The first stanza of the Song of Roland turns readers into admiring imperial subjects: “Charles the king, our great emperor” (l. 1).Footnote 1 It seems that “our king” has been successfully conquering Spain for seven years. As we learn of his success, though, we are also invited to doubt him: “There is no castle which can resist him / […] Except for Saragossa” (ll. 4–6). Having been cast into uncertainty over “our” future in Spain, we might become further puzzled by the narrator’s observation that the ruler of Saragossa, King Marsile, “serves Muhammad and calls upon Apollo” (l. 8). In an instant, the narrator looks rather ignorant because anyone who knows about Muhammad might also be expected to know that Apollo belongs to a completely different religious system. What else does the narrator not know? The closing affirmation that Marsile “cannot prevent disaster from overtaking him” (l. 9) is supposed to reassure us of Charles’s impending victory. And indeed, the resistant city falls to him before the end of the narrative—but not without exposing the profound vulnerability of “our” empire.
This short passage shows how even an old text associated with implacable imperial ideology opens itself to discussion of vital issues in postcolonial literary pedagogy. Of course, many other more recent and more provocative texts do, too. By engaging sources like the Song of Roland alongside more obviously postcolonial literature, students gain a sense of the longer histories of issues they often consider uniquely contemporary. The fact that the “Middle Ages aren’t old” means that we can still connect with them in human ways—and also that their regressive and violent dimensions have not been left behind by the “progress” of modernity. Students can thus learn to understand historical periods themselves as both colonial effects and targets of postcolonial critique. They can develop tactics for identifying the internal contradictions of hegemonic claims and political polarizations.
For this essay, I will sketch out some of the techniques I used to engage nonspecialist students with the Song of Roland—techniques that I hope nonspecialist instructors can use to expand their engagement with unfamiliar literatures in the classroom, especially medieval literatures. Integrating longer histories of colonialism, hybridity, and cosmopolitanism into postcolonial literary studies sharpens our understanding of discrete historical and aesthetic forms. In the end, I believe that reading and teaching texts associated with imperial ambitions keeps students attuned to the ongoing circulation of divisive ideologies, especially to the stealthy work that medievalizing discourse does every day in the news.
Refracted Reading
My discussion here stems largely from my most recent course, Global Medievalism. We explored the literature and politics of medievalism since 1990. Our questions addressed the intersections of history, globalization, and Orientalism. How has the idea of the Middle Ages been deployed in policy debates? How does it provide justification for both religious violence and pluralistic dialogue? We read a fair amount of international relations theory and social history along with a handful of historical novels set in the medieval Mediterranean—including Tariq Ali’s Book of Saladin (1998) and Radwa Ashour’s Granada (1994, trans. 2003). The first literature we read, though, was the Song of Roland.Footnote 2
This epic narrative was written down in the twelfth century following a phase of oral development. In the nineteenth century, it became a symbol of nationalist imperialism in France—proof of an ancient and illustrious history of world domination. Today, it remains part of the standard curriculum of French literary history. It is also frequently taught in history courses to illustrate the Christian ideology of crusade. My goal in assigning the Song of Roland was twofold: to illustrate how literature voices hegemonic claims that forestall intercultural understanding and also to demonstrate how those claims rely on discourse (and so actually create what they purport to describe).
The students in this course were all undergraduates, most of whom sought required credits in international studies; they were not particularly drawn to literature or postcolonial theory. They represented a broad range of majors—from engineering to political science to sociology to English literature; some were in their first year, others in their last. In this context, the course endeavored to bring the medieval into the postcolonial, the literary into other disciplines, and the specialized into the public sphere.
To prepare for reading the Song of Roland, we read several types of sources that emphasize oppositions between East and West, Christianity and Islam: historical overviews of the Crusades (eleventh to fourteenth centuries), selections from Edward Said’s Orientalism (1978), and articles by Bernard Lewis and Samuel Huntington from the 1990s articulating the “Clash of Civilizations” model. The Song of Roland certainly offers ample evidence of the long history of this rhetoric. It is perhaps most famous for the warrior Roland’s depressingly sharp declaration of cultural supremacy: “Christians are right, pagans are wrong”; it also depicts antiblack racism in ways familiar to many modern readers (ll. 1474–1479, 1932–1934). At the same time, the narrative is not monolithic in its articulations of Eurocentric superiority. Subtle evidence that the binaries don’t entirely hold appears in the figures of the blond Arab (Marsile’s son Jurfaleu, l. 2702) and the black-haired Frank (Charles’s champion Thierry, l. 3821).Footnote 3 The oppositional slogan, in other words, turns out to be just that—a slogan, not a description.
