The Flourishing of Romance and the Rise of Allegory / (Periods of European Literature, vol. II)

But immediately afterwards, and more or loss throughout, he redoubles and redoubles again every possible artifice—sound-repetition in the imminet, imminet, of the third line, alliteration in the recta remuneret of the fourth, and everywhere trills and roulades, not limited to the actually rhyming syllables of the same vowel—

He has instinctively discovered the necessity of varying as much as possible the cadence and composition of the last third of his verse, and carefully avoids anything like a monotonous use of his only spondee; in a batch of eighteen lines taken at random, there are only six end-words of two syllables, and these only once rhyme together. The consequence of these and other devices is that the whole poem is accompanied by a sort of swirl and eddy of sound and cadence, constantly varying, constantly shifting its centres and systems, but always assisting the sense with grateful clash or murmur, according as it is loud or soft, of word-music.

The vernacular languages were not as yet in case to produce anything so complicated as this, and some of them have never been quite able to produce it to this day. But it must be obvious at once what a standard was held up before poets, almost every one of whom, even if he had but small Latin in a general way, heard these hymns constantly sung, and what means of producing like effects were suggested to them. The most varied and charming lyric of the Middle Ages, that of the German Minnesingers, shows the effect of this Latin practice side by side, or rather inextricably mingled, with the effects of the preciser French and Provençal verse-scheme, and the still looser but equally musical, though half-inarticulate, suggestions of indigenous song. That English prosody—the prosody of Shakespeare and Coleridge, of Shelley and Keats—owes its origin to a similar admixture the present writer at least has no doubt at all, while even those who deny this can hardly deny the positive literary achievement of the best mediæval hymns. They stand by themselves. Latin—which, despite its constant colloquial life, still even in the Middle Ages had in profane use many of the drawbacks of a dead language, being either slipshod or stiff,—here, owing to the millennium and more during which it had been throughout Western Europe the living language and the sole living language of the Church Universal, shakes off at once all artificial and all doggerel character. It is thoroughly alive: it comes from the writers' hearts as easily as from their pens. They have in the fullest sense proved it; they know exactly what they can do, and in this particular sphere there is hardly anything that they cannot do.

The far-famed and almost more abused than famed Scholastic Philosophy[11] cannot be said to have added to positive literature any such masterpieces in prose as the hymn-writers (who were very commonly themselves Scholastics) produced in verse. With the exception of Abelard, whose interest is rather biographical than strictly literary, and perhaps Anselm, the heroes of mediæval dialectic, the Doctors Subtle and Invincible, Irrefragable and Angelic, have left nothing which even on the widest interpretation of pure literature can be included within it, or even any names that figure in any but the least select of literary histories. Yet they cannot but receive some notice here in a history, however condensed, of the literature of the period of their chief flourishing. This is not because of their philosophical importance, although at last, after much bandying of not always well-informed argument, that importance is pretty generally allowed by the competent. It has, fortunately, ceased to be fashionable to regard the dispute about Universals as proper only to amuse childhood or beguile dotage, and the quarrels of Scotists and Thomists as mere reductions of barren logomachy to the flatly absurd. Still, this importance, though real, though great, is not directly literary. The claim which makes it impossible to pass them over here is that excellently put in the two passages from Condorcet and Hamilton which John Stuart Mill (not often a scholastically minded philosopher) set in the forefront of his Logic, that, in the Scottish philosopher's words, "it is to the schoolmen that the vulgar languages are indebted for what precision and analytical subtlety they possess;" and that, as the Frenchman, going still further, but hardly exaggerating, lays it down, "logic, ethics, and metaphysics itself owe to Scholasticism a precision unknown to the ancients themselves."

There can be no reasonable or well-informed denial of the fact of this: and the reason of it is not hard to understand. That constant usage, the effect of which has been noted in theological verse, had the same effect in philosophico-theological prose. Latin is before all things a precise language, and the one qualification which it lacked in classical times for philosophic use, the presence of a full and exact terminology, was supplied in the Middle Ages by the fearless barbarism (as pedants call it) which made it possible and easy first to fashion such words as aseitas and quodlibetalis, and then, after, as it were, lodging a specification of their meaning, to use them ever afterwards as current coin. All the peculiarities which ignorance or sciolism used to ridicule or reproach in the Scholastics—their wiredrawnness, their lingering over special points of verbal wrangling, their neglect of plain fact in comparison with endless and unbridled dialectic—all these things did no harm but much positive good from the point of view which we are now taking. When a man defended theses against lynx-eyed opponents or expounded them before perhaps more lynx-eyed pupils, according to rules familiar to all, it was necessary for him, if he were to avoid certain and immediate discomfiture, to be precise in his terms and exact in his use of them. That it was possible to be childishly as well as barbarously scholastic nobody would deny, and the famous sarcasms of the Epistolæ Obscurorum Virorum, two centuries after our time, had been anticipated long before by satirists. But even the logical fribble, even the logical jargonist, was bound to be exact. Now exactness was the very thing which languages, mostly young in actual age, and in all cases what we may call uneducated, unpractised in literary exercises, wanted most of all. And it was impossible that they should have better teachers in it than the few famous, and even than most of the numerous unknown or almost unknown, philosophers of the Scholastic period.

