literature of Europe?'-and spent some time in pursuit of so elusive a quarry. My choice of a theme was decided by Edinburgh's connection with the revival of Romance, and my guesses at its origin. I must speak of Romance.

Some may feel that a definition of Romance should precede any survey of its inception and character. I respectfully demur. A definition of Romance would be easy if there were general agreement on the meaning of the word. Unfortunately there is not. Most people if asked, 'What is Romance?' would answer, as Augustine did of Time, I know when you do not ask me.' When dealing with the dimly apprehended we must discover before we can define. Columbus had no map of America.

One way of discovery would be to select an example of obvious, though undefined, Romance, and then to analyse its contents. But that plan if applied, for instance, to Ariosto's Orlando Furioso will be found to lead away from definition rather than towards it. Analysis of extreme romantic types yields a jumble of mythologies, refracted through several layers of history, all more or less distorted and opaque. There is plenty of fighting and lovemaking, a good deal of scenery and weather; and, apart from human interest, there are troops of animals and some strange inhuman forces masquerading as giants and dragons and warlocks. From such confusion a definition does not readily emerge. A better way of discovery is called, I believerather pompously-the historic method. It amounts to this. If you can establish When and Where a thing happened you may be able to guess Why it happened and, even, What it was. Let us, then, post

pone analysis of Romance, and set out by weighing the question with which the Cardinal of Este greeted Ariosto's presentation of his masterpiece. (1510.) The prelate asked the poet, 'quite simply,' 'where he had been for all that rot.' That is what I shall try to discover. If we begin by detecting when, and where, Romance first appeared in Europe we may be able to say why it appeared, and even to hazard a surmise at its nature. But the last is a fearsome enterprise, trenching on metaphysics, as the way is with all inquiry if you push it any distance. I shall seek in the main for origins, and call my address The Springs of Romance in the Literature of Europe.'

You can look for the advent of Romance either in literature that remains and can be studied; or else, in the theories of learned men who infer the pre-existence of earlier literature, that has certainly perished, and may never have been written. They cite the songs in which, Tacitus tells us, the Germans extolled the founders of their race; or the didactic poetry of the Druids, which the Druids were forbidden to write; or they point in later versions to a barbarous handling of stories treated with relative urbanity in earlier versions, and infer from the discrepancy a common origin for both of

more primitive character than either reveals. These deductions from contemporary references to songs that are lost, and from antique touches in later documents, are always ingenious and often delightful. But they present two difficulties. In the first place, hypothetical literature affords a foundation too insecure for the erection of theory that must itself partake of conjecture. In the second place, it is by no means certain that barbarous

legends are romantic to the races who invent them. I shall return to that view before I conclude. At the outset I must look for the advent of Romance in writings that still form part of the literature of Europe.

THE ADVENT OF ROMANCE

Keeping, then, to literature that remains, I advance the disputable proposition that the writings preserved from Greece and Rome are not romantic; briefly, that the classics are not romantic. If time permitted I could, I think, sustain that thesis, with qualifications, of course, and concessions to any who disputed its truth. I would readily admit that the Greeks were more romantic than the Romans. I would certainly concede Nausicaa in the Odyssey and Medea in the Argonauts; Dido and Camilla in the Æneid. But, excepting Virgil, whose peculiar romantic note caught the ear of the Middle Ages, I should point out that my concessions were mainly in respect of the earliest and latest poems of the Classic world, and that, including even the Eneid, all such touches of romance as do faintly transfigure the classics are to be found in stories of wandering through strange lands, and of encounters with alien customs and superstitions. I would give my 'heckler' the Golden Ass of Apuleius, and cut the argument short by taking refuge in the considered opinion of Professor W. P. Ker. He writes (The Dark Ages, p. 41): 'Classical literature perished from a number of contributory ailments, but of these none was more desperate than the want of Romance in the Roman Empire, and especially in the Latin language.'

The Latin world of the fifth century was unromantic, and notably so in northern Gaul, the most Roman, because the least invaded, province of the Western Empire. Latinised Gauls led an ordered existence of unchallenged convention, revolving round garrisons, townhalls, and schools. Their life was military and municipal; their literature, an affair of grammar and rhetoric, written in classical Latin which had diverged from vulgar Latin, so widely as to be unintelligible to all but the learned. From the people's Latin, spoken throughout the country, almost every trace of Celtic words and Celtic beliefs had been eliminated. We possess nothing that can be called Romance in either of these languages. Yet Latin Gaul was to be the nursery-garden of the first seedling of romantic literature, and that earliest growth was not to flourish until it had been transplanted. When, then, and where, does Romance arrive in European literature? The answer to the first question is, not before the second half of the eleventh century, and, to the second, probably in Great Britain. The first piece of obvious Romance in literature that remains is the 'Song of Roland,' as we have it in the Oxford MS. (Bodleian, Digby, 23). The composition of the poem is attributed to a Norman, and the date of it placed between the Norman conquest of England in 1066 and the Crusaders' conquest of Jerusalem in 1099. The handwriting, as distinguished from the composition, is dated about 1170. Romance arrives six centuries after the overthrow of the Western Empire, and appears where a province had been torn from it long before the Latin Gauls had ceased to speak or write in languages derived from Rome. We

know when and where Romance appeared. To understand why it came, and to surmise at what it was, we must sketch in the events of those six centuries which preceded and as I shall urgeprepared for the Advent of Romance after 1050 A.D., and for its rapid development a hundred years later.

In the fifth century two things happened which began the preparation of Gaul to be the nurserygarden of Romance. A Celtic people established themselves in the north-west of Gaul, thenceforward to be called Brittany, where their language is still spoken by the Bretons. They came in numbers, and the territory which they occupied ceased to be Latin. We are told that they sang lays to a little harp, called the rote. But none of their songs appears in literature for centuries. Again in the fifth century, a Teutonic nation, the Francs, invaded the north-east of Gaul, and soon ceased, for the most part, to be German. They were few in number, and their ambition was to be like the Latin aristocracy. Their mother-tongue, after a brief interval, contained more words of Latin than of Teutonic derivation. Their laws were written in learned Latin. Their religion, after 496 A.D., was orthodox Latin Christianity. Clovis, or Chlodoweg-if you like that name better-preferred his title of a Roman patrician' to the glory of his conquests. We are told that the Francs sang the deeds of their kings in poems, accompanied on harps. It may well be so. But none of these poems have ever appeared in literature. They may, or may not, have been romantic. We have no record of Frankish verse, save one. There are eight Latin lines in the life of a saint composed