In order to prompt a pluralistic engagement with the text—what I have called elsewhere “refracted reading”Footnote 4—I assigned a reading journal where students recorded two kinds of observations: evidence of Orientalism and civilizational clash; evidence that these ideas are not entirely dominant (such as resistance and identification across differences). At first, students found the latter harder to identify than the former, but with practice they became adept at discerning the cracks in the edifice of hegemony. At that point, the Song of Roland began to enliven students to diverse opportunities for challenging ideological claims and modes of “othering.” As they learned to recognize discourse as a characteristic of description, they saw how it operates in daily language as well as literature. In the literature classroom, discourse analysis is the lever to cultural and political critique—and ideally, the opening to pluralistic conceptions of society. This simple assignment generated plenty of discussion points over several days.
Students collected much evidence of Orientalism and Christian imperialism, from plot points to narrative structure to descriptive details. These observations help explain the Song of Roland’s appeal to French imperial nationalism.Footnote 5 Much of the oppositional discourse substantiates Roland’s claim of Christian “right” against pagan “wrong.” The contrast is evident from the first lines, where Marsile (“who does not love God” [l. 7]) and his men are called “pagan” (l. 22), and they themselves debate conversion to Christianity (l. 38). Students pointed out immediately that the lack of capitalization for pagan indicated lesser worth in the poem’s moral economy. Although capitalization is not a feature of the medieval source, the graphic forms have a legitimate impact on modern readers, and they are a perfectly valid point of entry into issues that clearly occupy the narrative. Students then puzzled over the connotations of the term pagan: Is it derogatory or merely descriptive (as when Baligant refers to his own men as pagan, l. 3136)? Is Saracen its synonym or subtly distinct? Students must also grapple with the (common) assumption that both terms mean Arab Muslim: What traces of historic Islam are actually visible within the Christianized frame? How can readers distinguish between representation and their own projections?
The theme of conversion reinforces the normative idea of Christianity throughout the narrative. Several times, pagans must choose between death and conversion (ll. 101–102, 3669–3674, 3978–3987). The result is a picture of Christian hegemony, a Spain in which all other faiths have been destroyed. Indeed, several scenes of divine intervention clearly demonstrate that the Franks are favored by the Christian God: the angel Gabriel takes Roland’s proffered glove when he dies and accompanies his soul to heaven (ll. 2389–2396), God stops the sun so that Charles has time to catch up to his fleeing enemy (ll. 2448–2459), and at the end Gabriel comes again to send Charles to fight further battles (ll. 3993–3998). These dramatic moments crystallize into a recognizable “crusading” mentality that students connect to the motivating paradigms of US and global policy since the 1990s—clash of civilizations and war on terror.
Despite the absolutist rhetoric, students can discern alternate histories. The first time we meet “our great emperor,” we learn that there is nothing he can do to avoid Marsile’s deceit (l. 95). In other words, he lacks something—intellectual capacity, political authority, or maybe just willpower. His ranks, moreover, harbor a traitor named Ganelon—identified as he “who committed the act of treason” (l. 178) even before Charles considers Marsile’s peace offering. The ensuing discussion reveals contradictory opinions, conflicting logics, and ultimately threats of violence between Ganelon and Roland. Clearly, Ganelon and Marsile both resent Roland—which makes Ganelon a traitor, but also shows that Frankish loyalty has its limits. This episode sets the frame for later ones in which Charles is alternately helpless, irrational, and vengeful. The culmination of this contradictory portrait comes with Ganelon’s trial after Roland’s death: Charles seeks a guilty verdict, not the truth; he refuses peace in favor of vengeance. Indeed, Charles looks more than ever like Marsile, readily sacrificing his own relatives: Charles judges death to thirty relatives of Ganelon, his own brother-in-law, just as Marsile sent his own son to certain death as a hostage (ll. 147–150, 3947–3959). At the end, Charles, like Marsile at the beginning, is weary of war. Charles’s very inconsistency casts doubt on the righteousness of his actions and thus on the moral superiority of the Christian polity.
On the other side, the narrative does not thoroughly demonize the pagans. They possess many items, for example, that the Franks clearly covet. The promiscuous sharing of material culture raises many questions. How should we interpret, for example, the wrapping of Roland’s body in Oriental silk (l. 2973)? Even without research, students can trace the symbolic impacts of luxury goods (silk, fur, ivory, swords, animals); with a little research, they can add historical texture.Footnote 6 Meanwhile, Saracen warriors are regularly described in positive terms as “worthy” and “valiant.” In one particularly deft passage, the narrator even calls upon readers’ emotional sympathy in the midst of battle: “You would have seen so much human grief there, / So many men dead, wounded, soaked in blood, / Men piled upon men, face down or on their backs. / The Saracens can endure it no longer” (ll. 1655–1658). Evoking identification through the second person address to “you,” this description humanizes the Saracens who grieve their fallen comrades. In this way, readers are encouraged to see more than one side to the battle of Roncevaux.
The most intriguing result of the reading journal was how students chose some of the same passages or traits as evidence for opposite conclusions: almost every example of Orientalism can also disrupt “othering.” The message of superior Christianity implied in conversion, for example, also includes the provocative message that the denigrated “other” can become just like “us”: the very fact that anyone can cross the line of difference undermines the line itself. And Gabriel, who intervenes so decisively, belongs as much to Islam as to Christianity. The very possibility of these dualities reflects a core insight of postcolonial literary theory: the inherent instability of the colonial enterprise, which augurs its dissolution within its very enunciation (which of course is not to say that colonial power is not also brutally and physically destructive, with durable effects even after the end of formal colonialist regimes). Instructors can get a lot of mileage out of a basic definition of the palimpsest.
One student, Ziad al-Shamsie, identified a powerful example in which difference converges on similarity in Archbishop Turpin’s speech on martyrdom (ll. 1124–1138). Addressing the Franks before battle, Turpin reminds them of their obligations to their king and their God. Couched as a defense of the Christian faith, Turpin promises: “If you die, you will be blessed martyrs” (l. 1134). He concludes with a trope closely identified with Christian crusade: “As penance [for their confessed sins] he orders them to strike” (l. 1138). For this student, the passage sounded every word like a description of Islamic martyrdom. He pointed out that fleeing battle is one of the unforgiveable major sins (Kaba’ir) in Islam; Christian and Muslim martyrs are both promised a “place in paradise” (l. 1135). In this unexpected convergence, students again connected the Song of Roland to current global issues. How do Roland’s motives—pride, vengeance—illuminate the motives of contemporaries who seek martyrdom through violence? Can insights that bridge the rhetoric of absolute differences eventually short circuit the cycle of retributive violence? Is the rivalry “eternal” or can it be ended by unraveling some of its origins? In the Song of Roland, “medieval pride” breeds martyrs and traitors; the same may be true in the twenty-first century (al-Shamsie).
The narrative’s most destabilizing characteristic is the persistent use of parallelisms. These mirroring effects translate the “menace of mimicry” through almost-but-not-quite repetitions that weaken the very foundations of Frankish ideology.Footnote 7 Descriptive vocabulary, shared values, and formal repetitions all forge similarities that break down binaries, introduce multiplicities, and erode unities. The formulaic repetition of stanzas, known in French as laisses similaires, makes parallelism a defining characteristic of the narrative. At regular intervals, two or three stanzas in a row recount the same event with slight variations. We notice first the exact repetition of a first line, and then scrutinize for the moment of divergence. Readers’ expectations are alternately confirmed and contradicted. In some sequences, time moves at different speeds; in others, the narrator offers differing interpretations of the events; sometimes, the sequence moves from general to specific (such as when a vague threat becomes a specific commitment [ll. 563–595]).
All this similarity was somewhat disorienting for students, who were prepared for a “clash of civilizations.” Saracens and Franks are both flawed, both admired. Parallels across and within each group abound: no one seems to need a translator despite the multiple origins of each army; large numbers of people speak in unison; everyone has the same brightly colored equipment and luxury goods; both sides follow the logic of feudal service; kinship structures seem to work the same. Do these similarities betray the narrator’s innocent lack of knowledge, arrogant erasure of distinctions, or cosmopolitan acceptance of commonalities? The Song of Roland is saturated with these kinds of self-canceling parallelisms. As one student observed, this technique shifts ethical responsibility on to readers: we can’t just follow the narrator or any one authoritative character.
Similar and even identical descriptions of Franks and Saracens create contradictory effects. Throughout the battle scenes, the two sides are repeatedly described together as if acting in exactly the same ways—equally fearsome, effective, and deadly (“Franks and pagans strike awesome blows” [l. 1397]). Blancandrin is introduced as “A most valiant and worthy knight” (l. 25); Grandonie as “a valiant and worthy man” (l. 1636). These kinds of descriptions enhance Christian valor by making victory an achievement over worthy opponents. They also suggest that the narrator has no conception other than Christian feudalism. At the same time, they recognize valiant qualities irrespective of religious or cultural differences. Even when the narrator distinguishes marshal qualities from moral ones, such as noting of a Saracen emir “Had he been a Christian, he would have been a worthy baron” (l. 899), religion becomes a relative rather than an absolute quality—one aspect of identity rather than the only determinant of value.
The epic technique of simile also functions through equivocation. On the one hand, the rhetorical figure reinforces true qualities: Charles “tugs at his beard like a man beset with grief” (l. 2414) because he is grief-stricken; he “sleeps like a weary man” (l. 2525) because he is tired. Elsewhere, however, similes seem to distance Saracens from the qualities mentioned (as one student noted): when Marsile is described as “like a true baron” (l. 1889) or other pagans ride forth “like valiant men” (l. 3264), they seem to appear as something they are not. They may look valiant, but actually they are traitors (as they are also described). Within this double-dealing discourse, should we cast some doubt on the seemingly positive description of Roland as speaking “like a true knight” (l. 752)?
Expert Amateurs
For students coming to the Song of Roland with little to no knowledge of medieval European literature or history, the concept of the “expert amateur” defines what it means to be a good learner—perhaps the most fundamental outcome teachers can hope to achieve with their students. I tell my students that experts are just amateurs who have studied a lot, so we shouldn’t be afraid to plunge in using whatever we do know and build from there. I have practiced two different ways of setting up a course to instill this ethos in the classroom and break down students’ hesitations to approach “old” books. In the case of Global Medievalism, I drew extensive inspiration from Therese Huston’s Teaching What You Don’t Know (Harvard, 2009). Concretely, I adapted her model of a background survey that asks students to assess their familiarity with about twenty terms, concepts, and facts relevant to the course (Huston 179–182). The scale from 1–5 defines degrees of prior knowledge from “never heard of it” to “could readily explain details to the class.” I was able to share with students that I myself registered a number of low scores before beginning my research for the course and that the purpose of the course was not to reach 5 on every survey item. Instead, the survey would enable us to draw on pockets of expertise within our classroom community and open space for each us to express our ignorance, thereby increasing our collective learning opportunities. Students completed the survey in about ten minutes.
Some sample items relevant to the Song of Roland (which wasn’t itself a survey item):
Crusades
1. Never heard of them
2. Have heard of them but not sure what they are
3. Learned about them once but can’t recall much now
4. Can recall some general ideas and a few facts
5. Can explain some characteristics, dates, and other details
Islam
1. Don’t know much about it
2. Have heard references but I haven’t paid much attention
3. Familiar with some characteristics but there’s probably a lot that I don’t know
4. Can explain general characteristics and a few details
5. Can explain details of history and characteristics
Feudalism
1. Never heard of it
2. Have heard of it but not sure what it is
3. Learned about it once but can’t recall much now
4. Can recall some general ideas and a few facts
5. Can explain some characteristics, dates, and other details
Orientalism
1. Never heard of it
2. Have heard of it but not sure what it is
3. Learned about it once but can’t recall much now
4. Can recall some general characteristics and a few examples
5. Have studied many examples and could explain them to the class
Cosmopolitanism
1. Never heard of it
2. Know the word but not sure what to discuss beyond a dictionary definition
3. Learned about it once but can’t recall much now
4. Can recall general issues and a few examples
5. Have studied debates around the definition and could explain them to the class
Throughout the course, I referred back to the survey, reminding students when they had indicated 4 or 5 for a topic, asking them to contribute their background knowledge while inviting those who had less background to ask them questions. It was also extremely helpful to me to know the distribution of familiarity as I designed approaches to each set of readings. Many students had not heard of feudalism (the bonds of service and protection that define social and political relationships in medieval Europe) whereas several were well versed in crusade history and two had extensive background in Islam. Because Orientalism turned out to be relatively unfamiliar for students—and highly familiar for me—I used the medieval epic to reinforce students’ grasp of the theory and to practice using textual analysis to illustrate, modify, and critique theory. Cosmopolitanism elicited little reaction at the beginning of the term but became a major theme of the course—with students using medieval sources such as the Song of Roland to develop nuanced definitions in dialogue with contemporary theorists. Ultimately, cosmopolitanism provided a flexible and complex framework for assessing cultural encounters in a wide variety of medieval and modern narratives.
My second approach to teaching the Song of Roland in the spirit of the “expert amateur” derives from Virginia Scott’s approach to literature as an applied linguist—that is, as a nonspecialist.Footnote 8 This approach can be valuable for instructors familiar with modern or contemporary literature who wish to try the adventure of including historical texts such as the Song of Roland—or really, any literature from outside one’s zone of familiarity (defined by period, genre, culture, language, or any other variable). Instead of focusing on distance and difference, Scott asks students to identify both ahistorically and aculturally. In other words, what are students’ current topical concerns—personal, social, political? What are they interested in or worried about? This survey of “current preoccupations” also forges social sharing among students and leverages their preexisting concerns into the literary experience. Instructors can give students a menu of options, along with the opportunity to define their own. Some popular examples: love, faith, violence, gender, freedom, environmentalism, and so forth. These topics then structure a first reading of each new text. Once the familiar themes have been discussed from a presentist or individualist perspective, the discussion can move to historicize or otherwise contextualize the topic. Once students pursue observations related to similarities between themselves and literary texts, they are more ready to observe and analyze differences. Over time, they can discuss the significance of the greater or lesser presence of their preferred themes in different literatures. In most cases, students identify personal concerns that are broadly humanistic, and so prevalent in almost any literature. Yet the fact that they begin their readings from this personal standpoint diminishes the barriers that can stand between them and historically or culturally distant literatures. The Song of Roland, then, becomes meaningful first because students find there what they already find meaningful. In this way, the poem can be pried loose from the nationalist framing that often overdetermines scholarship. We do not need to accept this framing, or even know very much about it. Instead, we can connect the poem directly to our contemporary concerns.
The preoccupations survey and the background survey can both become tools of community building in the most optimistic spirit of pedagogical pluralism, breaking down assumptions of what students do and don’t know while drawing on expertise students might not otherwise reveal (religious education, etc.). Both are pivotal to decolonizing and culturally alert pedagogies.Footnote 9 The background survey in particular is essential for interdisciplinary courses—and for the inherently multicultural dimension of teaching postcolonial literature (both the content of the reading materials and the diversity of the students who enroll). I have found that it helps students to invest in sharing their expertise and move beyond the “fake neutrality” that many maintain in order to perform “objectivity” in the classroom and not “get in trouble” for sharing views that differ from their peers or professor. The surveys can break down the anxieties that students and instructors alike can bring to reading “old” literature. I candidly shared with students my own low scores on some of the terms on the background survey—and also why I nonetheless felt entirely competent to engage with them on these topics. Just like learning a new language, we can accomplish a lot by building on what we do know rather than obsessing over what we don’t. We can read with feeling before reading with history, an approach that does not require specialization but may inspire it.