It has been said that of those most famous almost all belong specially to this our period. Before it there is, till its very latest eve, hardly one except John Scotus Erigena; after it none, except Occam, of the very greatest. But during the twelfth and thirteenth centuries there is scarcely a decade without its illustration. The first champions of the great Realist and Nominalist controversy, Roscellinus and William of Champeaux, belong to the eleventh century in part, as does their still more famous follower, Abelard, by the first twenty years of his life, while almost the whole of that of Anselm may be claimed by it.[12] But it was not till the extreme end of that century that the great controversy in which these men were the front-fighters became active (the date of the Council of Soissons, which condemned the Nominalism of Roscellinus as tritheistic is 1092), and the controversy itself was at its hottest in the earlier part of the succeeding age. The Master of the Sentences, Peter Lombard, belongs wholly to the twelfth, and the book which gives him his scholastic title dates from its very middle. John of Salisbury, one of the clearest-headed as well as most scholarly of the whole body, died in 1180. The fuller knowledge of Aristotle, through the Arabian writers, coincided with the latter part of the twelfth century: and the curious outburst of Pantheism which connects itself on the one hand with the little-known teaching of Amaury de Bène and David of Dinant, on the other with the almost legendary "Eternal Gospel" of Joachim of Flora, occurred almost exactly at the junction of the twelfth and thirteenth. As for the writers of the thirteenth century itself, that great period holds in this as in other departments the position of palmiest time of the Middle Ages. To it belong Alexander Hales, who disputes with Aquinas the prize for the best example of the Summa Theologiæ; Bonaventura, the mystic; Roger Bacon, the natural philosopher; Vincent of Beauvais, the encyclopædist. If, of the four greatest of all, Albert of Bolstadt, Albertus Magnus, the "Dumb Ox of Cologne," was born seven years before its opening, his life lasted over four-fifths of it; that of Aquinas covered its second and third quarters; Occam himself, though his main exertions lie beyond us, was probably born before Aquinas died; while John Duns Scotus hardly outlived the century's close by a decade. Raymond Lully (one of the most characteristic figures of Scholasticism and of the mediæval period, with his "Great Art" of automatic philosophy), who died in 1315, was born as early as 1235. Peter the Spaniard, Pope and author of the Summulæ Logicales, the grammar of formal logic for ages, died in 1277.

Of the matter which these and others by hundreds put in forgotten wealth of exposition, no account will be expected here. Even yet it is comparatively unexplored, or else the results of the exploration exist only in books brilliant, but necessarily summary, like that of Hauréau, in books thorough, but almost as formidable as the original, like that of Prantl. Even the latest historians of philosophy complain that there is up to the present day no "ingoing" (as the Germans say) monograph about Scotus and none about Occam.[13] The whole works of the latter have never been collected at all: the twelve mighty volumes which represent the compositions of the former contain probably not the whole work of a man who died before he was forty. The greater part of the enormous mass of writing which was produced, from Scotus Erigena in the ninth century to Gabriel Biel in the fifteenth, is only accessible to persons with ample leisure and living close to large and ancient libraries. Except Erigena himself, Anselm in a few of his works, Abelard, and a part of Aquinas, hardly anything can be found in modern editions, and even the zealous efforts of the present Pope have been less effectual in divulging Aquinas than those of his predecessors were in making Amaury of Bena a mystery.[14] Yet there has always, in generous souls who have some tincture of philosophy, subsisted a curious kind of sympathy and yearning over the work of these generations of mainly disinterested scholars who, whatever they were, were thorough, and whatever they could not do, could think. And there have even, in these latter days, been some graceless ones who have asked whether the Science of the nineteenth century, after an equal interval, will be of any more positive value—whether it will not have even less comparative interest than that which appertains to the Scholasticism of the thirteenth.

