1Dear Mr Administrator,
Dear colleagues,

2Deep forests, enchanted castles, monsters, damsels in distress, stout-hearted heroes and boundless love. The literature of the Middle Ages has everything it takes to appeal to the imagination of children and teenagers. However, can it do more than that? And does it even still have this appeal? Anatole France claims that, to the loves of Abeille des Clarides and Georges de Blanchelande and to the magnanimity of the King of the Dwarves, his pretty little neighbour prefers “scientific novels” where she can read the description of the cuttlefish, Sepia officinalis. To be sure, Abeille, whose prologue I evoke here, is not a real story of the Middle Ages; the prologue yields to complacent easiness and the way the author imagines medieval legends and positive science is indeed very old fashioned. Yet, if the only thrilling novel is that of the natural sciences on the move, what is left to us who claim to revive lost dreams? If this be the case, our colleagues the real scholars will truly have taken everything away from us, and it is only through excessive generosity or indulgence that they still grant us a place amongst them.

3At least this place has seniority going for it. In 1853 a Chair of French Language and Literature of the Middle Ages was created at the Collège de France for Paulin Paris. It kept the same name for his son, Gaston Paris, then for Joseph Bédier, and after having been given the title of History of the French Vocabulary for Mario Roques, it went back to its original name for Félix Lecoy. From 1876 to 1906, moreover, the Chair of Languages and Literatures of Southern Europe, which Mr Harald Weinrich presently holds, was held by Paul Meyer, most of whose work actually concerned French and Occitan philology. Finally, there was a Chair of Latin Literature of the Middle Ages from 1925 to 1954, held by Edmond Faral. Additionally, medieval studies took on a predominant place in the activities of the Chair of Celtic Languages and Literatures held from 1882 to 1930 first by Henry d’Arbois de Jubainville and then by Joseph Loth, and in those of the Chair of Languages and Literatures of Germanic Origin, especially between 1934 and 1956, under the impetus of Ernest Tonnelat and Fernand Mossé. Medievalists of other disciplines have also expressed an interest in literature, such as Étienne Gilson and those who teach Medieval History, in particular Mr Georges Duby and presently, Mr Pierre Toubert. To sum up, if we consider only the fields of French, Occitan and Latin, then we can say that the philology and literature of the Middle Ages were taught uninterruptedly at the Collège de France from 1853 to 1974, and sometimes simultaneously within several Chairs.

4And by what masters! Gaston Paris, the founding father of our studies, of whom Francesco Novati said that he could render the most complex and driest subject clear and attractive. His alter ego and conscience, the strict Paul Meyer, who used to “mark each mistake with a piercing stab of his nail”, as Anatole France – always he – described him through the mouth of his Sylvestre Bonnard – although Meyer, fortified by this rigour, also provided decisive expertise in favour of Dreyfus, to which Proust pays a magnificent tribute in Jean Santeuil. The brilliant, appealing and subtle Bédier, whose adaptation of Tristan and Iseult – done half in play – will have been one of the significant literary works of the century that is drawing to an end. Edmond Faral, who was an administrator of this institution. The ironic and kindly Mr Lecoy, with his unlimited knowledge, who taught us to take the greatest care in whatever we were doing, without however exaggerating its importance, for therein lies elegance and reason. Yes, the teaching of the Literatures of Medieval France, which is being revived today, can claim to belong to a long tradition – even an overwhelming one if one considers the quality of all those who have dedicated themselves to it in the past.

5This long tradition would certainly not be a reason in itself to keep or restore the subject, had it not something to do with the subject itself. All of the associations between the past and literature, all of the signs that point towards an essential link between the notion of literature and a feeling for the past, are crystallized in medieval literature. The curiosity that medieval literature has aroused since it was rediscovered at the dawn of Romanticism presupposes such associations. The very forms of this literature bear indications of them. They encourage us to consider jointly the interest of modern times in the medieval past and the signs of the past with which the Middle Ages marked its own literature. Even more, they invite us to seek in the relationship with the past a defining criterion for literature, a most necessary task with reference to a time when words are not understood in their modern sense, and there is no guarantee that a corresponding notion exists. The best reason to continue with this hundred-and-fifty-year-old teaching is that its object may not even exist.

6Let us, however, pretend that it does exist. We have the right to do so. The word litteratura refers either to grammar or to the commented reading of authors and the knowledge that it provides, but not to all works taken as a whole. Vernacular languages do not have any generic term for literary activity or literary works, yet in the Middle Ages an awareness of such an activity and of a corpus of works did exist. On the Latin side, this is demonstrated by the word litterae in the sense of literary culture, the continuous thought applied to the nature and legitimacy of the belles-lettres, the permanent canon of authors since Antiquity, and the systematic practice of borrowing and imitating. Concerning vernacular production, where the situation is more ambiguous, a collection of poems, also canonical in its own way, constitutes quite early on the forgotten repertoire of the jugglers. At the beginning of Cligès, while enumerating his own preceding works, Chrétien de Troyes intermingles them with translations of Ovid and Breton novels, as though the unity of writing prevailed in his eyes upon the ill-assorted subjects and sources. The troubadours never cease to cite and answer one another, marking by such play of intertextuality the closed and identifiable contours of a realm of poetry. Finally, the systematic undertakings in the thirteenth century to adapt or to translate the French novels and songs of gest into German or Norse, and later the way they are constantly circulated and brought up to date, also suggest a synthetic view of literary production.

