assimilated at the end of the poem and singles out the woman— conspicuously absent from the text — at the origin of the drama about to ensue:
Roland’s mother, Charlemagne’s sister, and Ganelon’s wife, who binds the
three together in feudal-familial allegiance and conflict.
In her subsequent appearances in the text, Bramimonde comes
more and more to resemble the stock character of the Saracen princess —
the young foreign woman who converts to Christianity, often for love of a
strong Frankish warrior.40 The case of Orable in La Prise d’Orange is exemplary. The wife of a powerful pagan emir, her beauty— like the emir of Balaguer’s valor — evokes a wistful sense of unrealized potential: “God, what
good are her body and her youth, since she doesn’t know God, the almighty
father!” [Dex! mar i fu ses cors et sa jovente, / Quant Dex ne croit, le pere
omnipotente!] (206– 7).41 In fact, her loveliness does not go to waste: baptized “Guiborc” and married to the valiant count Guillaume Fierebrace, she
is transformed into the ideal wife of a Frankish Christian warlord. This
common epic motif, I have suggested elsewhere, was particularly attractive
for representing the Franks’ contact with peoples they perceived as culturally
superior. Unlike the pagans on Latin Christendom’s northern and eastern
frontiers, Muslims from advanced Mediterranean cultures had nothing to
gain from conversion. They could be seduced only in the imagination, in
the figure of the bold princess ready to exchange a royal Saracen husband
for an intrepid Christian count. When Orable delivers the rich city of
Orange into the hands of Guillaume Fierebrace, accepts baptism, and marries him, it is at once an erotic fantasy and a political victory.42
Stripped of the heady eroticism surrounding a figure like Orable,
Bramimonde demystifies the function of the Saracen princess: to abandon
religion, family, and culture to embrace Frankish Christianity. When Marsile retreats to Saragossa, his hand severed by Roland, the queen takes in the
enormity of the pagans’ loss at a glance and leads an extraordinary revolt
against their gods:
Dedevant lui sa muiller, Bramimunde,
Pluret e criet, mult forment se doluset,
Ensembl’od li plus de .XX. mil humes,
Si maldient Carlun e France dulce.43
Ad Apolin en curent en une crute,
Tencent a lui, laidement le despersunent:
“E! malvais deux, por quei nus fais tel hunte?
92 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 93
Cest nostre rei por quei lessas cunfundre ?
Ki mult te sert, malvais luer l’en dunes!”
Puis si li tolent sun sceptre e sa curune,
Par les mains le pendent sur une culumbe,
Entre lur piez a tere le tresturnent,
A granz bastuns le batent e defruisent;
E Tervagan tolent sun escarbuncle
E Mahumet enz en un fosset butent
E porc e chen le mordent e defulent. (2576–91, my emphasis)
[Before him, his wife Bramimonde
Weeps and wails, mightily lamenting.
With her are more than twenty thousand men
And they curse Charles and fair France.
They rush off to Apollin in a crypt,
Rail against him and foully abuse him:
“O, wretched god, why do you cause us such shame?
Why did you permit our king to be destroyed?
Anyone who serves you well receives a poor reward.”
Then they grab his scepter and his crown
And hang him by his hands from a pillar;
Then they hurl him to the ground between their feet
And beat him and smash him to pieces with huge sticks.
They seize Tervagant’s carbuncle
And fling Muhammed into a ditch;
Pigs and dogs bite and trample on him.]
Though her role is partially subsumed in the plural verbs maldient, tencent,
despersunent, she is unmistakably in charge: the formulaic line “Ensembl’od
li plus de .XX. mil humes” (2578) echoes an earlier description of Marsile
amidst his assembly of barons: “twenty thousand men, and more, all around
him” [Envirun lui plus de vint milië humes] (13). Stepping into the place of
her maimed husband, Bramimonde leads the type of revolt usually attributed to male Saracens who wreak punishment upon their gods “in a futile
rage at the inefficacy of their . . . protection.”44 When Emir Baligant’s
ambassadors greet her in the name of the Saracen “trinity,” Bramimonde
accuses the gods of what amounts to a breach of feudal contract:
Dist Bramimunde: “Or oi mult grant folie!
Cist nostre deu sunt en recreantise;
Kinoshita / Alterity, Gender, Nation in the Roland 93
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 94
En Rencesval malvaises vertuz firent:
Noz chevalers i unt lesset ocire;
Cest mien seignur en bataille faillirent;
Le destre poign ad perdut, n’en ad mie,
Si li trenchat li quens Rollant, li riches.
Trestute Espaigne avrat Carles en baillie.
Que devendrai, duluruse, caitive?
E! lasse, que nen ai un hume ki m’ociet!” (2714–23)
[Said Bramimonde: “Now I hear great madness!
These gods of ours are vanquished !
At Rencesvals they performed miserably:
They let our knights be killed there,
They failed my lord here in battle.
He has lost his right hand, he no longer has it;
Count Roland the powerful cut it off.
Charles will have all of Spain in his power;
What will become of me, miserable wretch?
O, woe is me that I have no man to kill me.”]
For a society which articulated faith and feudal loyalty in the same discourse, nothing could be more damning than Bramimonde’s accusation of
recreantise. Mohammed, Apollin, Tervagant, she is saying, are bad lords:
their weak showing (“malvaises vertuz”) in allowing the pagans to perish at
Roncevaux contrasts miserably with the great miracle (“vertuz mult granz”
[2458]) the Christian God had performed at Charlemagne’s behest —stopping the sun so the emperor could pursue Roland’s killers. Her allusion to
Marsile’s severed hand calls attention to the symbolic castration of the entire
political order, while her woeful lament over her own future points to the
dire fate awaiting female enemy captives. In reciting this litany of misfortune, Bramimonde builds an unimpeachable case for the indefensibility of
the pagan spiritual order.
After denouncing her gods, Bramimonde next sets her sights on the
Saracens’ political hierarchy. When the ambassadors reassure her that Baligant will pursue the Franks all the way to France, she turns on them with
cutting sarcasm:
Dist Bramimunde: “Mar en irat itant!
Plus près d’ici purrez truver les Francs:
En ceste tere ad estet je .VII. anz.
94 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 95
Li emperere est ber e cumbatant:
Meilz voel murir que ja fuiet de camp;
Suz ciel n’ad rei qu’il prist a un enfant.
Carles ne creint nuls hom ki seit vivant.” (2734– 40)
[Bramimonde said: “He need not go so far!
