Читать книгу A Short History of French Literature - Saintsbury George - Страница 5

The Tale of Arthur. Its Origins.

The passion for narrative poetry, which at first contented itself with stories drawn from the history or tradition of France, took before very long a wider range. The origin of the Legend of King Arthur, of the Round Table, of the Holy Graal, and of all the adventures and traditions connected with these centres, is one of the most intricate questions in the history of mediaeval literature. It would be beyond the scope of this book to attempt to deal with it at length. It is sufficient for our purpose, in the first place, to point out that the question of the actual existence and acts of Arthur has very little to do with the question of the origin of the Arthurian cycle. The history of mediaeval literature, as distinguished from the history of the Middle Ages, need not concern itself with any conflict between the invaders and the older inhabitants of England. The question which is of historical literary interest is, whether the traditions which Geoffrey of Monmouth, Walter Map, Chrestien de Troyes, and their followers, wrought into a fabric of such astounding extent and complexity, are due to Breton originals, or whether their authority is nothing but the ingenuity of Geoffrey working upon the meagre data of Nennius51. As far as this question concerns French literature, the chief champions of these rival opinions were till lately M. de la Villemarqué and M. Paulin Paris. In no instance was the former able to produce Breton or Celtic originals of early date. On the other hand, M. Paris showed that Nennius is sufficient to account for Geoffrey, and that Geoffrey is sufficient to account for the purely Arthurian part of subsequent romances and chronicles. The religious element of the cycle has a different origin, and may possibly not be Celtic at all. Lastly, we must take into account a large body of Breton and Welsh poetry from which, especially in the parts of the legend which deal with Tristram, with King Mark, &c., amplifications have been devised. It must, however, still be admitted that the extraordinary rapidity with which so vast a growth of literature was produced, apparently from the slenderest stock, is one of the most surprising things in literary history. Before the middle of the twelfth century little or nothing is heard of Arthur. Before that century closed at least a dozen poems and romances in prose, many of them of great length, had elaborated the whole legend as it was thenceforward received, and as we have it condensed and Englished in Malory's well-known book two centuries and a half later.

Order of French Arthurian Cycle.

The probable genesis of the Arthurian legend, in so far as it concerns French literature, appears to be as follows. First in order of composition, and also in order of thought, comes the Legend of Joseph of Arimathea, sometimes called the 'Little St. Graal.' This we have both in verse and prose, and one or both of these versions is the work of Robert de Borron, a knight and trouvère possessed of lands in the Gâtinais52. There is nothing in this work which is directly connected with Arthur. By some it has been attributed to a Latin, but not now producible, 'Book of the Graal,' by others to Byzantine originals. Anyhow it fell into the hands of the well-known Walter Map53, and his exhaustless energy and invention at once seized upon it. He produced the 'Great St. Graal,' a very much extended version of the early history of the sacred vase, still keeping clear of definite connection with Arthur, though tending in that direction. From this, in its turn, sprang the original form of Percevale, which represents a quest for the vessel by a knight who has not originally anything to do with the Round Table. The link of connection between the two stories is to be found in the Merlin, attributed also to Robert de Borron, wherein the Welsh legends begin to have more definite influence. This, in its turn, leads to Artus, which gives the early history of the great king. Then comes the most famous, most extensive, and finest of all the romances, that of Lancelot du Lac, which is pretty certainly in part, and perhaps in great part, the work of Map; as is also the mystical and melancholy but highly poetical Quest of the Saint Graal, a quest of which Galahad and Lancelot, not, as in the earlier legends, Percival, are the heroes. To this succeeds the Mort Artus, which forms the conclusion of the whole, properly speaking. This, however, does not entirely complete the cycle. Later than Borron, Map, and their unknown fellow-workers (if such they had), arose one or more trouvères, who worked up the ancient Celtic legends and lays of Tristram into the Romance of Tristan, connecting this, more or less clumsily, with the main legend of the Round Table. Other legends were worked up into the omnium gatherum of Giron le Courtois, and with this the cycle proper ceases. The later poems are attributed to two persons, called Luce de Gast and Hélie de Borron. But not the slightest testimony can be adduced to show that any such persons ever had existence54.

These prose romances form for the most part the original literature of the Arthurian story. But the vogue of this story was very largely increased by a trouvère who used not prose but octosyllabic verse for his medium.

Chrestien de Troyes.