7If one admits that medieval literature exists, its most striking feature when one looks at European literature as a whole, is that it constitutes its beginning.

8In a sense, it is an absolute beginning, when literature finds its expression in the Romance languages that have just sprung to life. It is a beginning that seems to suppose at every moment an unreachable prehistory in the form of oral productions or of a mythical substrate of which we believe we can discern the traces or the wreckages. It is a beginning that is not really one, since this literature inherits from Ancient literature with no solution of continuity, making its imitation one of its founding principles, and keeping a living and privileged place for Latin within it. This birth and this legacy were at first the most salient features: a new literature in new languages; a literature born of the school of Latin literature, partially written in Latin, but whose uncompromising originality is never so manifest as when it claims to reproduce ancient models.

9What a wonderful discovery – and what novelty! More than one would think. Even the term “Middle Ages”, which appears in the 18th century and defines the era, not so much in itself, but rather negatively or by contrast with the ones that precede and follow it, conveys a historical representation of literature and the arts. This is especially so because it replaced such expressions as “barbarian times”, which express a value judgement. Antiquity could serve as a model for French and European classicism because forms of spirit and taste were believed to be permanent. The Middle Ages garnered attention as soon as the changes affecting these forms were brought into focus, changes which triggered reflection on aesthetics and history. It is no coincidence that the first person to have really considered the Middle Ages within a philosophy of history and within aesthetics was Hegel. Before that, Vico had seen in poetic imagination the mark of childhood, the childhood of civilizations and individuals alike, and Herder, in his disorderly and, sometimes contradictory, fashion had conferred on the poetry of the Middle Ages considerable importance in the linguistic, artistic and national development of the European peoples. He thus gave a theoretical basis to the appeal that the ancient poetic traditions had had throughout Europe since Percy and Macpherson. “Poetic traditions”, the expression encompasses everything: the past, which in order to be considered as such presupposes both a change of some sort and continuity; the association between poetry and something that is somewhat primitive. According to Herder, the national genius and collective spirit of all peoples are embodied within and express themselves through the spontaneous manifestations of their art, which mark the beginnings of their history, and they survive where outside and scholarly influences have had no impact. Hence, the combined interest for the poetry of the Middle Ages and for popular poetry which, in Germany, manifests itself not only in Herder himself – even though his notion of the Volkslied is strangely eclectic – but also, and mainly, in Brentano, and to a lesser extent in Arnim, Uhland, the Brothers Grimm of course, and even, with no scientific pretensions but with the acuity of irony and sensitivity, in Heine. Mérimée perfectly and pleasantly portrayed this combination in his depiction of professor Wittembach, the narrator of Lokis. In France, Hersart de la Villemarqué exhibits as if by some miracle “popular Breton songs”, that are the exact copy of the lay of the Laüstic by Marie de France or of the meeting between Perceval and the knights in Chrétien de Troye’s Perceval, the Story of the Grail.

10Is it then necessary to go through romanticism, and through German romanticism, in order to arrive at the literatures of medieval France? Yes, it is. It is the concepts inherited from Herder that gave the studies of medieval philology and literature the task of defining the identity of the peoples of Europe, in the two senses of the word “identity”: what distinguishes and what unites them. For what distinguishes them is their greater or lesser ability to retain, along with their primitive genius, the primitive trace of the original, heavenly unity which was that of humanity at its beginnings, before it was lost. This is the theory which established the connection between the Middle Ages and folklore, and which is still profoundly entrenched today. It is what sustained philology’s passion, and the profound taste for a kind of poetry perceived as naive. These are the ideas, which, in nineteenth-century France, go with the combined interest for German romanticism and for the new philological methods applied in German universities. To parody Martianus Capella, medieval studies were born of the marriage between romanticism and philology. It was this combination that led the first French medievalists to look towards Germany and to deplore the fact that the study of our ancient language and literature was more developed there than in France. On 20 March 1852, Paulin Paris, assistant librarian of the manuscripts at the Bibliothèque nationale, wrote a note to request the creation of a Chair of French Language and Literature of the Middle Ages at the Collège de France. His main argument was as follows:

This language and this literature have [...] (as is known today all over Europe) precedence and pre-eminence over all of the neo-Latin literatures. The French chroniclers, poets and prose writers have inspired the best geniuses of Germany, Italy, Spain and Belgium; and France alone appears to decline the homage that the whole world pays to its literary genius by excluding this great and peculiar source of studies from higher education.

11Paulin Paris added that he would not be unworthy of occupying this Chair himself were it to be created. We know that he ended up being satisfied in both respects. A few years later, he sent his son and future successor, Gaston Paris, to Bonn, where he studied under Friedrich Diez, a specialist in Romance languages.

