PREFACE

It is somewhat curious that there is, so far as I know, no complete handling in English of the subject of this volume, popular and important though that subject has been. Dunlop's History of Fiction, an excellent book, dealt with a much wider matter, and perforce ceased its dealing just at the beginning of the most abundant and brilliant development of the English division. Sir Walter Raleigh's English Novel, a book of the highest value for acute criticism and grace of style, stops short at Miss Austen, and only glances, by a sort of anticipation, at Scott. The late Mr. Sidney Lanier's English Novel and the Principle of its Development is really nothing but a laudatory study of "George Eliot," with glances at other writers, including violent denunciations of the great eighteenth-century men. There are numerous monographs on parts of the subject: but nothing else that I know even attempting the whole. I should, of course, have liked to deal with so large a matter in a larger space: but one may and should "cultivate the garden" even if it is not a garden of many acres in

extent. I need only add that I have endeavoured, not so much to give "reviews" of individual books and authors, as to indicate what Mr. Lanier took for the second part of his title, but did not, I think, handle very satisfactorily in his text.

I may perhaps add, without impropriety, that the composition of this book has not been hurried, and that I have taken all the pains I could, by revision and addition as it proceeded, to make it a complete survey of the Novel, as it has come from the hands of all the more important novelists, not now alive, up to the end of the nineteenth century.

GEORGE SAINTSBURY.

Christmas, 1912.

CONTENTS

I. THE FOUNDATION IN ROMANCE II. FROM LYLY TO SWIFT III. THE FOUR WHEELS OF THE NOVEL WAIN IV. THE MINOR AND LATER EIGHTEENTH-CENTURY NOVEL V. SCOTT AND MISS AUSTEN VI. THE SUCCESSORS—TO THACKERAY VII. THE MID-VICTORIAN NOVEL VIII. THE FICTION OF YESTERDAY—CONCLUSION INDEX

THE FOUNDATION IN ROMANCE

One of the best known, and one of the least intelligible, facts of literary history is the lateness, in Western European Literature at any rate, of prose fiction, and the comparative absence, in the two great classical languages, of what we call by that name. It might be an accident, though a rather improbable one, that we have no Greek prose fiction till a time long subsequent to the Christian era, and nothing in Latin at all except the fragments of Petronius and the romance of Apuleius. But it can be no accident, and it is a very momentous fact, that, from the foundation of Greek criticism, "Imitation," that is to say "Fiction" (for it is neither more nor less), was regarded as not merely the inseparable but the constituent property of poetry, even though those who held this were doubtful whether poetry must necessarily be in verse. It is another fact of the greatest importance that the ancients who, in other forms than deliberate prose fiction, try to "tell a story," do not seem to know very well how to do it.

The Odyssey is, indeed, one of the greatest of all stories, it is the original romance of the West; but the Iliad, though a magnificent poem, is not much of a story. Herodotus can tell one, if anybody can, and Plato (or Socrates) evidently could have done so if it had lain in his way: while the Anabasis, though hardly the Cyropædia, shows glimmerings in Xenophon. But otherwise we must come down to Lucian and the East before we find the faculty. So, too, in Latin before the two late writers named above, Ovid is about the only person who is a real story-teller. Virgil makes very little of his story in verse: and it is shocking to think how Livy throws away his chances in prose. No: putting the Petronian fragments aside, Lucian and Apuleius are the only two novelists in the classical languages before about 400 A.D.: and putting aside their odd coincidence of subject, it has to be remembered that Lucian was a Syrian Greek and Apuleius an African Latin. The conquered world was to conquer not only its conqueror, but its conqueror's teacher, in this youngest accomplishment of literary art.

It was probably in all cases, if not certainly, mixed blood that produced the curious development generally called Greek Romance. It is no part of our business to survey, in any detail, the not very numerous but distinctly interesting compositions which range in point of authorship from Longus and Heliodorus, probably at the meeting of the fourth and fifth centuries, to Eustathius in the twelfth. At one time indeed, when we may return to them a little, we shall find them exercising direct and powerful influence on modern European fiction, and so both directly and indirectly on English: but that is a time a good way removed from the actual beginning of our journey. Still, Apollonius of Tyre, which is probably the oldest piece of English prose fiction that we have, is beyond all doubt derived ultimately from a Greek original of this very class: and the class itself is an immense advance, in the novel direction, upon anything that we have before. It is on the one hand essentially a "romance of adventure," and on the other essentially a "love-story"—in senses to which we find little in

classical literature to correspond in the one case and still less in the other. