Song of Roland
Overall Analysis and Themes
To begin to analyze The Song of Roland, we must start with its smallest units. Like other chansons de geste—this term is French for "song of deeds" and refers to the epic poems of the Middle Ages recounting the exploits of heroes like Charlemagne, Guillaume, and Girart—The Song of Roland is divided into verse paragraphs of varying length called laisses. Many of the poetic features of The Song of Roland are lost when it is translated from the dialect of Old French in which it was composed. The first laisse, in the original, may serve to demonstrate these features:
Carles li reis, nostre emperere magnes, Set anz tuz plains ad estet en Espaigne: Tresqu'en la mer cunquist la tere altaigne. N'i ad castel ki devant lui remaigne; Mur ne citet n'i est remes a fraindre, Fors Sarraguce, ki est en une muntaigne. Li reis Marsilie la tient, ki Deu nen aimet. Mahumet sert e Apollin recleimet: Nes poet guarder que mals ne l'i ateignet. AOI.
Each laisse is held together by several poetic devices. Each line consists of ten syllables, divided roughly down the middle by a pause or rest. The rhythm of the line is formed by strong stresses falling on the fourth and tenth syllables. Within a single laisse, the separate lines are linked by assonance—a partial rhyme in which the accented vowel sounds are the same but the consonants differ, as in "brave" and "vain," for instance. The vowel sound repeated through one laisse never carries on to the next. Since the poet has divided his song into laisses according to the sense and not any standard length—for instance, a new laisse will begin when one combat or speech ends and the next begins—this use of assonance reinforces the divisions of plot, of action.
The repeated "AOI," found throughout the poem, usually but not always at the end of a laisse, is something of a mystery. Nothing of the sort is found in any other chanson de geste or Old French manuscript of any sort. There are many conjectures about it—perhaps it is an obscure abbreviation of alleluia or Amen or ainsi soit it ("so be it"), perhaps it is some sort of musical notation—but in any case it certainly marks out changes of scene or atmosphere and moments of special significance in the action. It calls on us to pay particular attention when it crops up.
The Song of Roland is structured so as to be symmetrical through and through. The poem is centered around four great scenes which balance each other perfectly. At the very beginning we have Ganelon's crime; at the very end we have his punishment. Around the center, Roland's martyrdom and Charlemagne's vengeance face and mirror each other, both taking the shape of great battles, presented in a parallel order, at Roncesvals. Ganelon's successful treachery and Roland's early death temporarily set the scales of good and evil askew; the events of the rest of the poem then set them right.
The many repetitions and parallel passages of the poem contribute to the total sense of purpose and symmetry. For instance, the similarities between how the battle between Roland's rear guard and Marsilla's army and the battle between Charlemagne's and Baligant's men reinforce the poet's point that one battle is the mirror-image of the other, that Charlemagne's triumph over Baligant is perfect revenge for the Saracen ambush. The order in which the two battles are presented is the same, as it must be for them to balance each other properly; first there is the inventory of the two opposing forces as they assemble themselves, then, when they meet on the field, the threats and boasts and first blows. Each one-on-one combat, besides the most remarkable and important ones, such as that between Charlemagne and Baligant, takes up one laisse, and all are described in the same language. Comparing the various rather gory ways in which the warriors kill each other, one sees immediately that each description is a slight variation on all the others. Ideally, the effect of such repetition is a sense of ceremonious consistency and rhythm.
Rather than running along at a consistent pace, the narrative consists of certain scenes where time is slowed down so much that it almost stands still, suspending the noble and the wicked gestures of the characters mid-air, with bits of quick summary providing the connection from one tableau to the next. This rhythm is particularly clear and easy to pick out toward the beginning of the poem, in the first fifty or so laisses. After some quick exposition in the first laisse, we get the council of Marsilla presented as if it were a drama. The poet summarizes nothing; he describes the stage of the action, the "terrace of blue marble" (2.12) and then gives us the speeches of Marsilla's advisors in full. The story is conveyed in this section by dialogue, not by running commentary. Then, after another quick laisse of summary, telling how Marsilla's messengers rode out to Charles's camp, we go back to the same slow, dramatic mode of presentation that was used for Marsilla's council for the conversation between Marsilla's envoys and Charlemagne. This alternating, fast-slow-fast-slow rhythm, interspersing quick pieces of narrative between long dramatic scenes at regular intervals, is characteristic.
Within each laisse, each sentence and phrase stands separate, on its own. Similarly, no grammatical connection is drawn between one laisse and the next. The reader must draw the connection between one element to the next on his own, for the author does not make the relation between the separate elements clear, but instead simply sets them side by side, without conjunctions. This technique is known as parataxis, which means "a placing side by side" in Greek. To see more clearly what this is, one might take a quick look at laisse 177, for instance, a particularly striking example. There is no connective tissue: "Roland is dead, his soul with God in Heaven. / The emperor arrives at Roncesvals" (177.2397-2398). The corollaries of this lack of relation between phrases include a propensity towards long lists and a lack of simile, aside from certain highly stylized and conventionalized comparisons which are repeated often—beards, for instance, are very frequently "white as April flowers." The elements are strung together like beads, one after another.
It is thought the The Song of Roland, like other medieval chansons de geste, was passed on orally, sung by wandering performers known as jongleurs at feasts and festivals, before it was ever written down. The written epic that we now have, based on a manuscript version set down by a medieval scribe, bears the marks of its origin in the performances of the jongleurs in its narration. The voice that tells the story is the voice of the jongleur. He does not take on the character of one who was there, nor does he take on any kind of neutral, third-person omniscience of observation. He tells the story as a story-teller.
While the events recounted in The Song of Roland are almost all myths and inventions, the jongleurs' medieval audiences accepted them as historical truth. Because of this, and because the heroic deeds described took place in what was the distant past for even those long-ago listeners (the centuries that separated the audience from the figures they heard about made those figures seem all the more grand and glorious), the jongleur could not take on the point of view of an eye-witness of the events he sings about. If he did, the whole story told would lose credibility in the face of the obvious impossibility of the jongleur having seen himself anything that he was describing. Thus, the effect that the narration aims for and achieves is a vividness without immediacy. The characters and events are brightly painted, to be sure, but there is none of the you-are-there feeling that one usually expects nowadays from a well-told story. Different eras want different effects from their literature.
The narrator does not pretend to know what he tells because he was there; he instead implies that he has his knowledge from chronicles and tales, which he alludes to in order to gain the best effect of credibility for the story he tells: for instance, he says of Olivier, Roland, and Turpin fighting at Roncesvals that "The number that they killed can be determined; / it is written in the documents and notes: / the Chronicle says better than four thousand" (127.1683-1685). It is probable that many of the historical chronicles he speaks of are as much his own inventions as many of the events he recounts, but this does not hinder his allusions to them from creating the desired effect of a past both mythic and historical.
That the telling of The Song of Roland does not aim for surprise or suspense is a result of the way in which it, like other chansons de geste, was passed about orally, told again and again, varied but still recognizable in each new performance. The narrator assumes that his audience is already thoroughly familiar with the story he is telling them; he knows they have already heard it plenty of times, but that they enjoy hearing it again. The interest of the audience is not bound up in the question of what's going to happen next; the listeners already know that Ganelon will betray Roland but that Charlemagne will avenge him in the end. Familiarity was part of the story's charm for medieval listeners. And so the element of surprise is absent, and suspense is not cultivated; in the very first laisse, we are told that Marsilla will be clobbered by Charlemagne's men, and Ganelon is called a traitor before he makes a single treacherous move.
The story of The Song of Roland is essentially the very old, inexhaustible story of the struggle between good and evil. The sides are as clearly marked as they come: the Christian Franks, led by Charlemagne, represent the good and the will of God, while the Muslim Saracens, led by Marsilla and Baligant, represent the purest evil. The good, in the medieval world-view, will always triumph in the end; this is the inevitable result of a good and all-powerful God who takes a real interest in human events. The characters clearly have free will; God didn't step down and stop Ganelon's dastardly plot before it took lethal effect. All the same, God intervenes fairly often to make sure that good comes out in the end; thus, for instance, Thierry miraculously beats Pinabel in a duel. Although he is the weaker man, he is just and right, and God makes sure he wins.