12One may object that, even so, this is attributing too much to Romanticism and to Germany. During the Enlightenment period, a kind of Italian patriotism was already motivating Muratori’s medieval research. In France, from Fauchet to Du Cange, interest in medieval history, mores and literature never waned completely. During the course of the 18th century, the Académie des Inscriptions et Belles-Lettres encouraged research on the Middle Ages. That of La Curne de Sainte-Palaye counted amongst the most pre-eminent. The wealth of the medieval collection in the Bibliothèque universelle des romans is well known, as is the importance that L’Esprit des lois [The Spirit of the Laws] gives to Frank and Carolingian legislation in the constitution of French law. We know that the Génie du christianisme [Genius of Christianity], which sees the Middle Ages as “the only poetic times of our history”, proposes a medieval and Christian model whose ideological tone is already perceptible in the short stories of Baculard d’Arnaud or the novels of Madame de Genlis, who dresses the misfortunes of emigration in medieval rags. Not to mention more illustrious works. In 1826 the academician Parseval – almost Perceval! – published a Philippe Auguste, poème héroïque en douze chants, in the first song of which the council of the gods is replaced by a council of fairies presided over by Melusine – for that matter wicked foreign anti-French and pro-English fairies. However, this bright idea sets the tone of the poem: it is an antique-style epic tale whose subject is medieval, the equivalent of a Franciade or a Henriade. Despite the author’s scholarly precautions, the era represented here has little more influence on the literary form than in Le Cid, Attila or Zaïre. Is it because the lamentable Parseval is a belated neo-classicist? Not at all, for does one find a reflection of medieval literary art in Les Burgraves? Does the poetry of each past era colour in different ways the plays that followed the Poèmes antiques et modernes or the Légende des siècles? A little, but so very little. The Middle Ages may actually provide French romanticism with the themes of its inspiration, but the prevalent idea that its literary monuments deliver the raw, pure essence of poetry, and that they can for this reason be compared to popular art, did not actually have a very profound impact. In fact this idea was soon ridiculed. In his first edition of Éducation sentimentale [Sentimental Education], the young Flaubert writes of one of his characters, a budding author who, we are told, reads Schelling and Herder, but has returned to sounder concepts:

He re-read as much as he could understand of the bards and troubadours, and frankly admitted that only a pretty queer fish would regard all this as sublime, though at the same time the real beauties he discovered in them struck him more forcibly than ever.

In short, he made a clean sweep of all the fragments of folk-song, translations of foreign poetry, barbaric hymns, cannibal odes, Eskimo ditties and all the other unpublished rubbish that they have been boring us to death with for the last twenty years.

13The increasingly detailed and precise knowledge of medieval literature made available thanks to the progress of philology at the end of the century will lead to the challenge of such associations. However, that progress will have little influence in itself over aesthetics reflection on the Middle Ages. Gaston Paris believes that the teaching of medieval literature “needs to be wisely restricted”, for “it does not meet the requirements of classical taste”. He is, however, always careful to include his discipline in the historical sciences and within a historical perspective; always mindful also, in the spirit of his times, to see literary history impose itself as a science, to the detriment of rhetoric. However, he does not go so far as to include aesthetics in the scope of his historical thinking. He takes an interest – too great an interest, his successors will say – in the encounter between the literature of the Middle Ages and popular poetry. He is very sensitive to the charm of the latter wherein he sees “the faithful and spontaneous expression of the French genius”. However, he hardly seeks to define a theoretical basis for this encounter and admiration. For him, as for the obscure Parseval sixty or eighty years earlier, the Middle Ages can be an object of study, even of inspiration, but its poetic forms are in no way a model. The historical depth of the literary past does little as such to inspire his thinking on the essence of poetry.

14Gaston Paris is still influenced by the romantic concepts, which inspire his theories on epic cantilenas and on the Indian origin of the fabliaux. Born of positive critique, his immediate successors, who are used to making hypotheses with reference to specific texts alone, demonstrate their loyalty to him by opposing him. For example, in every field Joseph Bédier supports theses that oppose his own. All of them emphasize the deliberate character of literary creation and the decisive role that particular individuals, great personalities and geniuses played in the appearance and the evolution of the poetic forms. This attitude does not only reflects the will to take into account preserved documents only, and the refusal to accept unverifiable hypotheses that are the basis for faith in the existence of a continuous oral tradition, which, by definition, is elusive; it also conveys, on the part of the man that Bédier was, gifted as he was with an exceptional sensitivity to texts, the affirmation of a certain idea of poetry. However, this idea too is dated. It is that of the first twentieth century, which tends to glorify the poet and even regard him as sacred, while at the same time attributing to poetry an extreme self-consciousness induced by the work that it demands. Such an idea also presupposes that aesthetic values, through the permanence of poetic genius and toil, are themselves permanent. It is not essentially concerned with integrating temporal considerations as a constituent factor of these values.

15Without being in any way guided by the taste for provocation or the love of paradox, we measure the originality of the romantic position in terms of contrasts. One imagines how fecund the medievalists’ reflection on the third age of Hegelian aesthetics might have been – had it taken shape, which it did not really do, despite Benedetto Croce’s atypical example. That was the romantic age, the age of subjectivity, which appeared with Christianity and whose true birth coincides with the Middle Ages. One finds oneself regretting being unable to adhere to the romantic vision of the Middle Ages, although it is erroneous in almost every way. For this form of thinking has an interest in the medieval past that is genuinely historical, since the interpretation of the history of the peoples is at stake here, but with the historical theory being based on poetics. In a sense, it is poetics that throws light on history, and not the contrary. Philology is necessary for understanding history in that it examines the shaping of a language, the birth of a literature and aesthetics. However, it is quite true that, based on the belief that by going back in time one finds the people, this reflection seeks the essence of poetry in its original and fundamental surge, flowing from buried depths – with all the associations that go hand in hand with the simple and the primitive, the old and the archaic of which Flaubert makes fun.