You will find the Franks closer by;
He has now been in this land for seven years.
The emperor is valiant and a fine warrior;
He would sooner die than flee the field.
There is no king on earth he ranks as more than a child.
Charles fears no man alive.”]
Wresting the initiative from her incapacitated husband, the indignant queen
devastatingly exposes the bankruptcy of the pagan feudal order: her allusion
to the Franks’ seven-year siege of Saragossa impugns Baligant’s failure to
come to the aid of his vassal; her insistence that Charlemagne would never
flee the battlefield calls attention to Marsile’s own flight. In strategically
praising the Christian emperor, she signals her contempt for all of pagan
society. Belatedly, her hapless husband rouses himself to speech: “ ‘Now let
that be,’ said King Marsile; and to the messengers: ‘Lords, speak to me.’ ”
[“Laissez ço ester” dist Marsilies li reis. / Dist a messages: “Seignurs, parlez
a mei! ”] (2741– 42, emphasis added).
Bramimonde’s verbal aggressivity — her unruly speech — has fascinated generations of critics. In the late nineteenth and early twentieth century, her outspokenness marked her, above all, as foreign. For Léon Gautier,
the “shameless behavior” of such women was a sure trace of the Germanic
origins of epic. F. M. Warren, writing in 1914, found the “masterful nature”
of the Saracen princess “foreign to France and to the feminine ideal of the
French.”45 More recently, Kay has emphasized how such speech exceeds
male control: the Saracen princess “does not merely ventriloquize a controlling masculine fantasy: she helps to shape it, and thereby disrupts assumed
hierarchies.”46
But as postcolonial feminism has taught us, gender serves the construction of cultural difference just as cultural difference determines the
construction of gender. The ideological work Bramimonde performs
depends precisely on the “strategic deployment of sexual difference in . . .
narratives of an East-West encounter.”47 Her brazenness disrupts, not all
hierarchies, but the pagan religious and social order. For the Franks, on the
Kinoshita / Alterity, Gender, Nation in the Roland 95
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 96
other hand, her outspokenness is not merely acceptable but desirable: her
dissent exposes the bankruptcy of the Saracen system, at the same time that
her aggressivity, unthinkable in a proper Christian wife, instantiates pagan
deviance and prepares the way for her conversion. Whatever feminist edge
her words seem to have works to the interest of Frankish Christendom: she
must speak, for she is the site where pagan society turns against itself.
Therefore it is not surprising that after Charlemagne’s rout of Baligant’s army it is Bramimonde who delivers the city into the emperor’s
hands:
E Carles ad sa bataille vencue.
De Sarraguce ad la porte abatue:
Or set il ben que n’est mais defendue.
Prent la citet, sa gent i est venue:
Par poestet icele noit i jurent.
Fiers est li reis a la barbe canue,
E Bramidonie les turs li ad rendues:
Les dis sunt grandes, les cinquante menues. (3649–56 )
[And Charles has won his battle.
He has knocked down the gate of Saragossa,
He knows well now: it will not be defended.
He takes the city, his people entered in,
By virtue of their power they slept there that night.
Proud is this white-bearded king,
And Bramimonde has surrendered the towers to him:
Ten of them are big and fifty are small.]
It is in this resolution to Charlemagne’s seven-year siege that the stakes of
the Roland ’s poetics of conquest become visible, along with the role alloted
the Saracen queen. In the late-eleventh- and early-twelfth-century Reconquest, Spanish kings accustomed to Iberian traditions of convivencia and
anxious to maintain the prosperity of their new possessions treated their
Muslim subjects with tolerance. When Alfonso I of Aragon captured the
city of Saragossa (1118) in fact rather than fiction, he guaranteed its Muslim inhabitants “the exercise of their own religion, their rule by Muslim officials, their lives, property, and freedom of movement within the realm or
emigration from it if they wished.”48 Such leniency, though “normal among
Spaniards,” was alien to the French. When Alfonso VI of León-Castile conquered Toledo in 1085, his Frenchborn queen, Constance of Burgundy, and
96 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 97
Archbishop Bernard de Sédirac, were scandalized at the favorable terms the
kings granted his new Muslim subjects. During the king’s absence, and in
violation of his promise to the Muslims, they “seized the great mosque and
rededicated it as a cathedral.”49
In the Roland, the fictional inhabitants of Saragossa are not so
lucky. When the emperor storms the gates and occupies the city, his troops
wreak havoc in its mosques and synagogues, forcibly baptizing more than
100,000 pagans:
Li emperere ad Sarraguce prise.
A mil Franceis funt ben cercer la vile,
Les sinagoges e les mahumeries;
A mailz de fer e a cuignees qu’il tindrent
Fruissent les ymagenes e trestutes les ydeles:
N’i remeindrat ne sorz ne falserie.
Li reis creit en Deu, faire voelt sun servise,
E si evesque les eves beneïssent,
Meinent paien entesqu’al baptisterie:
S’or i ad cel qui Carle cuntredie,
Il le fait pendre o ardeir ou ocire. (3660 –70)
[The emperor has taken Saragossa.
He sends a thousand French to search the city,
The synagogues, and the mosques.
With iron mauls and hatchets in their hands,
They break the images, shatter all idols:
There shall be no more magic and no more fraud.
The king believes in God, and wants to impose his worship:
And so bishops bless the waters
And lead the pagans to the baptismal font:
If anyone refuses Charles,
He has that man hanged, or burned, or killed.]
For Bramimonde, on the other hand, Charlemagne has something different
in mind:
Baptizet sunt asez plus de .C. milie
Veir chrestien, ne mais sul la reïne.
En France dulce iert menee caitive:
Ço voelt li reis par amur cunvertisset. (3671– 74, my emphasis)
Kinoshita / Alterity, Gender, Nation in the Roland 97
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 98
[More than a hundred thousand are baptized
True Christians, but not the queen:
She will be led, a captive, to sweet France:
The king wants her to convert for love.]
Learning of Charlemagne’s annihilation of Marsile’s army, Bramimonde
had lamented that she had no one to slay her, yet with the actual conquest
of Saragossa, she is the only survivor not immediately forced to choose
between conversion and death. This fate sets her apart from Climborin and
Valdabrun, the two pagans who had preceded her in exchanging greetings
and gifts with Ganelon; both die at Roncevaux (killed by Oliver and
Roland, respectively), with the poet explicitly evoking their role in the treasonous pact.50 Bramimonde, on the other hand, survives, to be taken back
to Aix-la-Chapelle and instructed in the Christian faith, that she might
actively choose it. No trace of the carnal or profane compromises Charlemagne’s desire to convert Bramimonde “par amur”—“love” signals not
romantic desire but the voluntary nature of her conversion. The narrator
forestalls any suspicion to the contrary:
Muntet li reis e si hume trestuz
E Bramidonie, qu’il meinet en sa prisun;
Mais n’ad talent que li facet se bien nun. (3679 –81)
[Now the king mounts his horse, all his men mount,
And Bramimonde, whom he leads off to captivity,
But he only wishes to do her good.]
For untold thousands of Saracen warriors, refusal to convert meant instant
death; in contrast, their queen is taken to the heart of the Carolingian
Empire, that she might embrace Charlemagne’s God as actively as she has
denounced her own. At Aix, Bramimonde — previously the lone woman in
a host of pagan men — is paired with Roland’s fiancée Aude, the contrast
between the Frankish maiden and the Saracen queen serving as final confirmation of cultural difference.
III
In symbolic terms, the convert is the obverse of the traitor: Bramimonde’s
baptism, signaling her acceptance both of Christianity and of the feudal
Frankish world it entails, sutures the breach opened by Ganelon’s treasonous
98 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 99
commerce with Marsile.51 To this point, Bramimonde has functioned as the
site of disruption: denouncing Saracen culture from within, her unruly
speech (re)establishes difference between the two barely distinguishable
sides. Now, as the scene shifts from Saragossa to Aix, Bramimonde is paired
with the poem’s other female character, belle Aude. However small their
place in the Roland as a whole, the two women are prominent here at its
end. Their appearances bracket the climactic episode of Ganelon’s trial,
Aude’s preceding it (laisses 268 – 69) and Bramimonde’s immediately following (laisse 290). For many critics, gender supersedes culture: the two,
Christian and pagan, are linked by their common marginalization.52 But the
contrast between Frankish and Saracen femininity is crucial to the Roland ’s
resolution. Reversing the binarism of later colonial discourse, it is the foreign woman who displays a new feminine agency while the Frankish woman
is consigned to passivity and silence.
Returning to his capital, Charlemagne is met by Roland’s fiancée
Aude. Though this is her first appearance in the text, her name had come up
during the battle at Roncevaux when her brother Oliver, angry at Roland for
sounding the olifant, threatens to break off his friend’s engagement to his
sister (1719– 21). Once the two patch up their differences, however, she is
forgotten. As an object of exchange, she is less important in forging the
homosocial bond than in threatening to disrupt it. Now, given the chance to
speak for herself, her only thought is for Roland:
Li empereres est repairet d’Espaigne
E vient a Ais, al meillor sied de France;
Muntet el palais, est venut en la sale.
As li Alde venue, une bele damisele.
Ço dist al rei: “O est Rollant le catanie,
Ki me jurat cume sa per a prendre?”
Carles en ad e dulor e pesance,
Pluret des oilz, tiret sa barbe blance:
“Soer, cher’ amie, d’hume mort me demandes.” (3705–13)
[The emperor has returned from Spain
And he comes to Aix, the best seat of France.
He goes up to the palace and came into the hall.
See now, Aude came to him, a beautiful girl.
She said to the king: “Where is Roland the captain,
Who swore to take me as his wife?”
Charles reacts with grief and distress;
Kinoshita / Alterity, Gender, Nation in the Roland 99
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 100
He weeps and tugs at his white beard:
“Sister, dear friend, you ask me about a dead man.”]
Through his grief, Charlemagne struggles to imagine a new future, for himself, for the Franks, and for Aude: “I shall give you a very fine replacement:
that is Louis, I do not know of better. He is my son, and he will possess my
marches” [Jo t’en durai mult esforcet eschange; / Ço est Loewis, mielz ne sai
a parler. / Il est mes filz e si tendrat mes marches] (3714 –16). Aude, however, rejects the proffered exchange — not because she objects to Roland’s
designated replacement (Louis is destined to cut a poor figure in the epic
tradition) but because her love is nontransferable. “Preux” like her fiancé
rather than “sage” like her brother, she drops dead at the emperor’s feet.53
Refusing her prescribed role in the feudal politics of lineage, she elects
the dramatic gesture of absolute devotion over the option of a measured,
longer-term advantage.54
Alde repunt: “Cest mot mei est estrange.
Ne place Deu ne ses seinz ne ses angles
Après Rollant que jo vive remaigne!”
Pert la culor, chet as piez Carlemagne,
Sempres est morte. Deus ait mercit de l’anme!
Franceis barons en plurent e si la pleignent. (3717– 22)
[Aude replies: “These words are strange to me.
May it not please God or his saints or his angels,
That I live on after Roland!”
She pales and falls at Charlemagne’s feet,
Already dead. May God have mercy of her soul!
The French barons weep for her and mourn her.]
In feudal terms, Aude’s decision is short-sighted: in dismissing as mot
estrange Charles’s offer to exchange a dead Roland for a live Louis, she
rejects the opportunity to join her lineage to the emperor’s direct agnatic
line. But in dying, she demonstrates the incontrovertibility of her loyalty to
Roland just as Roland had acted out his fidelity to Charles. The very briefness of the episode concentrates its effect: with drama and pathos, this scene
conjures the enormity of Charlemagne’s loss and adds to the tally for which
Ganelon must pay.55
If Aude’s ideological task is to refuse all exchange, Bramimonde’s is
to embrace it. Having denounced her husband, his overlord, and their gods,
100 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 101
she has been brought to Aix to renounce further her homeland, her name,
and her religion. She is baptized in the poem’s penultimate laisse, just after
Ganelon’s execution. The conversion of the Saracen queen provides closure
to this song of feudal loyalty and heroic sacrifice, as if her integration into
Frankish society “is as necessary for France’s salvation as it is for Bramimonde’s own.”56 No male pagan of stature remains alive: offered a chance at
conversion, Baligant has categorically refused, while those like the emir of
Balaguer who would have made good Christians have been slain with the
rest.57 Only Bramimonde survives, her conversion par amur bearing witness
to the superiority of Frankish Christianity. That she converts for love, but
not for love of Charles, is sometimes perceived as a lack.58 But the fact that
she is not Charlemagne’s romantic interest is crucial: her love is not for a
man but for a religious and social order.