However this may be, the claim, modest and even meagre as it may seem to some, which has been here once more put forward for this Scholasticism—the claim of a far-reaching educative influence in mere language, in mere system of arrangement and expression, will remain valid. If, at the outset of the career of modern languages, men had thought with the looseness of modern thought, had indulged in the haphazard slovenliness of modern logic, had popularised theology and vulgarised rhetoric, as we have seen both popularised and vulgarised since, we should indeed have been in evil case. It used to be thought clever to moralise and to felicitate mankind over the rejection of the stays, the fetters, the prison in which its thought was mediævally kept. The justice or the injustice, the taste or the vulgarity, of these moralisings, of these felicitations, may not concern us here. But in expression, as distinguished from thought, the value of the discipline to which these youthful languages were subjected is not likely now to be denied by any scholar who has paid attention to the subject. It would have been perhaps a pity if thought had not gone through other phases; it would certainly have been a pity if the tongues had all been subjected to the fullest influence of Latin constraint. But that the more lawless of them benefited by that constraint there can be no doubt whatever. The influence of form which the best Latin hymns of the Middle Ages exercised in poetry, the influence in vocabulary and in logical arrangement which Scholasticism exercised in prose, are beyond dispute: and even those who will not pardon literature, whatever its historical and educating importance be, for being something less than masterly in itself, will find it difficult to maintain the exclusion of the Cur Deus Homo, and impossible to refuse admission to the Dies Iræ.

CHAPTER II.

CHANSONS DE GESTE.[15]

When we turn from Latin and consider the condition of the vernacular tongues in the year 1100, there is hardly more than one country in Europe where we find them producing anything that can be called literature. In England Anglo-Saxon, if not exactly dead, is dying, and has for more than a century ceased to produce anything of distinctly literary attraction; and English, even the earliest "middle" English, is scarcely yet born, is certainly far from being in a condition for literary use. The last echoes of the older and more original Icelandic poetry are dying away, and the great product of Icelandic prose, the Saga, still volitat per ora virum, without taking a concrete literary form. It is in the highest degree uncertain whether anything properly to be called Spanish or Italian exists at all—anything but dialects of the lingua rustica showing traces of what Spanish and Italian are to be; though the originals of the great Poema del Cid cannot be far off. German is in something the same trance between its "Old" and its "Middle" state as is English. Only in France, and in both the great divisions of French speech, is vernacular literature active. The northern tongue, the langue d'oïl, shows us—in actually known existence, or by reasonable inference that it existed—the national epic or chanson de geste; the southern, or langue d'oc, gives us the Provençal lyric. The latter will receive treatment later, the former must be dealt with at once.

It is rather curious that while the chansons de geste are, after Anglo-Saxon and Icelandic poetry, the oldest elaborate example of verse in the modern vernaculars; while they exhibit a character, not indeed one of the widest in range or most engaging in quality, but individual, interesting, intense as few others; while they are entirely the property of one nation, and that a nation specially proud of its literary achievements,—they were almost the last division of European literature to become in any degree properly known. In so far as they were known at all, until within the present century, the knowledge was based almost entirely on later adaptations in verse, and still later in prose; while—the most curious point of all—they were not warmly welcomed by the French even after their discovery, and cannot yet be said to have been taken to the heart of the nation, even to the limited extent to which the Arthurian romances have been taken to the heart of England, much less to that in which the old, but much less old, ballads of England, Scotland, Germany, and Spain have for periods of varying length been welcomed in their respective countries. To discuss the reason of this at length would lead us out of our present subject; but it is a fact, and a very curious fact.

The romances of Charlemagne, or, to employ their more technical designation, the chansons de geste, form a large, a remarkably homogeneous, and a well-separated body of compositions. These, as far as can be decided, date in time from the eleventh to the thirteenth century, with a few belated representatives in the fourteenth; but scarcely, as far as probability shows, with any older members in the tenth. Very little attention of any kind was paid to them, till some seventy years ago, an English scholar, Their age and
history.Conybeare, known for his services to our own early literature, following the example of another scholar, Tyrwhitt, still earlier and more distinguished, had drawn attention to the merit and interest of, as it happens, the oldest and most remarkable of all. This was the Chanson de Roland, which, in this oldest form, exists only in one of the MSS. of the Bodleian Library at Oxford. But they very soon received the care of M. Paulin Paris, the most indefatigable student that in a century of examination of the older European literature any European country has produced, and after more than half a century of enthusiastic resuscitation by M. Paris, by his son M. Gaston, and by others, the whole body of them has been thoroughly overhauled and put at the disposal of those who do not care to read the original, in the four volumes of the remodelled edition of M. Léon Gautier's Epopées Françaises, while perhaps a majority of the actual texts are in print. This is as well, for though a certain monotony is always charged against the chansons de geste[16] by those who do not love them, and may be admitted to some extent even by those who do, there are few which have not a more or less distinct character of their own; and even the generic character is not properly to be perceived until a considerable number have been studied.