As is the case with most of these early writers, little or nothing is known of Chrestien de Troyes but his name. He lived in the last half of the twelfth century, he was attached to the courts of Flanders, Hainault, and Champagne, and he wrote most of his works for the lords of these fiefs. Besides his Arthurian work he translated Ovid, and wrote some short poems. Chrestien de Troyes deserves a higher place in literature than has sometimes been given to him. His versification is so exceedingly easy and fluent as to appear almost pedestrian at times; and his Chevalier à la Charrette, by which he is perhaps most generally known, contrasts unfavourably in its prolixity with the nervous and picturesque prose to which it corresponds. But Percevale and the Chevalier au Lyon are very charming poems, deeply imbued with the peculiar characteristics of the cycle – religious mysticism, passionate gallantry, and refined courtesy of manners. Chrestien de Troyes undoubtedly contributed not a little to the popularity of the Arthurian legends. Although, by a singular chance, which has not yet been fully explained, the originals appear to have been for the most part in prose, the times were by no means ripe for the general enjoyment of work in such a form. The reciter was still the general if not the only publisher, and recitation almost of necessity implied poetical form. Chrestien did not throw the whole of the work of his contemporaries into verse, but he did so throw a considerable portion of it. His Arthurian works consist of Le Chevalier à la Charrette, a very close rendering of an episode of Map's Lancelot; Le Chevalier au Lyon, resting probably upon some previous work not now in existence; Erec et Énide, the legend which every English reader knows in Mr. Tennyson's Enid, and which seems to be purely Welsh; Cligès, which may be called the first Roman d'Aventures; and lastly, Percevale, a work of vast extent, continued by successive versifiers to the extent of some fifty thousand lines, and probably representing in part a work of Robert de Borron, which has only recently been printed by M. Hucher. Percevale is, perhaps, the best example of Chrestien's fashion of composition. The work of Borron is very short, amounting in all to some ninety pages in the reprint. The Percevale le Gallois of Chrestien and his continuators, on the other hand, contains, as has been said, more than forty-five thousand verses. This amplification is produced partly by the importation of incidents and episodes from other works, but still more by indulging in constant diffuseness and what we must perhaps call commonplaces.

Spirit and Literary value of Arthurian Romances.

From a literary point of view the prose romances rank far higher, especially those in which Map is known or suspected to have had a hand. The peculiarity of what may be called their atmosphere is marked. An elaborate and romantic system of mystical religious sentiment, finding vent in imaginative and allegorical narrative, a remarkable refinement of manners, and a combination of delight in battle with devotion to ladies, distinguish them. This is, in short, the romantic spirit, or, as it is sometimes called, the spirit of chivalry; and it cannot be too positively asserted that the Arthurian romances communicate it to literature for the first time, and that nothing like it is found in the classics. In the work of Map and his contemporaries it is clearly perceivable. The most important element in this – courtesy – is, as we have already noticed, almost entirely absent from the Chansons de Gestes, and where it is present at all it is between persons who are connected by some natural or artificial relation of comradeship or kin. Nor are there many traces of it in such fragments and indications as we possess of the Celtic originals, which may have helped in the production of the Arthurian romances. No Carlovingian knight would have felt the horror of Sir Bors when the Lady of Hungerford exercises her undoubted right by flinging the body of her captive enemy on the camp of his uncle. Even the chiefs who are presented in the Chanson d'Antioche as joking over the cannibal banquet of the Roi des Tafurs, and permitting the dead bodies of Saracens to be torn from the cemeteries and flung into the beleaguered city, would have very much applauded the deed. Gallantry, again, is as much absent from the Chansons as clemency and courtesy. The scene in Lancelot, where Galahault first introduces the Queen and Lancelot to one another, contrasts in the strongest manner with the downright courtship by which the Bellicents and Nicolettes of the Carlovingian cycle are won. No doubt Map represents to a great extent the sentiments of the polished court of England. But he deserves the credit of having been the first, or almost the first, to express such manners and sentiments, perhaps also of having being among the first to conceive them.

These originals are not all equally represented in Malory's English compilation. Of Robert de Borron's work little survives except by allusion. Lancelot du Lac itself, the most popular of all the romances, is very disproportionately drawn upon. Of the youth of Lancelot, of the winning of Dolorous Gard, of the war with the Saxons, and of the very curious episode of the false Guinevere, there is nothing; while the most charming story of Lancelot's relations with Galahault of Sorelois disappears, except in a few passing allusions to the 'haughty prince.' On the other hand, the Quest of the Saint Graal, the Mort Artus, some episodes of Lancelot (such as the Chevalier à la Charrette), and many parts of Tristan and Giron le Courtois, are given almost in full.