This is not to say, however, that Bramimonde can be recuperated as
a feminist subject. Her display of self-determination is less an act of feminist agency than part of a scripted role in the construction of Frankish
Christianity.59 For Bramimonde, conversion means submitting both to the
Christian God and to a discursive regime that demands women’s silent
acquiescence. As Saracen queen, she enjoyed the prerogative of strong
speech. Her rebelliousness against the pagan gods distinguished her from
her husband, predestined for a bad end (8), as well as from Aude, who
tersely refused marriage to a future emperor in favor of immediate death. As
a Christian, however, she must relinquish the defiant spirit that has defined
her. Though she elects Christianity par amur as Charlemagne intended, he
is the one to articulate her will:
Quant li empereres ad faite sa venjance,
Sin apelat ses evesques de France,
Cels de Baviere e icels d’Alemaigne:
“En ma maisun ad une caitive franche.
Tant ad oït e sermuns e essamples
Creire voelt Deu, chrestientet demandet.
Baptisez la, pur quei Deu en ait l’anme.”
Cil li respundent: “Or seit fait par marrenes!”
. . . . . . . . . . . . . . .
As bainz ad Ais mult sunt granz les c . . .
La baptizent le reïne d’Espaigne:
Truvé li unt le num de Juliane.
Chrestienne est par veire conoisance. (3975 –87)
Kinoshita / Alterity, Gender, Nation in the Roland 101
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 102
[When the emperor has taken his revenge,
He summoned the bishops of France
And those of Bavaria and Germany.
“In my house there is a noble captive;
She has heard so many sermons and parables
That she wishes to believe in God and seeks
Christianity.
Baptize her, so that God may have her soul.”
They reply: “May it be done through godmothers”
. . . . . . . . . . . . . . . .
In the baths at Aix there is a vast gathering;
There they baptize the queen of Spain.
They found for her the name of Juliana;
She is a Christian, through true understanding.]
In giving Ganelon the necklace for his wife, Bramimonde had momentarily
made visible the otherwise neglected world of Frankish women. Now she is
silently absorbed into the gyneceum of godmothers (marrenes) waiting to
oversee her baptism in the waters of Aix.
Despite enormous losses at Roncevaux, Charlemagne’s symbolic
victory over the Saracens is complete. The bankruptcy of pagan society has
been revealed not only in the failure of its kings and recreantise of its gods,
but in the unruliness of the woman willing to denounce them. Bramimonde’s dissent shores up the distinction between Saracens and Christians
at a crucial moment of the threat of nondifferentiation. Brought from Spain
to the heart of the Frankish empire, she is like the saint choosing translatio:
her successful removal from Saragossa to Aix vindicates Charlemagne’s
cause.60 The olifant and the bodies of Roland, Oliver, and Turpin have been
deposed along the way in the churches of Bordeaux and Blaye, objects of
veneration to serve as lieux de mémoire of Charlemagne’s loss and sacrifice.
Bramimonde, in contrast, survives as the living tribute to the Franks’ victory. Brought to the center of empire, her conversion and lapse into silence
punctuate Charlemagne’s Spanish campaign and definitively substantiate
Roland’s vision of the world: “Paien unt tort e crestiens unt dreit” (1015).
In this article, I have rather perversely argued that the role of women is most
central in the very place where it has most automatically been dismissed, in
the Chanson de Roland, enshrined since the nineteenth century as the
canonical text of the French Middle Ages. The deeroticization of the Sara102 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 103
cen princess motif, far from stripping it of its interest, reveals its ideological
core: the pagan queen who chooses to embrace Frankish Christianity confirms it, not only by the fact of her conversion but because, contrasted to
Saracen men on the one hand and a Frankish woman on the other, she is the
site where gender distinctions and cultural difference meet, collapse, and are
finally reinscribed. The masculinist, feudal-Christian monologism usually
assumed to be the Roland ’s point of departure instead turns out to be its
most significant product. In the person of Bramimonde/Juliane, the specter
of nondifferentiation linked to the prospect of the mass conversion of Saracen warriors is put to rest: disconcertingly aggressive, she instantiates pagan
difference even as she abandons it.
The case of the Chanson de Roland reveals how powerfully our
lieux de mémoire both determine and are determined by the ongoing process
of interpretation. It reveals how strongly the male homosociality that has
haunted the discipline of medieval studies from its inception has inflected
our vision of feudal society. It reveals the specificity of modern notions of
alterity, often rooted in unexamined assumptions of racial or biological difference. In The Subject of Violence, Peter Haidu observes that to date nearly
all work on the Roland ’s “historical contextualization” or “hidden historical
meaning” has been done by foreign scholars, while studies undertaken in
France remain dominated by “traditional French ideological values of
nationalism and superficial religiosity.”61 For the past century and more,
medievalists have cultivated the Chanson de Roland as a lieu de mémoire of
the origin of a precocious French national sentiment. Unpacking that site
reveals the poem’s critical history to be inseparable from a history of FrancoGerman conflict and colonial ideology. I began this essay by looking at the
Roland as an example of what we might call second-degree Orientalism
which, though not (like Heart of Darkness or Passage to India ) produced
during the era of European colonial expansion, nevertheless became central
to the colonial imaginary. Disengaging the Roland from this colonial context, in which alterity is implicitly or explicitly cast in the taxonomic categories of racialized difference, brings into focus the fluidity characterizing
medieval notions of difference. This in turn reveals “France” and “Europe”
to be not geographical entities given in advance, but ideological constructs
with their own deeply complicated history of conquest, colonization, and
acculturation, in ways that continue to resonate, for example, in political
debates on multiculturalism in France or in the emergence of the European
Community.62 The Roland, it turns out, is an exemplary text — not for its
depiction of heroic sacrifice or its articulation of a precocious national senKinoshita / Alterity, Gender, Nation in the Roland 103
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 104
timent, but for the way it concentrates and imbricates questions of self and
other, gender and genre, history and ideology.
That is why, for each generation, to be a medievalist has meant taking a stand on the Chanson de Roland.
a
Notes
1
2
3
4
5
6
An early version of this essay was delivered at the 1997 University of California
Humanities Research Institute Seminar, “Theorizing Race in Pre- and Early Modern
Contexts,” organized by Margo Hendricks at UC Santa Cruz. I thank the participants,
especially Geraldine Heng and Oumelbanine Zhiri, for their encouragement. Thanks
also to Chris Connery, Carla Freccero, and Monique Young for their comments. This
research was supported by faculty research funds granted by the University of
California, Santa Cruz.