16Yes, the study of medieval literatures had a significant role at the time: to understand the spirit of the peoples and the nature of poetry. Within this perspective, the first manifestations of the literatures of Europe took on an extreme importance, greater in a sense than their later development, as they were supposed to reflect a state of art from which genius had not yet been erased by taste. What is however the situation today? Who still believes that the national identity of peoples reveals itself in their first artistic productions, and that these are collective and spontaneous? No one. Who would describe medieval literature as popular? No one. The development of philology soon dissipated the romantic illusions that had encouraged it. Yet even then the purpose of these studies did not seem to vanish, and the interest they aroused did not weaken. Why is that? Why has the fascination aroused by this literary past outlived the philosophical foundations that gave rise to and justified it? Was it because national quarrels were being fought on the remains of romantic thought and because, during that period of Franco-German conflicts, each nation sought superiority in the anteriority or pre-eminence of its literature? These concerns did exist, it is true, and even among the greatest medievalists, but it would be wrong to exaggerate them. A more accurate explanation, here as in all fields, might be the positivist faith in knowledge for the sake of knowledge and the (perhaps unjustified) conviction that all that can be known deserves to be known. This does not answer the question, however. Why did we seek, above all else, to find out what there was before what we already knew – or thought we knew – existed? Why this almost exclusive interest over decades for the insoluble question of origins: the origins of the chansons de geste (epic poems), of the lyrical forms, of courtly love and of the sources of Breton tales? One could certainly invoke the constant tendency to look for a key to ourselves in every beginning – that of the universe, of life, of civilization – and for a justification of our choices in the past, as if the latter resonated with our personal past and our own childhood. And it is true that the Middle Ages, which are so closely linked to childhood, have for two centuries been forcibly used to support various ideologies. However, beyond this overly general and facile answer, an examination of how medieval literature achieves its effects would undoubtedly reveal that the prehistory of literature is above all a trompe-l’œil of literature itself. It is the texts themselves that mislead the reader as to their antecedents.

17This trompe-l’œil has probably existed in every age and in all literary systems. It meets an aesthetic need and an expectation of the mind. However, the Middle Ages, an era which is actually marked by the signs of a commencement, by the anguish of a break from the past and a concern for continuity, made use of this effect with predilection. The romantic in quest of the roots of a collective identity, the philologist in search of positive sources, and the anthropologist on the trail of the primitive forms of art and of the imaginary: they all found in medieval literature a domain that was all the more promising that it had held up to them for centuries a mirror ready to reflect their own illusions back at them. They were true illusions, however, since these false perspectives are really contrived by this literature.

18This is the first proposition that I would presently like to put forward. The second is that this illusion is true, this illusion where the nature of the object that gives rise to it reveals itself equally to the ethnologist and to the folklorist, such that the tenacious association between the literature of the Middle Ages and folklore is quite different from the deplorable relic of romantic thought. If all the conjurations have been powerless to dissipate it, it is for excellent reasons, but not because folklore – whatever name one gives to it or whatever reality one places under that name – can offer an explanation of some kind for medieval literature. The third proposition is that it is in literature’s very nature to produce this illusion, to the point where one is allowed to seek therein a defining criterion for it.

19That the literature of the Middle Ages constantly refers to its own past is evident. That it might itself weave the backdrop in front of which it poses, and give the illusion of depth to a backdrop that it has itself painted, seems to be an arbitrary presupposition. For after all, the vast Latin section of this literature really does carry on from the ancient literatures – pagan, biblical, patristic – and rightly presents itself as their descendant at the same time as it assures its living preservation. The same goes for whole sections of vernacular literature that are based on Latin models. As for the new forms which emerge before our very eyes from the end of the ninth to the end of the eleventh century, after the appearance of Romance languages and their entry into the field of writing, and which go on to flourish as we know they did, it would be as unreasonable to deny that they were rooted in a tradition that necessarily mostly escapes us, as to reduce them to the emergence of such a tradition in the written form. But let us consider things from a different angle. Why are we constantly led to suppose that a tradition escapes us? Why, as we go back in time from text to text, up to the most ancient ones, do we always have the feeling that we are left at the doorstep of an enigma? Because the texts multiply the clues that give us that feeling. They do not let these clues escape [from their pen]; they sow and place them so that their effects will comply with the poetic requirements of their time, not with those inherited from the past.

20And so it is for a good many manifestations of Romance lyricism, whose origins have intrigued and where the contrast between the scholarly courtly poetry and the simplicity of the so-called women’s songs has been so striking. In the latter we witnessed the remains of an ancient tradition sprung from the depths of a people, and which the new, demanding, complicated art of the troubadours might have shrouded before letting scraps of it reappear. This might be true. What is certain, however, is that medieval literary art makes a point of obtaining contrasting effects by giving part of its lyrical production marks of simplicity and of rusticitas, and that in order to do so it provides clues that associate this production with the past. As far back in time as one goes, one sees in every period of literature, in successive synchronous layers, the poetic system as a whole integrate within itself songs which are constantly marked with the same signs that oppose them to the rest of poetry, the latter being concerned with refinement and novelty. Which signs? The signs of that which is fragmented and discontinuous, and which produce two effects. The first is that of citation: the song only exists between scholarly inverted commas. The second is that of the trace: the fragment is unfailingly perceived as a vestige from a whole that is no more.