It seems also probable that considerable portions of the original form of the Arthurian legends are as yet unknown, and have altogether perished. The very interesting discovery in the Brussels Library, of a prose Percevale not impossibly older than Chrestien, and quite different from that of Borron, is an indication of this fact. So also is the discovery by Dr. Jonckbloet in the Flemish Lancelot, which he has edited, of passages not to be found in the existing and recognised French originals. The truth would appear to be that the fascination of the subject, the unusual genius of those who first treated it, and the tendency of the middle ages to favour imitation, produced in a very short space of time (the last quarter or half of the twelfth century) an immense amount of original handling of Geoffrey's theme. To this original period succeeded one of greater length, in which the legends were developed not merely by French followers and imitators of Chrestien, but by his great German adapters, Wolfram von Eschenbach, Gottfried of Strasburg, Hartmann von der Aue, and by other imitators at home and abroad. Lastly, as we shall see in a future chapter, come Romans d'Aventures, connecting themselves by links more or less immediate with the Round Table cycle, but independent and often quite separate in their main incidents and catastrophes.

The great number, length, and diversity of the Arthurian romances make it impossible in the space at our command to abstract all of them, and useless to select any one, inasmuch as no single poem is (as in the case of the Chansons) typical of the group. The style, however, of the prose and verse divisions may be seen in the following extracts from the Chevalier à la Charrette of Map, and the verse of Chrestien: —

Atant sont venu li chevalier jusqu'au pont: lors commencent à plorer top durement tuit ensamble. Et Lanceloz lor demande porquoi il plorent et font tel duel? Et il dient que c'est por l'amor de lui, que trop est perillox li ponz. Atant esgarde Lanceloz l'ève de çà et de là: si voit que ele est noire et coranz. Si avint que sa véue torna devers la cité, si vit la tor où la raïne estoit as fenestres. Lanceloz demande quel vile c'est là? – 'Sire, font-il, c'est le leus où la raïne est.' Si li noment la cité. Et il lor dit: 'Or n'aiez garde de moi, que ge dont mains le pont que ge onques mès ne fis, nè il n'est pas si périlleux d'assez comme ge cuidoie. Mès moult a de là outre bele tor, et s'il m'i voloient hébergier il m'i auroient encor ennuit à hoste.' Lors descent et les conforte toz moult durement, et lor dit que il soient ausinc tout asséur comme il est. Il li lacent les pans de son hauberc ensenble et li cousent à gros fil de fer qu'il avoient aporté, et ses manches méesmes li cousent dedenz ses mains, et les piez desoz; et à bone poiz chaude li ont péez les manicles et tant d'espès comme il ot entre les cuisses. Et ce fu por miauz tenir contre le trenchant de l'espée.

Quant il orent Lancelot atorné et bien et bel si lor prie que il s'en aillent. Et il s'en vont, et le font naigier outre l'ève, et il enmainent son cheval. Et il vient à la planche droit: puis esgarde vers la tor où la raïne estoit en prison, si li encline. Après fet le signe de la verroie croiz enmi son vis, et met son escu derriers son dos, qu'il ne li nuise. Lors se met desor la planche en chevauchons, si se traïne par desus si armez comme il estoit, car il ne li faut ne hauberc ne espée ne chauces ne heaume ne escu. Et cil de la tor qui le véoient en sont tuit esbahï, ne il n'i a nul ne nule qui saiche veroiement qui il est; mès qu'il voient qu'il traïne pardesus l'espée trenchant à la force des braz et à l'enpaignement des genouz; si ne remaint pas por les filz de fer que des piez et des mains et des genous ne saille li sanz. Mès por cel péril de l'espée qui trenche et por l'ève noire et bruiant et parfonde ne remaint que plus ne resgart vers la tor que vers l'ève, ne plaie ne angoisse qu'il ait ne prise naient; car se il à cele tor pooit venir il garroit tot maintenant de ses max. Tant s'est hertiez et traïnez qu'il est venuz jusqu'à terre.

This becomes in the poem a passage more than 100 lines long, of which the beginning and end may be given: —

* * * * * *

Romances of Antiquity. Chanson d'Alixandre.