“[L]a Chanson de Roland joue pour les études médiévales en tant que discipline, en
tant qu’institution, un rôle fondateur. . . . Etre médiéviste c’est, au plus vrai, prendre
position sur la C. R.” Bernard Cerquiglini, “Roland à Roncevaux, ou la trahison des
clercs,” Littérature 42 (1981): 40.
La Chanson de Roland, ed. Joseph Bédier (Paris: Piazza, n.d.), hereafter cited
parenthetically in the text by line numbers. Translations are my own. I have consulted
The Song of Roland, trans. Frederick Goldin (New York: Norton, 1978); and The Song
of Roland, trans. Glyn Burgess (Harmondsworth: Penguin, 1990).
Lagarde and Michard, for example, devote thirty pages to the Roland (3– 32) and a
combined ten (33 – 42) to Huon de Bordeaux, Le Couronnement de Louis, Aliscans, and
Raoul de Cambrai. They write: “A lire nos extraits des autres Gestes, dont l’intérêt est
surtout documentaire ou pittoresque, on sentira la superiorité de la C 
R par son art, par sa valeur humaine, par sa spiritualité plus grande. Les moeurs
y sont moins barbares, l’idéal chevaleresque plus pur. C’est comme si la dégradation
même de la féodalité, contre laquelle réagira l’esprit courtois, se reflétait dans les
épopées postérieures.” André Lagarde and Laurent Michard, Le Moyen Age: Les Grands
Auteurs Français du Programme (Paris: Bordas, 1962), 6, emphases in the text.
Gaston Paris characterizes the Roland on this occasion as being a root of French
identity, despite his earlier description of epic as a fusion of the Teutonic and the
Latin. David Hult, “Gaston Paris and the Invention of Courtly Love,” in Medievalism
and the Modernist Temper, ed. R. Howard Bloch and Stephen G. Nichols (Baltimore:
Johns Hopkins University Press, 1996 ), 195.
Promoted by monarchical and commercial interests, the invasion was launched on the
pretext of an incident that had taken place three years before, when the dey of Algiers
struck the French consul with a flyswatter.
In 1871 the archbishop of Algiers, former bishop of Nancy, launched the following
appeal: “Christian populations of Alsace and Lorraine, now on the roads of France,
Switzerland, Belgium, fleeing your burned out houses, your ruined fields, Algeria—
104 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
7
8
9
10
11
12
13
14
2/26/01
6:59 PM
Page 105
African France— through my voice as bishop, opens its doors and its arms to you. . . .
Come, and contribute a hard-working, moral, Christian population to this still-infidel
soil; you will be its true apostles, before God and the fatherland [la patrie]. In France
as in Algeria, it is easily understood . . . how much your arrival would serve the very
cause of the mother country [la mère-patrie] in developing its influence in a land
almost as large as France itself, which in the past has numbered more than twenty
million inhabitants, and which is closer to Marseilles than Marseilles is to Paris.” In
Raoul Girardet, ed., Le Nationalisme français (Paris: Seuil, 1983), 93– 94, my
translation.
Jamil M. Abun-Nasr, A History of the Maghrib in the Islamic Period (Cambridge:
Cambridge University Press, 1987 ), 268.
Girardet, Le nationalisme français, 85, 104.
This was an important move in the nationalization of medieval studies. Previously,
those seeking to imbue their work with scientific rigor were obliged to travel abroad
to study with German masters. In the 1870s and 1880s, the French established
hundreds of university professorships and founded journals like Romania (1872) and
Le Moyen Age (1888), signaling the professionalization of medieval studies in France.
R. Howard Bloch, “842: The First Document and the Birth of Medieval Studies,” in
A History of French Literature, ed. Denis Hollier et al. (Cambridge: Harvard University
Press, 1989), 11. See also Hans Ulrich Gumbrecht, “‘Un Souffle d’Allemagne Ayant
Passé’: Friedrich Diez, Gaston Paris, and the Genesis of National Philologies,”
Romance Philolog y 40 (1986 ): 1– 37.
As Duggan notes, three major editions of the Roland (beginning with Léon Gautier’s
in 1872) were conceived, not coincidentally, at high moments of Franco-German
conflict. Joseph J. Duggan, “Franco-German Conflict and the History of French
Scholarship on the Song of Roland,” Hermeneutics and Medieval Culture, ed. Patrick
Gallacher and Helen Damico (Albany: State University of New York Press, 1989),
97 –106. The term lieu de mémoire is Pierre Nora’s. See his “General Introduction:
Between Memory and History,” in Pierre Nora, ed., Realms of Memory: Rethinking the
French Past, trans. Arthur Goldhammer (New York: Columbia University Press,
1996 ), 1–20.
Edward W. Said, Orientalism (New York: Random House, 1978).
Ann Laura Stoler and Frederick Cooper, “Between Metropole and Colony: Rethinking
a Research Agenda,” in Tensions of Empire: Colonial Cultures in a Bourgeois World, ed.
Frederick Cooper and Ann Laura Stoler (Berkeley: University of California Press,
1997 ), 7, 11, 34.
“It is indicative that one way to disguise oneself as a member of another race was to
adopt their hair-styles.” Robert Bartlett, The Making of Europe: Conquest, Colonization
and Cultural Change, 950–1350 (Princeton: Princeton University Press, 1993),
197– 98.
These figures reemerge vividly in the travel literature of the late Middle Ages. See John
Block Friedman, The Monstrous Races in Medieval Art and Thought (Cambridge:
Harvard Univeristy Press, 1981), chap. 1; and Jeffrey Jerome Cohen, “The Order of
Monsters: Monster Lore and Medieval Narrative Traditions,” in Francesca Canadé
Sautman et al., eds.,Telling Tales: Medieval Narratives and the Folk Tradition (New
Kinoshita / Alterity, Gender, Nation in the Roland 105
JMEMS31.1-04 Kinoshita
15
16
17
18
19
20
21
22
2/26/01
6:59 PM
Page 106
York: St. Martin’s, 1998), 37 – 58. On the stereotyped view of Christians in Spanish
Muslim writings of the same period, see Aziz al-Azmeh, “Mortal Enemies, Invisible
Neighbours: Northerners in Andalusi Eyes,” in Salma Khadra Jayyusi, ed., The Legacy
of Muslim Spain (Leiden: E. J. Brill, 1992), 259–72, esp. 264.
Though Mal-, like Tort-, is not limited to Africans. Paula Gilbert Lewis, “The
Contemporary Relevance of the Teaching of La Chanson de Roland: The Christian
European Mind versus ‘the Other,’” College Language Association Journal 25 (1982):
344– 45. See also Gerald Herman, “Some Functions of Saracen Names in Old French
Epic Poetry,” Romance Notes 11 (1970): 431.
Compare Robert de Clari’s chronicle of the Fourth Crusade, in which he describes the
crusaders’ astonishment at meeting the black Christian king of Nubia in the palace of
the Byzantine emperor: “a king came there whose skin was all black, and he had a
cross in the middle of his forehead that had been made with a hot iron. . . . And the
barons gazed at this king with very great wonder” [si vint illueques uns rois qui toute
avoit le char noire, et avoit une crois en mi le front qui lui avoit esté faite d’un caut fer.
. . . Si esgarderent li baron chu roi a molt grant merveille]. Robert de Clari, La
Conquête de Constantinople, ed. Philippe Lauer (Paris: Honoré Champion, 1924),
LIV, 3 –32; Robert of Clari, The Conquest of Constantinople, trans. Edgar Holmes
McNeal (New York: Columbia University Press, 1936; repr. Toronto: University of
Toronto Press, 1996 ), 79 – 80. In the visual arts, Spanish Muslims are first depicted as
black in the second half of the thirteenth century. Turkish Muslims, however,
continue to be depicted as white. See Philippe Sénac, L’Image de l’Autre: l’Occident
médiéval face à l’Islam (Paris: Flammarion, 1983), 71– 72.
Joseph J. Duggan, A Concordance of the Chanson de Roland (Columbus: Ohio State
University Press, 1969), 282 – 83, 417–18. The former term from the Latin paganus,
meaning a rustic “from the country or village,” originally referred to an internal other:
Greco-Latin polytheists, as opposed to city-dwelling Christians. Saracens, on the other
hand, named an external other: originally an Egyptian term for the people of Arabia,
it became a Byzantine designation for Muslims.
Arjun Appadurai, “Here and Now” in Modernity at Large: Cultural Dimensions of
Globalization (Minneapolis: University of Minnesota Press, 1996 ), 12–14.
Contrast the Prise d’Orange, in which Guillaume makes use of a vassal who, by virtue
of having spent three years as a prisoner of war, has acquired a knowledge of “turquois
et Aufriquant, bedoïn et basclois.” Claude Régnier, ed., La Prise d’Orange: Chanson de
geste de la fin du XIIe siècle, 7th ed. (Paris: Klincksieck, 1986 ), 328.
Although in laisse 2, as Catherine Yu pointed out to me, Marsile conducts his council
from a recumbent position: “Sur un perrun de marbre bloi se culchet” (12).
Erich Auerbach, “Roland against Ganelon,” in his Mimesis: The Representation of
Reality in Western Literature, trans. Willard R. Trask (Princeton: Princeton University
Press, 1953), 96 –122.
As Comfort writes, with a telling choice of metaphor: “As the likelihood is ever
present that a Saracen may change his faith, it is important that he shall not be
painted in advance in too black colors. The evidence would hardly show that the
Christians thought of the Saracens as ethically or culturally inferior to themselves.”
106 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
23
24
25
26
27
28
29
30
2/26/01
6:59 PM
Page 107
William Wistar Comfort, “The Literary Role of Saracens in the French Epic,” PMLA
55 (1940): 633.
As Jonin writes: “En général dans nos épopées, lorsque le païen a une nature morale
foncièrement bonne, il supporte mal la religion musulmane. Il comprend
obscurément que le bien et le christianisme ne peuvent pas se dissocier, et que l’un
appelle l’autre. Plus il réfléchit, et plus il reconnaît, les chrétiens aidant, qu’il est
nécessaire et logique qu’un homme de bien s’adhère à une religion du bien, c’est-à-dire
la religion chrétienne.” Pierre Jonin, “Le climat de croisade des chansons de geste,”
Cahiers de civilisation médiévale 7 (1964 ): 285.
Before this, attempts at conversion were directed at the preliterate pagans on the
northern and eastern frontiers of Christian Europe—the groups to whom
“Christianity” represented not simply a religion but a culture comprising technologies
of literacy, education, administration, and social organization. Drawn by the prestige
of such “modern” Carolingian and post-Carolingian practices, many Baltic, Slavic,
and Scandinavian leaders converted voluntarily, and their entire people with them. In
the Mediterranean, on the other hand, Latin Christians encountered Muslims and
Greek Christians whose sophistication surpassed their own and who had little
incentive to embrace Latin Christianity as part of a larger cultural package. See
Bartlett, The Making of Europe, chap. 11. However, for some anecdotal accounts of
conversions in the Iberian peninsula, see Benjamin Z. Kedar, Crusade and Mission:
European Approaches toward the Muslims (Princeton: Princeton University Press,
1984 ), chap. 1. On literary representations of conversion, see Norman Daniel, Heroes
and Saracens: An Interpretation of the Chansons de Geste (Edinburgh: Edinburgh
University Press, 1984 ), chap. 9.
These attitudes soon work their way into vernacular literature. See Sharon Kinoshita,
“The Poetics of Translatio: French-Byzantine Relations in Chrétien de Troyes’s Cligés,”
Exemplaria 8 (1996 ): 315–54.
Fulcher of Chartres’s chronicle of the First Crusade is cited in Edward Peters, ed.,The
First Crusade: The Chronicle of Fulcher of Chartres and Other Source Materials
(Philadelphia: University of Pennsylvania Press, 1971), 48 – 49.
“It came to have a quasi-ethnic nuance, as in the phrase gens latina, ‘the Latin
people.’. . . The category ‘Latin’ thus had a role in the self-description of the people
of western Europe and obviously helped lend a kind of conceptual cohesion to
groups of very varied national origin and language” (Bartlett, The Making of Europe,
19).
Ibid., 101– 5. These institutions and practices included universities, the international
religious orders, the minting of coins, the use of court chanceries— even the
preference for universal Christian names like John or Henry over local ones like
Duncan or Pribislaw. Regions within Latin Christendom but outside this core were
subject to “Frankish” conquest and colonization, as in the Anglo-Norman expeditions
in Wales and Ireland.
See Duggan, Concordance of the Chanson de Roland, s.v. Franc and paien.
These descriptions of the batallions contrast the individual presentation of the twelve
peers, each with his personal entourage, before the rearguard’s battle at Roncevaux.