The presentation of the other side, of the Muslims, poses a problem for the poet: he must make them unquestionably evil and base, clearly less noble, less manly, and less brave than the Christians, but nonetheless a worthy enemy. Accuracy, of course, figures in not at all. His solution is to make the Saracens the reverse image, the evil twins, of the Christians, both opposite and identical. For instance, the Muslims of The Song of Roland worship Mohammed, Termagant, and Apollo, mirroring the form of the Christian Trinity but signifying the most villainous idolatry. Never mind that the Muslims are actually much more strictly monotheistic than Christians; the Christians of the early Middle Ages saw in Islam, which they knew next to nothing about, just another form of paganism, as signified by the notion that Muslims worshipped the ancient Greek god Apollo. Likewise, there are twelve Saracen peers to match the twelve Frankish peers on the battlefield; the opposing armies organize themselves into the same forms, but the Christians fight more nobly. At times the need to make the Muslims a worthwhile adversary, to make the struggle more compelling and the eventual Christian victory more satisfying, comes to the fore. The description of Baligant in particular is shaped by the need to make him worthy of fighting Charlemagne. And so he is praised—"God, what a lord, if he were but a Christian!" (228.3164). The poet also enjoys spicing his depiction of the Saracens with a touch of the exotic; the description, while Baligant assembles his forces, of curious peoples from far-away lands, with all their peculiarities, is accomplished with an eye for the colorful detail.
The depiction of the larger-than-life heroism of the Christians is organized around the idea of vassalage. The obedience that a knight owes his lord is a model for the obedience that a man owes the Lord, and vice versa. The relation between man and God was truly feudal. And thus it made as much sense to go to war for God as it did for a local baron, only it was of course much holier. One can think of feudal society as a pyramid scheme; the peasants offered the local seigneur allegiance, fealty, and hard work in return for protection and so on up through the various ranks of nobility—the lesser noble owed the greater noble fealty in return for protection, and the greater noble had the same deal going with a still greater noble—all the way up to the king and then to God at the apex. This total synthesis of economics, military strategy, and religion into the same basic feudal pattern, in which each aspect justified and served as a model for the others, is brilliantly summed up by Roland's dying gesture. He lifts up to the heavens and to God his right-hand glove—the gesture of a vassal's fealty to his lord, repeated throughout the poem in the more ordinary, earth-bound context—and Saint Gabriel comes down to accept it (176.2389-2390). He dies the true vassal of the Lord, which he demonstrates by the same gestures he would make to an earthly lord, and thus is taken to Paradise.
Some have tried to fit the Roncesvals massacre into the pattern of the classical tragedy. Up to a point, this seems to work. Roland seems to be in the situation of the tragic hero; it is his pride that causes him not to blow the oliphant and call back Charlemagne's troops before it is too late. And he dies of a self- inflicted wound—not by a well-aimed blow from a Saracen, but by his burst temples, the result of his blowing his horn so hard. But the glory of Roland's death contradicts such an interpretation. He dies a martyr. The significance seems to lie less in his flaws than in how his perfect fealty to the Lord, as shown in particular by his recognition of the absoluteness of the values that the Franks are fighting for in Spain and his refusal to compromise with the Saracens, makes his flaws irrelevant. The passion with which he fights for Christendom saves him. We see the great value that the author of The Song of Roland ascribes to passion by how he describes the weeping, the moaning, even the fainting away of his bravest characters. It is their capacity for strong emotions, not self-discipline or stoic virtues or anything of that sort, that makes them great. Roland's closest comrade, Olivier, is a fine and noble man, and he does not make the mistakes of pride that Roland does—"Roland is bold, Olivier is wise" (87.1093)—but he lacks the great passion, and thus the great heroism and the great reward, of Roland, who goes in the end beyond wisdom.
Readers' Notes allow users to add their own analysis and insights to our SparkNotes—and to discuss those ideas with one another. Have a novel take or think we left something out? Add a Readers' Note!