21“As far back in time as one goes”: the expression must be understood literally. Among the most ancient texts preserved in the Romance language there are fragments of songs that are cited in completely different poems written in another language. These are the recognisable khardjas which conclude the muwashshahs of Al-Andalus. Unlike the rest of the poem, which is written in Arabic – or in Hebrew – some of the khardjas are written in the Mozarabic dialect, the Romance language of the indigenous population, that of the Christians of Spain. All of these Mozarabic khardjas express in a few words the complaint of a young girl who is in love: “Oh you who are brown, oh delight for the eyes! Who will be able to bear [your] absence, my friend?” The change of language, the change of style and the change of subject: everything makes the contrast between the poem and this final point intense. However, it is the contrast that highlights the pointedness of the citation. The greater the distance between the poem and its khardja, the more ingenuity and wit the poet had to employ to dare to make a link between the two which is at once relevant and audacious. The khardja, which metrically and musically both founds and ends its new, brilliant, complex poem, must appear fragmented and stammering, coming from the dawn of time and from the depths of the soul, with an insisting, over-determined simplicity, that of a language that is not one of culture, that of rudimentary poetics, that of an ignorant young girl expressing herself.

22One cannot seek to reconstitute a past of which these Romance khardjas would be at once the outcome and the monuments: they emerge from a world that escapes us. Some of them have evidently been reshaped, if not invented, by the Arab poet who cites them. However, they find their place in a form of poetry that uses them by playing with fragmentary citation and cultural distance in order to give them the flavour of a traditional art which contrasts with a scholarly art.

23Far from being unique, this kind of effect constantly reappears during the rest of the Middle Ages, and within Romania itself. It is no doubt what made the success – limited though it was – of the chansons de toile at the beginning of the thirteenth century, with their ancient feel, their stanzas similar to those of the epic laisse in their simplicity, their elliptic and wooden narration, and this grave young girl whom they portray as handier with needlework than with the subtleties of lovers’ talk and who only knows how to offer herself in silence to a nonchalant seducer. All of these characteristics oppose them to courtly lyricism, but certain details – an aspect of social mores, a melody which is suddenly embellished through fioritura, a slightly heavy-handed imitation of the epic style – make one suspect a forced archaism. The mischievous Jean Renart takes great care to stress both their oldness and their old-fashioned character. When he inserts such details in his Roman de la Rose or Guillaume de Dole, he gets them sung by an old chatelaine buried in the depths of her province, who herself says: “It was in former days that the ladies and queens were in the habit of doing tapestry work and singing historical songs.” At the same time, at the court, the Emperor’s juggler sings an exaggerated and vaguely bawdy pastiche of them. It is likely that these songs actually derive from an ancient tradition. However, it is especially obvious that they claim the characteristics of oldness and simplicity, even of clumsiness. In the first third of the thirteenth century, a public used to the sophistication of the grand chant courtois would have found them charming. They had the charm of the archaism which they played upon. They also had the charm of what is fragmentary, which is linked to the former and is evidenced in the songs’ truncated citations, mysterious laconism and knowing tone [ton entendu] – knowing tone, in the sense of having already been heard, and what is a song if not a ton entendu [literally: a heard tone, one that is known]?1

24We have there, at the scale of a set of lyrical forms which balance themselves through contrast, the same aesthetical effects that existed three centuries earlier within the muwashshah. The only difference is that the opposition between cultures and languages, which has disappeared, has been replaced by an emphasis on the depth of the past. Simplicity, tradition, archaism, oldness and naivety are notions which begin to be associated with one another.

25Let us go on. Some songs preserved in the collections dating from the second half of the fifteenth century have often been called “popular”. The manuscripts are carefully made, even luxurious. Many of these songs differ little, formally, from the ones that sustained the polyphonies of the court of Burgundy during the same period. They are sometimes the same. Thematically, they lie within the tradition of the lyrical genres cultivated by the trouvères. What then is so “popular” about them? Only the fact that they carry on the tradition of the trouvères, a scholarly tradition if there be one, but an already ancient one, which had then been abandoned by courtly poetry, hungry as it was for rhetorical, metrical, linguistic and musical novelty. By contrast to this tradition, these songs in the fashion of distant times, these stanzaic, monodic songs, with spring gardens, nightingales, picked roses and shepherdesses, combine simplicity with a recollection of the past. They are sometimes mutilated, sometimes incoherent. They appear to be remnants while at the same time favouring a rustic universe. They contain small archaisms of language which will later seem characteristic of popular literature. They thus lead to ancient tradition being confused with popular rootedness. Giving the illusion of a refreshing plunge into a rustic past, they compensate for and counterbalance the tense aridity of the scholarly poetry of their time. One must therefore not be surprised to find them collected in beautiful manuscripts, since they are the necessary complement of courtly poetry, to which they belong on this account.

26It would be easy to go on this way right up to modern times, beginning with Marot, who associates young love with old song:

The song is (without giving the tune):
“Alegez moy, doulce, plaisant brunette.” / “Relieve me, sweet and pleasant brunette.”
It is sung in the old-fashioned way.
But that is all the same, the brunette is young.

27Or with another “old song”, that of Alceste, a simple song, which attests to an art that is at once coarse, ancient and superior to the scholarly and artificial art of today, with its “figurative style” and its “trinkets”: “The rhyme is not rich and the style is old because of it”; “Our forefathers, all crude, had it (i.e. taste) far better.” In the same idealization, Alceste confuses oldness with naive simplicity, as Rousseau does soon after in the New Héloïse:

Most of these songs are old romances whose tunes lack piquancy; but something ancient and sweet about them is ultimately touching. The words are simple, naive...

28Not to mention Nerval, who uses the songs of the Valois, of which he only ever cites snatches, as a counterpoint to his journeys to the land of his childhood, to his childhood memories, to his dives into history, in short as a fleeting brush with his awareness of a cloudy past? Or the delightful lines where George Sand remembers the impression made on the little girl that she was by the enigmatic and melancholic suspense of the song “No longer will we go to the wood, / The laurels have been cut”? Forgetting is a sign of the past. Songs need to be fragmentary, so that one may suppose them to be half lost, on the verge of being lost.