About the same time as the flourishing of the Arthurian cycle there began to be written the third great division of Jean Bodel, 'la matière de Rome la grant55.' The most important beyond all question of the poems which go to make up this cycle (as it is sometimes called, though in reality its members are quite independent one of the other) is the Romance of Alixandre. Of the earliest French poem on this subject only a few fragments exist. This is supposed to have been a work of the eleventh or very early twelfth century, composed in octosyllabic verses, and in the mixed dialect common at the time in the south-east, by Alberic or Auberi of Besançon or Briançon. The Chanson d'Alixandre is, however, in all probability a much more important work than Alberic's. It is in form a regular Chanson de Geste, written in twelve-syllabled verse, of such strength and grace that the term Alexandrine has cleaved ever since to the metre. Its length, as we have it56, is 22,606 verses, and it is assigned to two authors, Lambert the Short57 and Alexander of Bernay, though doubt has been expressed whether any of the present poem is due to Lambert; if we have any of his work, it is not later than the ninth decade of the twelfth century. Lambert, Alexander, and perhaps others, are thought to have known not Alberic, but a later ten-syllabled version into Northern French by Simon of Poitiers. The remoter sources are various. Foremost among them may undoubtedly be placed the Pseudo-Callisthenes, an unknown Alexandrian writer translated into Latin about the fourth century by Julius Valerius, who fathered upon the philosopher a collection of stories partly gathered from Plutarch, Quintus Curtius, and a hundred other authorities, partly elaborated according to the fashion of Greek romancers. Some oriental traditions of Alexander were also in the possession of western Europe. Out of all these, and with a considerable admixture of the floating fables of the time, Lambert and Alexander wove their work. There is, of course, not the slightest attempt at antiquity of colour. Alexander has twelve peers, he learns the favourite studies of the middle ages, he is dubbed knight, and so forth. Many interesting legends, such as that of the Fountain of Perpetual Youth, make their first appearance in the poem, and it is altogether one of extraordinary merit. A specimen laisse may be given: —

While the figure of Alexander served as centre to one group of fictions, most of which were composed in Chanson form, the octosyllabic metre, which had made the Arthurian romances its own, was used for the versification of another numerous class, most of which dealt with the tale of Troy divine.

Roman de Troie.

Here also the poems were neither entirely fictitious, nor on the other hand based upon the best authorities. Dares Phrygius and Dictys Cretensis, with some epitomes of Homer, were the chief sources of information. The principal poem of this class is the Roman de Troie of Benoist de Sainte More (c. 1160). This work58, which extends to more than thirty thousand verses, has the redundancy and the long-windedness which characterise many, if not most, early French poems written in its metre. But it has one merit which ought to conciliate English readers to Benoist. It contains the undoubted original of Shakespeare's Cressida. The fortunes of Cressid (or Briseida, as the French trouvère names her) have been carefully traced out by MM. Moland, Héricault59, and Joly, and form a very curious chapter of literary history. Nor is this episode the only one of merit in Benoist. His verse is always fluent and facile, and not seldom picturesque, as the following extract (Andromache's remonstrance with Hector) will show: —

Other Romances on Classical subjects.

The poems of the Cycle of Antiquity have hitherto been less diligently studied and reprinted than those of the other two. Few of them, with the exception of Alixandre and Troie, are to be read even in fragments, save in manuscript. Le Roman d'Enéas, which is attributed to Benoist, is much shorter than the Roman de Troie, and, with some omissions, follows Virgil pretty closely. Like many other French poems, it was adapted in German by a Minnesinger, Heinrich von Veldeke. Le Roman de Thèbes, of which there is some chance of an edition, stands to Statius in the same relation as Enéas to Virgil. And Le Roman de Jules César paraphrases, though not directly, Lucan. To these must be added Athis et Prophilias (Porphyrias), or the Siege of Athens, a work which has been assigned to many authors, and the origin of which is not clear, though it enjoyed great popularity in the middle ages. The Protesilaus of Hugues de Rotelande is the only other poem of this series worth the mentioning.

Neither of these two classes of poems possesses the value of the Chansons as documents for social history. The picture of manners in them is much more artificial. But the Arthurian romances disclose partially and at intervals a state of society decidedly more advanced than that of the Chansons. The bourgeois, the country gentleman who is not of full baronial rank, and other novel personages appear.

Note to Third Edition.– Since the second edition was published M. Gaston Paris has sketched in Romania and summarised in his Manuel, but has not developed in book form, a view of the Arthurian romances different from his father's and from that given in the text. In this view the importance of 'Celtic' originals is much increased, and that of Geoffrey diminished, Walter Map disappears almost entirely to make room for divers unknown French trouvères, the order of composition is altered, and on the whole a lower estimate is formed of the literary value of the cycle. The 'Celtic' view has also been maintained in a book of much learning and value, Studies on the Legend of the Holy Grail (London, 1888), by Mr. Alfred Nutt. I have not attempted to incorporate or to combat these views in the text for two reasons, partly because they will most probably be superseded by others, and partly because the evidence does not seem to me sufficient to establish any of them certainly. But having given some years to comparative literary criticism in different languages and periods, I think I may be entitled to give a somewhat decided opinion against the 'Celtic' theory, and in favour of that which assigns the special characteristics of the Arthurian cycle and all but a very small part of its structure of incident to the literary imagination of the trouvères, French and English, of the twelfth century. And I may add that as a whole it seems to me quite the greatest literary creation of the Middle Ages, except the Divina Commedia, though of course it has the necessary inferiority of a collection by a great number of different hands to a work of individual genius.