The old habit of reading this division of romance in late and travestied versions naturally and necessarily obscured the curious traits of community in form and matter that belong to it, and indeed distinguish it from almost all other departments of literature of the imaginative kind. Its members are frequently spoken of as "the Charlemagne Romances"; and, as a matter of fact, most of them do come into connection with the great prince of the second race in one way or another. Yet Bodel's phrase of matière de France[17] is happier. For they are all still more directly connected with French history, seen through a romantic lens; and even the late and half-burlesque Hugues Capet, even the extremely interesting and partly contemporary set on the Crusades, as well as such "little gestes" as that of the Lorrainers, Garin le Loherain and the rest, and the three "great gestes" of the king, of the southern hero William of Orange (sometimes called the geste of Montglane), and of the family of Doon de Mayence, arrange themselves with no difficulty under this more general heading. And the chanson de geste proper, as Frenchmen are entitled to boast, never quite deserts this matière de France. It is always the Gesta Francorum at home, or the Gesta Dei per Francos in the East, that supply the themes. When this subject or group of subjects palled, the very form of the chanson de geste was lost. It was not applied to other things;[18] it grew obsolete with that which it had helped to make popular. Some of the material—Huon of Bordeaux, the Four Sons of Aymon, and others—retained a certain vogue in forms quite different, and gave later ages the inexact and bastard notion of "Charlemagne Romance" which has been referred to. But the chanson de geste itself was never, so to speak, "half-known"—except to a very few antiquaries. After its three centuries of flourishing, first alone, then with the other two "matters," it retired altogether, and made its reappearance only after four centuries had passed away.

This fact or set of facts has made the actual nature of the original Charlemagne Romances the subject of much mistake and misstatement on the part of general historians of literature. The widely read and generally accurate Dunlop knew nothing whatever about them, except in early printed versions representing their very latest form, and in the hopelessly travestied eighteenth-century Bibliothèque des Romans of the Comte de Tressan. He therefore assigned to them[19] a position altogether inferior to their real importance, and actually apologised for the writers, in that, coming after the Arthurian historians, they were compelled to imitation. As a matter of fact, it is probable that all the most striking and original chansons de geste, certainly all those of the best period, were in existence before a single one of the great Arthurian romances was written; and as both the French and English, and even the German, writers of these latter were certainly acquainted with the chansons, the imitation, if there were any, must lie on their side. As a matter of fact, however, there is little or none. The later and less genuine chansons borrow to some extent the methods and incidents in the romances; but the romances at no time exhibit much resemblance to the chansons proper, which have an extremely distinct, racy, and original character of their own. Hallam, writing later than Dunlop, and if with a less wide knowledge of Romance, with a much greater proficiency in general literary history, practically passes the chansons de geste over altogether in the introduction to his Literature of Europe, which purports to summarise all that is important in the History of the Middle Ages, and to supplement and correct that book itself.

The only excuse (besides mere unavoidable ignorance, which, no doubt, is a sufficient one) for this neglect is the curious fact, in itself adding to their interest, that these chansons, though a very important chapter in the histories both of poetry and of fiction, form one which is strangely marked off at both ends from all connection, save in point of subject, with literature precedent or subsequent. As to their own origin, the usual abundant, warm, and if it may be said without impertinence, rather futile controversies have prevailed. Practically speaking, we know nothing whatever about the matter. There used to be a theory that the Charlemagne Romances owed their origin more or less directly to the fabulous Chronicle of Tilpin or Turpin, the warrior-Archbishop of Rheims. It has now been made tolerably certain that the Latin chronicle on the subject is not anterior even to our existing Chanson de Roland, and very probable that it is a good deal later. On the other hand, of actual historical basis we have next to nothing except the mere fact of the death of Roland ("Hruotlandus comes Britanniæ") at the skirmish of Roncesvalles. There are, however, early mentions of certain cantilenæ or ballads; and it has been assumed by some scholars that the earliest chansons were compounded out of precedent ballads of the kind. It is unnecessary to inform those who know something of general literary history, that this theory (that the corruption of the ballad is the generation of the epic) is not confined to the present subject, but is one of the favourite fighting-grounds of a certain school of critics. It has been applied to Homer, to Beowulf, to the Old and Middle German Romances, and it would be very odd indeed if it had not been applied to the Chansons de geste. But it may be said with some confidence that not one tittle of evidence has ever been produced for the existence of any such ballads containing the matter of any of the chansons which do exist. The song of Roland which Taillefer sang at Hastings may have been such a ballad: it may have been part of the actual chanson; it may have been something quite different. But these "mays" are not evidence; and it cannot but be thought a real misfortune that, instead of confining themselves to an abundant and indeed inexhaustible subject, the proper literary study of what does exist, critics should persist in dealing with what certainly does not, and perhaps never did. On the general point it might be observed that there is rather more positive evidence for the breaking up of the epic into ballads than for the conglomeration of ballads into the epic. But on that point it is not necessary to take sides. The matter of real importance is, to lay it down distinctly that we have nothing anterior to the earliest chansons de geste; and that we have not even any satisfactory reason for presuming that there ever was anything.

One of the reasons, however, which no doubt has been most apt to suggest anterior compositions is the singular completeness of form exhibited by these poems. It is now practically agreed that—scraps and fragments themselves excepted—we have no monument of French in accomplished profane literature more ancient than the Chanson de Roland.[20] And the form of this, though from one point of view it may be called rude and simple, is of remarkable perfection in its own way. The poem is written in decasyllabic iambic lines with a cæsura at the second foot, these lines being written with a precision which French indeed never afterwards lost, but which English did not attain till Chaucer's day, and then lost again for more than another century. Further, the grouping and finishing of these lines is not less remarkable, and is even more distinctive than their internal construction. They are not blank; they are not in couplets; they are not in equal stanzas; and they are not (in the earliest examples, such as Roland) regularly rhymed. But they are arranged in batches (called in French laisses or tirades) of no certain number, but varying from one to several score, each of which derives unity from an assonance—that is to say, a vowel-rhyme, the consonants of the final syllable varying at discretion. This assonance, which appears to have been common to all Romance tongues in their early stages, disappeared before very long from French, though it continued in Spanish, and is indeed the most distinguishing point of the prosody of that language. Very early in the chansons themselves we find it replaced by rhyme, which, however, remains the same for the whole of the laisse, no matter how long it is. By degrees, also, the ten-syllabled line (which in some examples has an octosyllabic tail-line not assonanced at the end of every laisse) gave way in its turn to the victorious Alexandrine. But the mechanism of the chanson admitted no further extensions than the substitution of rhyme for assonance, and of twelve-syllabled lines for ten-syllabled. In all other respects it remained rigidly the same from the eleventh century to the fourteenth, and in the very latest examples of such poems, as Hugues Capet and Baudouin de Seboure—full as enthusiasts like M. Gautier complain that they are of a spirit very different from that of the older chansons—there is not the slightest change in form; while certain peculiarities of stock phrase and "epic repetition" are jealously preserved. The immense single-rhymed laisses, sometimes extending to several pages of verse, still roll rhyme after rhyme with the same sound upon the ear. The common form generally remains; and though the adventures are considerably varied, they still retain a certain general impress of the earlier scheme.

That scheme is, in the majority of the chansons, curiously uniform. It has, since the earliest studies of them, been remarked as odd that Charlemagne, though almost omnipresent (except of course in the Crusading cycle and a few others), and though such a necessary figure that he is in some cases evidently confounded both with his ancestor Charles Martel and his successor Charles the Bald, plays a part that is very dubiously heroic. The character of Charlemagne. He is, indeed, presented with great pomp and circumstance as li empereres à la barbe florie, with a gorgeous court, a wide realm, a numerous and brilliant baronage. But his character is far from tenderly treated. In Roland itself he appears so little that critics who are not acquainted with many other poems sometimes deny the characteristic we are now discussing. But elsewhere he is much less leniently handled. Indeed the plot of very many chansons turns entirely on the ease with which he lends an ear to traitors (treason of various kinds plays an almost ubiquitous part, and the famous "trahis!" is heard in the very dawn of French literature), on his readiness to be biassed by bribes, and on the singular ferocity with which, on the slightest and most unsupported accusation, he is ready to doom any one, from his own family downwards, to block, stake, gallows, or living grave. This combination, indeed, of the irascible and the gullible tempers in the king defrays the plot of a very large number of the chansons, in which we see his best knights, and (except that they are as intolerant of injustice as he is prone to it) his most faithful servants, forced into rebellion against him, and almost overwhelmed by his own violence following on the machinations of their and his worst enemies.

Nevertheless, Charlemagne is always the defender of the Cross, and the antagonist of the Saracens, and the part which these latter play is as ubiquitous as his own, and on the whole more considerable. A very large part of the earlier chansons is occupied with direct fighting against the heathen; and from an early period (at least if the Voyage à Constantinoble is, as is supposed, of the early twelfth century, if not the eleventh) a most important element, bringing the class more into contact with romance generally than some others which have been noticed, is introduced in the love of a Saracen princess, daughter of emperor or "admiral" (emir), for one of the Christian heroes. Here again Roland stands alone, and though the mention of Aude, Oliver's sister and Roland's betrothed, who dies when she hears of his death, is touching, it is extremely meagre. There is practically nothing but the clash of arms in this remarkable poem. But elsewhere there is, in rather narrow and usual limits, a good deal else. Charlemagne's daughter, and the daughters of peers and paladins, figure: and their characteristics are not very different from those of the pagan damsels. It is, indeed, unnecessary to convert them,—a process to which their miscreant sisters usually submit with great goodwill,—and they are also relieved from the necessity of showing the extreme undutifulness to their more religiously constant sires, which is something of a blot on Paynim princesses like Floripas in Fierabras. This heroine exclaims in reference to her father, "He is an old devil, why do you not kill him? little I care for him provided you give me Guy," though it is fair to say that Fierabras himself rebukes her with a "Moult grant tort avès." All these ladies, however, Christian as well as heathen, are as tender to their lovers as they are hard-hearted to their relations; and the relaxation of morality, sometimes complained of in the later chansons, is perhaps more technical than real, even remembering the doctrine of the mediæval Church as to the identity, for practical purposes, of betrothal and marriage. On the other hand, the courtesy of the chansons is distinctly in a more rudimentary state than that of the succeeding romances. Not only is the harshest language used by knights to ladies,[21] but blows are by no means uncommon; and of what is commonly understood by romantic love there is on the knights' side hardly a trace, unless it be in stories such as that of Ogier le Danois, which are obviously late enough to have come under Arthurian influence. The piety, again, which has been so much praised in these chansons, is of a curious and rather elementary type. The knights are ready enough to fight to the last gasp, and the last drop of blood, for the Cross; and their faith is as free from flaw as their zeal. Li Apostoiles de Rome—the Pope—is recognised without the slightest hesitation as supreme in all religious and most temporal matters. But there is much less reference than in the Arthurian romances, not merely to the mysteries of the Creed, but even to the simple facts of the birth and death of Christ. Except in a few places—such as, for instance, the exquisite and widely popular story of Amis and Amiles (the earliest vernacular form of which is a true chanson de geste of the twelfth century)—there are not many indications of any higher or finer notion of Christianity than that which is confined to the obedient reception of the sacraments, and the cutting off Saracens' heads whensoever they present themselves.[22]

In manners, as in theology and ethics, there is the same simplicity, which some have called almost barbarous. Architecture and dress receive considerable attention; but in other ways the arts do not seem to be far advanced, and living is still conducted nearly, if not quite, as much in public as in the Odyssey or in Beowulf. The hall is still the common resort of both sexes by day and of the men at night. Although gold and furs, silk and jewels, are lavished with the usual cheap magnificence of fiction, very few details are given of the minor supellex or of ways of living generally. From the Chanson de Roland in particular (which, though it is a pity to confine the attention to it as has sometimes been done, is undoubtedly the type of the class in its simplest and purest form) we should learn next to nothing about the state of society depicted, except that its heroes were religious in their fashion, and terrible fighters. But it ought to be added that the perusal of a large number of these chansons leaves on the mind a much more genuine belief in their world (if it may so be called) as having for a time actually existed, than that which is created by the reading of Arthurian romance. That fair vision we know (hardly knowing why or how we know it) to have been a creation of its own Fata Morgana, a structure built of the wishes, the dreams, the ideals of men, but far removed from their actual experience. This is not due to miracles—there are miracles enough in the chansons de geste most undoubtingly related: nor to the strange history, geography, and chronology, for the two divisions are very much on a par there also. But strong as the fantastic element is in them, the chansons de geste possess a realistic quality which is entirely absent from the gracious idealism of the Romances. The emperors and the admirals, perhaps even their fair and obliging daughters, were not personages unknown to the contemporaries of the Norman conquerors of Italy and Sicily, or to the first Crusaders. The faithful and ferocious, covetous and indomitable, pious and lawless spirit, which hardly dropped the sword except to take up the torch, was, poetic presentation and dressing apart, not so very different from the general temper of man after the break up of the Roman peace till the more or less definite mapping out of Europe into modern divisions. More than one Vivien and one William of Orange listened to Peter the Hermit. In the very isolation of the atmosphere of these romances, in its distance from modern thought and feeling, in its lack (as some have held) of universal quality and transcendent human interest, there is a certain element of strength. It was not above its time, and it therefore does not reach the highest forms of literature. But it was intensely of its time; and thus it far exceeds the lowest kinds, and retains an abiding value even apart from the distinct, the high, and the very curious perfection, within narrow limits, of its peculiar form.

It is probable that very few persons who are not specially acquainted with the subject are at all aware of the enormous bulk and number of these poems, even if their later remaniements (as they are called) both in verse and prose—fourteenth and fifteenth century refashionings, which in every case meant a large extension—be left out of consideration. The most complete list published, that of M. Léon Gautier, enumerates 110. Of these he himself places only the Chanson de Roland in the eleventh century, perhaps as early as the Norman Conquest of England, certainly not later than 1095. Twelfth century. To the twelfth he assigns (and it may be observed that, enthusiastic as M. Gautier is on the literary side, he shows on all questions of age, &c., a wariness not always exhibited by scholars more exclusively philological) Acquin, Aliscans, Amis et Amiles, Antioche Aspremont, Auberi le Bourgoing, Aye d'Avignon, the Bataille Loquifer, the oldest (now only known in Italian) form of Berte aus grans Piés, Beuves d'Hanstone (with another Italian form more or less independent), the Charroi de Nîmes, Les Chétifs, the Chevalerie Ogier de Danemarche, the Chevalerie Vivien (otherwise known as Covenant Vivien), the major part (also known by separate titles) of the Chevalier au Cygne, La Conquête de la Petite Bretagne (another form of Acquin), the Couronnement Loys, Doon de la Roche, Doon de Nanteuil, the Enfances Charlemagne, the Enfances Godefroi, the Enfances Roland, the Enfances Ogier, Floovant, Garin le Loherain, Garnier de Nanteuil, Giratz de Rossilho, Girbert de Metz, Gui de Bourgogne, Gui de Nanteuil, Hélias, Hervis de Metz, the oldest form of Huon de Bordeaux, Jérusalem, Jourdains de Blaivies, the Lorraine cycle, including Garin, &c., Macaire, Mainet, the Moniage Guillaume, the Moniage Rainoart, Orson de Beauvais, Rainoart, Raoul de Cambrai, Les Saisnes, the Siège de Barbastre, Syracon, and the Voyage de Charlemagne. In other words, nearly half the total number date from the twelfth century, if not even earlier.

By far the larger number of the rest are not later than the thirteenth. They include—Aimeri de Narbonne, Aiol, Anséis de Carthage, Anséis Fils de Gerbert, Auberon, Berte aus grans Piés in its present French form, Beton et Daurel, Beuves de Commarchis, the Département des Enfans Aimeri, the Destruction de Rome, Doon de Mayence, Elie de Saint Gilles, the Enfances Doon de Mayence, the Enfances Guillaume, the Enfances Vivien, the Entrée en Espagne, Fierabras, Foulques de Candie, Gaydon, Garin de Montglane, Gaufrey, Gérard de Viane, Guibert d'Andrenas, Jehan de Lanson, Maugis d'Aigremont, the Mort Aimeri de Narbonne, Otinel, Parise la Duchesse, the Prise de Cordres, the Prise de Pampelune, the Quatre Fils d'Aymon, Renaud de Montauban (a variant of the same), Renier, the later forms of the Chanson de Roland, to which the name of Roncevaux is sometimes given for the sake of distinction, the Siège de Narbonne, Simon de Pouille, Vivien l'Amachour de Montbranc, and Yon.

By this the list is almost exhausted. The fourteenth century, though fruitful in remaniements, sometimes in mono-rhymed tirades, but often in Alexandrine couplets and other changed shapes, contributes hardly anything original except the very interesting and rather brilliant last branches of the Chevalier au Cygne—Baudouin de Seboure, and the Bastart de Bouillon; Hugues Capet, a very lively and readable but slightly vulgar thing, exhibiting an almost undisguised tone of parody; and some fragments known by the names of Hernaut de Beaulande, Renier de Gennes, &c. As for fifteenth and sixteenth century work, though some pieces of it, especially the very long and unprinted poem of Lion de Bourges, are included in the canon, all the chanson-production of this time is properly apocryphal, and has little or nothing left of the chanson spirit, and only the shell of the chanson form.

It must further be remembered that, with the exception of a very few in fragmentary condition, all these poems are of great length. Only the later or less genuine, indeed, run to the preposterous extent of twenty, thirty, or (it is said in the case of Lion de Bourges) sixty thousand lines. But Roland itself, one of the shortest, has four thousand; Aliscans, which is certainly old, eight thousand; the oldest known form of Huon, ten thousand. It is probably not excessive to put the average length of the older chansons at six thousand lines; while if the more recent be thrown in, the average of the whole hundred would probably be doubled.

This immense body of verse, which for many reasons it is very desirable to study as a whole, is still, after the best part of a century, to a great extent unprinted, and (as was unavoidable) such of its constituents as have been sent to press have been dealt with on no very uniform principles. It was less inevitable, and is more to be regretted, that the dissensions of scholars on minute philological points have caused the repeated printing of certain texts, while others have remained inaccessible; and it cannot but be regarded as a kind of petty treason to literature thus to put the satisfaction of private crotchets before the "unlocking of the word-hoard" to the utmost possible extent. The earliest chansons printed[23] were, I believe, M. Paulin Paris's Berte aus grans Piés, M. Francisque Michel's Roland; and thereafter these two scholars and others edited for M. Techener a very handsome set of "Romances des Douze Pairs," as they were called, including Les Saisnes, Ogier, Raoul de Cambrai, Garin, and the two great crusading chansons, Antioche and Jérusalem. Other scattered efforts were made, such as the publication of a beautiful edition of Baudouin de Seboure at Valenciennes as early as 1841; while a Belgian scholar, M. de Reiffenberg, published Le Chevalier au Cygne, and a Dutch one, Dr Jonckbloët, gave a large part of the later numbers of the Garin de Montglane cycle in his Guillaume d'Orange (2 vols., The Hague, 1854). But the great opportunity came soon after the accession of Napoleon III., when a Minister favourable to literature, M. de Fourtou, gave, in a moment of enthusiasm, permission to publish the entire body of the chansons. Perfect wisdom would probably have decreed the acceptance of the godsend by issuing the whole, with a minimum of editorial apparatus, in some such form as that of our Chalmers's Poets, the bulk of which need probably not have been exceeded in order to give the oldest forms of every real chanson from Roland to the Bastart de Bouillon. But perfect wisdom is not invariably present in the councils of men, and the actual result took the form of ten agreeable little volumes, in the type, shape, and paper of the "Bibliothèque Elzévirienne" with abundant editorial matter, paraphrases in modern French, and the like. Les Anciens Poètes de la France, as this series was called, appeared between 1858, which saw the first volume, and 1870, which fatal year saw the last, for the Republic had no money to spare for such monarchical glories as the chansons. They are no contemptible possession; for the ten volumes give fourteen chansons of very different ages, and rather interestingly representative of different kinds. But they are a very small portion of the whole, and in at least one instance, Aliscans, they double on a former edition. Since then the Société des Anciens Textes Français has edited some chansons, and independent German and French scholars have given some more; but no systematic attempt has been made to fill the gaps, and the pernicious system of re-editing, on pretext of wrong selection of MSS. or the like, has continued. Nevertheless, the number of chansons actually available is so large that no general characteristic is likely to have escaped notice; while from the accounts of the remaining MSS., it would not appear that any of those unprinted can rank with the very best of those already known. Among these very best I should rank in alphabetical order—Aliscans, Amis et Amiles, Antioche, Baudouin de Seboure (though in a mixed kind), Berte aus grans Piés, Fierabras, Garin le Loherain, Gérard de Roussillon, Huon de Bordeaux, Ogier de Danemarche, Raoul de Cambrai, Roland, and the Voyage de Charlemagne à Constantinoble. The almost solitary eminence assigned by some critics to Roland is not, I think, justified, and comes chiefly from their not being acquainted with many others; though the poem has undoubtedly the merit of being the oldest, and perhaps that of presenting the chanson spirit in its best and most unadulterated, as well as the chanson form at its simplest, sharpest, and first state. Nor is there anywhere a finer passage than the death of Roland, though there are many not less fine.

It may, however, seem proper, if not even positively indispensable, to give some more general particulars about these chansons before analysing specimens or giving arguments of one or more; for they are full of curiosities.