29And here we have simply kept to the example of lyricism. There are many others. “The rhyme is not rich”, says Alceste, who turns poorness into something rich/wealth. Similarly, one sees the chansons de geste clinging on to assonance, while they appear very early on to be attracted to rhyme. One sees them preserve the epic decasyllable in spite of everything, even when faced with the new fashion of the Alexandrine, to which they occasionally yield. One sees them become contaminated by the fashion of the romantic account, but not without resistance. One sees them preserving their primary identity longer than any other genre, as if they knew that their appeal was due to their coarseness; as if they defined themselves, in comparison with the other genres, through their rustic conservatism. Gregory Nagy has strongly supported an analogous hypothesis concerning epic Greek poetry in relation to lyricism. If the literary forms were a concert, the part of the epic would be that of the simple harmonies and the affected dissonances. These are harmonies and dissonances which the famous cantilene of Saint Faron (which historically is a fake – thereby guaranteeing its poetic truth) cultivates within the Latin world on the brink of fragmentation and before even the appearance of Romance literature, with a remarkably just feeling of what Paul Zumthor calls “the sidestep” (how to resign oneself to say “used to call”?).

30However, might this insisting and illusory trace of the past not be itself an illusion? How can we be certain that the texts really aim to have the effects that have been ascribed to them here if they never explicitly state their intention of resorting to them? Here is a text, however, also very well known – too well known – and endlessly commented upon, which clearly indicates the trick by which literary work ostensibly obtains its justification from a past whose contours are only drawn in its own writing and which are at the same time blurred by the latter. It is the prologue to the Lais of Marie de France. We know its progression: the Ancients composed works that were deliberately obscure, counting upon time and the readers’ ever increasing penetration for their meaning to emerge. Marie had at first thought of dedicating herself to a translation from Latin into French. Instead of this too ordinary a project, she undertook to transcribe the Breton lays which she had heard, so that they would not be forgotten. Clearly, in her mind, the translation from Latin that she gave up would have been a matter of slowly and progressively elucidating the antique literature that she had first defined, thus granting a hermeneutical value to imitatio. It is also clear that she considers her two projects, the one that she abandoned and the one that she successfully carried through, to be comparable. The displayed proposition is that the new work is the heir and the outcome of the works of the past whose meaning it undertakes to unfurl or whose memory it promises to save. In the first case, the veil of the integumentum can be lifted only by the passing of time, which threatens the survival of the lays in the second case. However, it is in fact Mary’s very prologue which casts behind it the shadow of a doubt – a doubt stemming from the uncertainty of the meaning or the frailty of memory, and from which it claims to derive its raison d’être and appeal. The text constitutes itself by positing both the existence of a model taken from the past and its shortcomings, such that neither the one nor the others can be separated from what the text is in itself. It would be pointless to go and look for them without it. One would only find the abstruseness of meaning or the darkness of what is forgotten. In the same way, in the Lai du Chèvrefeuille by the same author, it is impossible to distinguish the remnants of the inscription engraved on the hazelnut stick covered with honeysuckle, or maybe only signified by it, from the commentary which claims to embrace and develop it, but which feeds off it and absorbs it to the point of making its identification impossible. Certainly, and to get back to the prologue to the Lays, Latin literature does exist. And so too, probably, do Breton lays.

31After all, the new work bases itself on these earlier models. However, it only takes shape by shrouding the forms and throwing them back into the shadows, while exhibiting them in this indistinct state.

32Whether it be in the picture constituted by all of the literary forms of a given moment, in the discrepancy of the citation or in the distance from the model that a particular work makes use of, everywhere medieval literature places the shadows and arranges the perspectives that suggest the depth of the past. At every period of its history, part of itself affects an archaic simplicity and thus opposes itself to the novelty that the other part claims. In order to do so, it takes on the appearance of the fragmentary and the discontinuous, symptoms of having been forgotten, which is in itself a mark of the past, while at the same time it employs poetic and linguistic effects meant to place it juxta rusticitatem, as Hildegaire would say about her cantilene of saint Faron. From the khardjas of El-Andalus to the songs of the Valois, we have not ceased to observe that the suggestion of incompleteness, that of rustic simplicity and that of archaism are interlinked. Such effects bring us back to our second proposition, which is, we recall, that the illusory depth of the past also offers itself to the ethnologist and the folklorist.

33Our brief cavalcade just now through the songs and centuries has already illustrated this proposition. All that has been said up to here comes down to applying to literature the famous saying that traditional societies “invent traditions in order to justify innovation”. Of course, the permanence in time of narrative or poetic themes is not an illusion but a fact, as is their extension in space. We are not concerned here with the fact itself, but with the fascination that it arouses. To seek in the past the truth of popular culture and the traces of its now lost coherence is not a distinctive feature of romantic illusions. Even contemporary ethnology, hardly inclined to naivety as it is, gives it up unwillingly. The investigator always has the feeling that he is arriving just a little too late, that the object that he is pursuing is already half gone, but that traces of it nevertheless subsist. His informants share this impression, and also suggest it to him. Singers and storytellers of both sexes have always protested that songs and stories have been lost, and that they have recently been lost, that they remember only snatches of them, but that this whole world was still alive when they were young: “ My poor mother, yes! She would have filled a sack of them, with the bottom cut off. As for me, it’s gone: the stories, I’ve let them disappear”, is what Henri Pourrat makes the typical storyteller say. We hear the same line of reasoning at the dawn of the nineteenth and the end of the twentieth centuries. It has therefore more to do with representation than reality. The representation of what exactly? Of the past. Here again, the oldness of something is guaranteed by the fact that it has been forgotten. “The ideology of the good old times” and “the illusion that the productions ... in pieces today, used to be coherent before” remain present, as Nicole Belmont put it. Boas, in a passage cited and commented by Lévi-Strauss, observes that a myth always appears like a fragment that has survived. The ethnological object offers its form to the gaze of whoever is studying it only by inviting this gaze to look towards the past. In our civilization, by playing this game the ethnologist soon finds himself within the field of the medievalist, who experiences the same dizziness from the texts that he has before him.