Kinoshita / Alterity, Gender, Nation in the Roland 107
JMEMS31.1-04 Kinoshita
31
32
33
34
35
36
2/26/01
6:59 PM
Page 108
The current list is reprised near the end of the text in the groups of “judges”
Charlemagne convokes for the trial of Ganelon (3700– 3702).
Peter Haidu, The Subject of Violence: The Song of Roland and the Birth of the State
(Bloomington: Indiana University Press, 1993), 149 – 51, 189– 91. In this reading,
“the monarchical class is the subject, God is the destinator, and féodalité is the
antagonist. The Saracens, enemies on the actorial level, are adjuvants on actantial
level, helping to decimate the feudal class, the anti-subject” (175). The Roland
“creates and kills Roland, the peers, Ganelon, the 20,000 troops of the rear-guard, in
order to perform the subjection of the warrior class, subjection to a monarchy that
does not yet exist, which awaits in the wings until the scene is set for its entrance” (190).
On Roland as the repository of memory in the poem, see Eugene Vance, “Roland and
Charlemagne: The Remembering Voices and the Crypt,” in Mervelous Signals: Poetics
and Sign Theory in the Middle Ages (Lincoln: University of Nebraska Press, 1986 ),
51–85, esp. 60.
Geraldine Heng notes that eyewitness chroniclers of the First Crusade often used the
name Hispania to designate interior regions of Syria. “Cannibalism, the First Crusade,
and the Generation of Romance,” differences 10 (1998): 118, 154 – 55 n. 39. In the
eleventh and twelfth centuries, the Pyrenees marked the frontier between lands bound
together by post-Carolingian institutions and practices (see above n. 28) and the
Christian but not yet fully “europeanized” Spanish kingdoms. Only in 1080, for
example, was the distinctive Mozarabic rite replaced by the Roman rite which had
been standard elsewhere in the West since the time of Charlemagne. The kingdom of
Navarre, which straddled the Pyrenees, was the site of intense acculturation. French
presence was strong in pilgrim traffic on the route to Santiago and in settler
communities in Estella (from 1090) and in the new coastal towns of San Sebastián
and Bayonne (twelfth century). Artistic and cultural influence was so strong among
the elite that by the twelfth century both “the ecclesiastical and the court culture of
the kingdom of Navarre became predominantly French.” Roger Collins, The Basques,
2nd. ed. (Oxford: Basil Blackwell, 1990), 211–19.
This discussion is drawn from Hult, “Gaston Paris,” 200–206; and R. Howard Bloch,
“Mieux Vaut Tard que Jamais,” Representations 36 (1991): 64– 86. The two articles in
question are Gaston Paris, “Etudes sur les romans de la table ronde,” Romania 10
(1881): 465 – 96; and “Etudes sur les romans de la table ronde: Lancelot du Lac,”
Romania 12 (1883): 459– 534.
I borrow this phrase from John Benton, “The Song of Roland and the Enculturation
of the Warrior Class,” Olifant 6 (1979): 237 – 58.
Simon Gaunt, Gender and Genre in Medieval French Literature (Cambridge:
Cambridge University Press, 1995), 38. As Sarah Kay notes, this marginalization of
women extends even to the critical tradition. No woman has edited the Roland or
written a book-length study of it; instead, female medievalists heve been drawn to
texts that problematize the heroic world and highlight its flaws. See Sarah Kay, The
Chansons de geste in the Age of Romance: Political Fictions (Oxford: Clarendon Press,
1995), 25 – 26. See also E. Jane Burns et al., “Feminism and the Discipline of Old
French Studies: Une Bele Disjointure,” in Bloch and Nichols, Medievalism and the
Modernist Temper, 225– 66, esp. 232– 37.
108 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
2/26/01
6:59 PM
Page 109
37 This formulation is Simon Gaunt’s in Gender and Genre, 23.
38 Kay draws on the following poems: Aiol, Anseïs de Cartage, Aspremont, Boeve de
Haumtone, Daurel et Beton, Elie de Saint Gille, Fierabras, Floovant, Folque de Candie,
Guibert d’Andrenas, Mainet, La Prise de Cordres et de Sebille, La Prise d’Orange, Les
Saisnes, and Le Siège de Barbastre.
39 On the semantic complexities of amer in Old French epic, see George Fenwick Jones,
The Ethos of the Song of Roland (Baltimore: Johns Hopkins University Press, 1963),
36 – 45. “Although derived from Latin amare, this verb does not always imply personal
affection or emotional attachment . . . [;] in many cases amer means ‘to keep pace
with’ or ‘to make peace with,’ or ‘to form an alliance with’” (36). On the kiss as a part
of the ritual of vassalage, see Jacques Le Goff, “The Symbolic Ritual of Vassalage,” in
Time, Work, and Culture in the Middle Ages, trans. Arthur Goldhammer (Chicago:
University of Chicago Press, 1980), 242 – 43.
40 For F. M. Warren, the “essential elements” of the motif of the “enamoured Moslem
princess” include “the release of a [Christian] prisoner by the daughter of his
[Moslem] captor; her conversion to his faith; her return with him to his native land.”
F. M. Warren, “The Enamoured Moslem Princess in Orderic Vital and the French
Epic,” PMLA 29 (1914): 346. Variations abound: some of the converts are maidens,
some married; some abandon their families, others persuade them to follow their
example.
41 See also La Prise d’Orange, ed. Régnier, 258– 59 and 281– 82. English translation is
from Guillaume d’Orange: Four Twelfth-Century Epics, trans. Joan M. Ferrante (New
York: Columbia University Press, 1974), 147.
42 Sharon Kinoshita, “The Politics of Courtly Love: La Prise d’Orange and the
Conversion of the Saracen Queen,” Romanic Review 86 (1995): 265– 87. W. W.
Comfort writes: “It was pleasing to the French, as it was to the poets of other nations,
to portray their heroes as irresistible breakers of hearts. The Saracen princess must be
fair and intelligent, worthy to become the consort of a Christian hero; she must be
passionately enamoured of his manly beauty and prowess; she must be willing to
sacrifice all— even her home and her gods —to serve her lover and to marry him.”
Comfort, “Literary Role,” 658.
43 Ann Tukey Harrison points out that each of Bramimonde’s appearances is “directly
and explicitly linked with the emperor Charles.” In “Aude and Bramimonde: Their
Importance in the Chanson de Roland,” The French Review 54 (1981): 674.