I’m also affirming the rigorous validity of reading in translation. Medieval and postcolonial literary studies share this very tangible methodological issue. When I began teaching Comparative Literature a few years ago, I started wondering: How can we teach serious literary inquiry—attentive to language and the nuances of representation—with translated texts?Footnote 10 Too much attention to source languages delegitimizes every interpretation before it can be made (most discouraging for students), and not enough attention seems to license all manner of irresponsible pillaging in the misunderstood foreign (191). Making translation viable requires a supple pedagogy that shifts focus away from the conditions of production (author, scribe) and toward the conditions of reception (students’ reading experience). Instead of developing “why” questions, students gain traction from “how” questions. And although I caution students not to make much of specific word choices as evidence for what the author was trying to communicate, translated words are fair game for explaining impact on a reader. In the Song of Roland, students picked up the word awesome as a multivalent description: “Franks and pagans strike awesome blows” (l. 1397); “The battle is awesome” (ll. 1412, 1653, 1661). The adjective implies both fear and admiration, a combination that effectively translates something fundamentally true about the narrative regardless of the source word or words that may have inspired awesome. The term’s double connotation, moreover, translates the poem’s persistent equivocations. I intentionally did not bring in Old French vocabulary for comparison, so as to keep focus on how meaningful students’ amateur expertise could be on its own terms.
In preparation for a writing assignment, I asked students to distinguish between two types of questions about the Song of Roland: ones that require research or expertise to answer and ones that could be answered by spending more time studying the text. Assessing the nature of questions is essential for reinforcing the idea of the amateur expert and for instructors to integrate unfamiliar texts in their courses with confidence. Any of us trained in literary analysis can bring those skills to text-based questions, which often have more than one compelling answer. And any of us trained in research can ferret out answers to cultural and historical questions, given the time and the motivation. Students can learn to assess their questions and so not address questions to the text that it can’t answer, and not expect answers to questions they don’t have the time or motivation to pursue. The ability to classify questions also enables students to develop more meaningful textual analyses and avoid common pitfalls of argumentation. In many cases, the difference between an answerable and unanswerable question requires only a pivot of perspective. What does it mean when Ganelon drops Charles’s glove while accepting the mission to speak with Marsile (l. 333)? The Franks ask the same question (l. 334), and so we can look for answers via textual analysis rather than via cultural history. We don’t need an expert in feudalism to get started.
Finally, I asked: “What is the value of reading the Song of Roland today?”Footnote 11 For some students, the opportunity to expand the discourse on Orientalism beyond modernity proved compelling. For others, the epic helps understand how people today deploy charges of medievalism against the Middle East and Islam. In other words, it points out the regressive nature of accusations of backwardness precisely because they are so similar to the modes of “othering” in an actual medieval source. “The UN needs to read this book,” offered one student, Gabriel Corso; another found that reading the poem could help downplay the “apocalyptic” sensationalism of the clash of civilizations; Ziad al-Shamsie observed in conclusion: “I still can’t believe how relevant that era is.” It seemed that we could all agree that by developing the ability to understand prior history, we might also discern how it is still affecting us. The Song of Roland helps see the operations of ideology all around us.Footnote 12 Analyzing a text that seems to offer few positive models for creative thought and little ideological space for free thinking can teach us to look for—and find—openings in other seemingly monolithic discourses. The fact that we also unearthed signs of resistance within the poem to some of its own messages made the quest for radical optimism all the more promising. In this postcolonial frame, the Song of Roland appears as an explicit node in Édouard Glissant’s “poetics of relation.” For Glissant, the epic tells a story of defeat and exile, not triumph; foundational literary forms do not function in isolation from the diversities that would be their undoing.Footnote 13
Conclusion
The Middle Ages aren’t old, and that’s both good and bad, an inspiration and a warning. We are surrounded, wherever we may be, by oppositional discourse—binary divisions of the world meant to establish who is modern and who isn’t, who deserves human rights and who doesn’t. One of the loftiest purposes of literary study can be teaching students how to recognize these discourses as discourse. Literature can show us that hegemony is usually aspirational, rarely achieved. Historical literature, including modern novels set in distant times, can help address the “historical amnesia” that Ali Behdad has diagnosed as an obstacle to democratic historicism.Footnote 14 The Song of Roland, as described here, bears witness to cultural complexities that illuminate the long and enduring histories of colonial imperialism: hegemony and its resistance; unequal power relations and the quest for mutual understanding; binary hierarchies and their destabilization. Within these seams, we can find margins for hope in a polarized world, openings for human understanding amid all the pressures for destructive closure.