34The ethnologist is thus led to ask himself what his object was before he seized it, and is tempted to believe that, had he only arrived a little earlier, everything would have been explained. In the same way, the medievalist is brought to suppose that the most ancient text that he has before him is only the continuation, the trace, the scattered limbs of a text, or of an oral production, or even of a myth, that are older, lost or disfigured. If only he knew the Song of Roland which preceded the Song of Roland, the legend of Tristan before Béroul and Thomas, Geoffrey of Monmouth’s Celtic source, the book on the Grail that Philippe of Alsace showed to Chrétien de Troyes, Eble de Ventadour’s lost poems, the stanzaic models of William IX, Duke of Aquitaine, or the songs from which the khardjas were taken! Here we have the storyteller working the shipwreck of memory into a clue of tradition, just as Marie de France did with the Breton lays that were in danger of being forgotten. Here we have the myth claiming to be broken up by time in order to offer itself to its interpreter’s sagacity, in the same way that ancient literature, always according to Marie de France, needed to hide in the shadow of the past for a gloss to shed light on it and complete it.

35It is not surprising, then, that the association between medieval literature and ethno-literature has survived the hypothesis that they might both be collective and spontaneous productions revealing a national identity. This is because it comes first. Far from stemming from it, it is the latter which, in particular historical and ideological circumstances, suggested such a hypothesis. It is based on the impression that they both still carry the more or less indistinct trace of a past which founds their authenticity and whose knowledge would reveal their secret. This impression is indestructible, for it is of a poetic nature. Once again, it is an effect produced by the works themselves, an effect made perceptible by the contrast with other works, which have a different effect. Which other works? The part of medieval literature which lays claim to novelty and sophistication; the official, the canonical, the scholarly “highbrow” literature, – or, to put it more simply, the literature of the investigator – without whose opposition to which ethno-literature or the said popular literature could not be perceived as literature as such. If this is the case, however, then this effect is one of the constituents which, grouped together, form the very notion of literature. That was our third proposition. If it is sound, this notion whose mere application to the Middle Ages had hitherto seemed problematic to us emerges only by appealing to a particularly remarkable trait in medieval literature.

36Let us actually summarize our observations one last time. The literary art of each age takes care to mark certain productions with signs that associate them with the past. These signs of course include explicit references to the past, but not exclusively. They are also signs of simplicity and rusticity, with the implicit hypothesis of a progressive refinement of literature. And signs of the fragmented, producing on this account a dual effect of citation and survival. They thus suggest that these productions are remnants of a body of work buried in the past and they prepare the association of the archaic with the popular. The productions thus marked according to the era as archaic, popular or both at the same time, and which seem to be out of fashion and out of date, contrast with others, which are felt to be scholarly, new and fashionable, and which constitute the heart of intellectual and literary life, giving it its lustre. Their most apparent function is to give balance and relief to all of the literary forms, by contrast, and to open up the perspectives of deep-rootedness and of an evolution. The ideological representations that such a system does not fail to arouse would thus be subordinate to aesthetics, and not the other way round.

37However, one could also recognize in these productions another function, which, in the equilibrium of the literary system, would place them no longer on the fringes, but at the centre. They would not only be the dark side through which the light comes out, but in a way the conscience of literature, the reminder of what it essentially is. Is it actually absurd to apply to literary activity itself the characteristics that define wild thinking [pensée sauvage] according to Lévi-Strauss: activity of the sign rather than of the concept, largely founding its effects and discourse on analogy? No wonder then that it feels the need to remind us that it is fundamentally linked to something primitive, just as poetry is perpetually led to situate itself according to intellectuality, and to confront it. Perhaps literature, whose task is to give words their weight, confusedly sees and experiences itself as an ongoing struggle against the levelling out of the signs of language, like the nostalgia of a past state where, as Lévi-Strauss again would say, they were values as well as signs. These were at once clumsier and richer than what a greater suitability for the needs of communication has made of them. Literature would then find its justification at every moment of its existence only if part of itself brought out, by contrast, the trace of a past on which it claims to found itself. Is it not true that the metaphor of the imaginative and irrational infancy of peoples, like the conviction that imagination is rooted in memory, as Vico would put it, find their best justification in literary productions, which never cease to provide them not only with a general confirmation, but also with a particular denial? A confirmation, seeing as they unceasingly refer back to this childhood and substantiate the idea that “the world in its infancy was constituted of poetic nations”, which is what one of Vico’s axioms says, verbatim this time. A denial, since, if closely examined and taken each in their singularity, all the literary productions situate themselves beyond such infancy and designate it or seem to keep a trace of it only through a conscious device.