44 Comfort, “Literary Role,” 643– 44.
45 Gautier, La Chanson de Roland: texte critique accompagné d’une traduction nouvelle et
précédé d’une introduction historique (Tours: Alfred Mame et fils, 1972), xxxii; cited in
Duggan, “Franco-German Conflict,” 102; Warren, “Enamoured Moslem Princess,”
356–57.
46 Kay, Chansons de geste, 46.
47 Jenny Sharpe, Allegories of Empire: The Figure of Woman in the Colonial Text
(Minneapolis: University of Minnesota Press, 1993), 8.
48 Bernard F. Reilly, The Medieval Spains (Cambridge: Cambridge University Press,
1993), 110. Proclaimed a crusade by Pope Gelasius II at the Council of Toulouse, this
campaign drew many French nobles, including northerners like Count Rotrou of
Kinoshita / Alterity, Gender, Nation in the Roland 109
JMEMS31.1-04 Kinoshita
49
50
51
52
53
54
55
56
57
58
2/26/01
6:59 PM
Page 110
Perche along with southerners like the counts of Toulouse and Béarn and the
viscounts of Bigorre and Carcassonne. Reilly, Medieval Spains, 110; Lomax,
Reconquest, 84.
Alfonso promised that any Muslim who wished might leave the city without forfeiting
his property, that any Muslims who stayed would owe the same taxes they paid to
their previous kings, and that the Muslims might keep their great mosque. Lomax,
Reconquest, 65– 66.
Valdabrun is associated with another violation as well: “Jerusalem prist ja par traïsun, /
Si violat le temple Salomon, / Le patriarche ocist devant les funz” [He took Jerusalem
by treason: he violated the temple of Solomon, he killed the patriarch before the
fonts] (1566 –68). While ostensibly a description of Muslim conquest of Jerusalem in
637, it also recalls the First Crusaders’ conquest of Jerusalem in 1099, where the
slaughter was so indiscriminate that in the Temple of Solomon, according to
chronicler Fulcher of Chartres, they “waded in blood up to their ankles.” Peters, The
First Crusade, 209. Compare the siege of Béziers (1209) during the Albigensian
crusade, when according to legend papal legate Arnaud Amaury ordered the massacre
of everyone who had taken refuge in church, with the infamous words, “Kill them all:
God will recognize his own.”
See John A. Stranges, “The Significance of Bramimonde’s Conversion in the Song of
Roland,” Romance Notes 16 (1974 ): 190– 96.
Comfort, “Character Types,” 423.
See, for example, Robert Francis Cook, The Sense of the Song of Roland (Ithaca:
Cornell University Press, 1987 ), 111.
Peter Haidu notes, “Aude takes the position that no exchange is possible for Roland: it
is an anomalous position in the codes of her society.” His analysis, however, places
Aude exclusively in relationship to Roland, not to Bramimonde. See The Subject of
Violence, 131– 32.
In the rhymed versions of the Roland legend, collectively known as Roncevaux, the
episode of Aude is expanded to upwards of 800 lines, much of it devoted to Aude’s
sensational prophetic dreams. See Joseph J. Duggan, “L’épisode d’Aude dans la
tradition en rime de la Chanson de Roland,” in Charlemagne in the North: Proceedings
of the Twelfth International Conference of the Société Rencesvals (Edinburgh: Société
Rencesvals British Branch, 1993), 273 – 79.
Joan Hawkins, “The Social Significance of Bramimonde’s Conversion,” unpublished
essay (Berkeley, 1983). For Ann Tukey Harrison, Bramimonde is “the sole
individualized Saracen survivor, and by her baptism, arranged at Charles’ behest, she
embodies the primary theme of the chanson: the Christians are right, the pagans are
wrong.” Harrison, “Aude and Bramimonde,” 675.
For later chansons de geste, more concerned with internal traitors than external
enemies, holy war loses its ideological urgency. As Sarah Kay describes, Saracens are
wealthy, leisured, and refined: their “availability for conversion marks the possibility of
romance values being assimilated to the Frankish world, just as the desirability—and
desire — of the Saracen princesses allow for the peaceful appropriation of those
values” (Chansons de geste, 179– 80).
For Comfort, “an excellent opportunity for a romantic marriage is neglected”
110 Journal of Medieval and Early Modern Studies / 31.1 / 2001
JMEMS31.1-04 Kinoshita
59
60
61
62
2/26/01
6:59 PM
Page 111
(“Character Types,” 362). “At a later date,” he adds, “we should have had the bravest of
the French knights woo and wed her” (421– 22).
Here my reading differs from that of Kay, for whom the Saracen princess “does not
merely ventriloquize a controlling masculine fantasy: she helps to shape it, and thereby
disrupts assumed hierarchies” (Chansons de geste, 46 ).
In the Middle Ages, the successful theft of a relic provided its own justification: if a
saint allowed his or her relic to be “translated” from one institution to another, it was
because the saint preferred the new one. Patrick Geary likens the theft of relics to the
ritual kidnapping of brides. Furta Sacra: Thefts of Relics in the Central Middle Ages
(Princeton: Princeton University Press, 1978; rev. ed., 1990), 133.
Haidu, The Subject of Violence, 214 n. 4. Cook notes the lip-service paid to the
necessity for historical interpretation of medieval literary texts: “We cannot help
noticing that our Rolands, our Lancelots, our Eliducs move in a particular kind of
society, but we avoid its vitality. If we hope to keep our historical curiosity alive, we
will do well to remind ourselves that the game of politics is not abstract in medieval
French epics” (Sense of the Song of Roland, 201). Haidu’s own compelling reading
emphasizes the poem’s production, not so much of the nation as of the monarchical
state: a quarrelsome nobility whose propensity for private warfare continually contests
their feudal allegiance to the emperor is superseded by citizen-vassals like Rabel,
Guineman, and Thierry, interpellated by a “nascent State . . . requiring the creation of
new Subjects, and a new [modern] form of Subjectivity” (210).
On multiculturalism in contemporary France, see Post-Colonial Cultures in France, ed.
Alec G. Hargreaves and Mark McKinney (London: Routledge, 1997 ). Several
reviewers comment on the connection between the historical process described in
Bartlett’s Making of Europe and the recent history of “Europe.” See, for example, the
reviews by Robert Landes in The Journal of Social History 30 (1996 ): 546– 52; R. I.
Moore in History Today 44 (1994): 53 – 54; and Leo Muray in Contemporary Review
263 (1993): 272 – 73.
Kinoshita / Alterity, Gender, Nation in the Roland 111