38Must one then, going back to the sources of medieval studies and to those of the present paper, invoke Hegel’s famous formula, according to which “from the point of view of its supreme destination, art is something of the past”? Within the framework of Hegelian aesthetics, outside of which it makes no sense, it means that since the Christian era the purpose of art has no longer been to reveal the truth of the world. It therefore has nothing to do with the considerations we are engaged in here, in theory. And yet, not only does it apply, within Hegelian divisions, to the modern period of art, of which the Middle Ages are the first original stage; it also assumes that a kind of nostalgia is inherent in art, as is concern regarding its purpose. It thus conjointly designates and associates together two essential traits of medieval literature, i.e. the insisting allusion to its own fragmented past, and its constant questioning about its legitimacy with regard to faith. If art belongs to the past, it is because it has in a way been discredited by the present of the Revelation.

39A new order of relations and contrasts is thus added to the one suggested here. It is not only a part of poetry which claims to be turned towards the past in order to highlight another part which lays claim to novelty. Poetry as a whole, while it claims to reflect the truth, finds itself relegated by the light of the Christian truth into the shadow of the past and derives from this its obscure appeal. The literary effect through which the suggestion of the residual and the primitive are contrasted with the claim of novelty draws its strength from the implication that a truth lies the past. The medievalists have responded to this by hastening to scrutinize the prehistory of their texts. However, the medieval reader could measure this supposed truth against truth itself only. From considerations on the particular and disconcerting nature of pagan morality which marks certain adaptations of Ancient works, to the ambiguous effort to constitute around the Grail a parallel story, a justificatory archaeology of Christian chivalry; from the fragment of an innocent and provoking cry of love that is the song of the troubadour, to the patient completion of lyrical crowns to the Virgin Mary: everywhere can be perceived the avowed, veiled, sometimes denied and often anguished concern to confront “the love of literature and the desire for God”, according to the admirable and famous expression of Dom Jean Leclercq, a love that is nostalgia and a desire that is hope, the intuition of a revelatory reminiscence of self and the call to a truth which tears away from oneself. The incessant assertion of faith in medieval literature is therefore not the insipid gangue which surrounds and hides the nuggets of pure poetry or the revelatory clues of profound mentalities and sunken myths. This is not only because it alone allows one to measure accurately the relationship between Latinity and vernacular. It is also because it accentuates and doubles the contrast which shapes literature by inscribing upon it the vanishing lines of a past in trompe-l’œil.

40Trompe-l’œil, illusion: one will say that impression is given too great a part here. Do we not have anything more solid to present? No, because nothing is more solid than impression. To be sure, the positive knowledge of medieval literature is progressing, and that is good. Dates, attribution, localizations are more precise and certain. The history of the texts is coming to light. Discoveries are being made: have one hundred and fifty-four verses of Thomas’ Tristan not just been found? However, conjectures still play a great part, and some uncertainties are absolute. By contrast, impression exists without any doubt. It deceives us, of course. The Middle Ages are a distant era that has become foreign to us. If they appear close, therein lies the illusion: an undisputed truism that is gravely repeated to us. However, this illusion is true, since we fall prey to it. It deserves to be confronted with the facts as we know them, reconstitute them and suppose them to be. Certainly, the forms of thought and sensitivity have changed, as have their expression, and language itself. Does that mean that we must refuse to recognize that the Middle Ages remain half familiar to us? But it is true that the right distance can be found only at the cost of successive, contradictory and forever challenged adjustments. Whoever examines the poetry of the past has no other choice, unlike perhaps the historian, than first to run the risk of proximity. Certainly, he or she cannot claim to read the old texts like their contemporaries would or the way their authors had meant them to be read. Which only makes it all the more miraculous that they still touch the readers. They do so at the cost of their misinterpretations, they mislead them, and yet they do indeed mislead them. Let us leave these mistakes and seek their truth in the Middle Ages and in ourselves.

41It is a truth that will always be fragmentary – a fragmentariness that lies at the heart of literature –, but a truth which for that very reason gives the seeker a pleasure similar to its own. It is a truth whose scattered pieces we interminably put together, but of which others than ourselves may sometimes have the dazzling revelation. This is a discouraging revelation for whoever dedicates his or her life to attempt patiently, ploddingly, to reconstitute the aim, the meaning and the flavour of these faraway texts. It is also a fortunate one, for at the end of this labour we have no other resource left than to put ourselves in its hands. For in the end a voice other than that of the pedant needs to make these texts resonate so that they can arouse the pleasure and meditation without which there can be no good reason to read them. To finish off, one must listen to the voice of the poet:

Iseut sojourns in her very poorly lit lower room,
Her dress is the colour of the dead’s attent,
And it is the dullest blue in the world,
Flaking off, revealing the ochre of the naked stones.

42The medieval legend offers itself like a fresco half discoloured by time. The “dull blue”, the colour of time, is flaking off. Only traces of it still remain on the ochre of what is forgotten. The lack of light, the dull colour that is almost gone, “the dead’s attent”: this is the mortal oppression which weighs on Iseut’s destiny. It is the threatened memory that the medieval poems claim to preserve; the endangered memory that we keep of these poems, and which makes them so valuable to us. It is our idea of the Middle Ages – hammered out capitals, bits of frescoes emerging from under the whitewash –, a period which touches us to the very extent that we perceive it as distant and blurred. It is Iseut, but in a later version it is the “orans”. It is the past of the legend and it is the wait for the prayer. It is a poem by Yves Bonnefoy. It must be feared that the plodding gloss of medieval literature will never shed light on them this way.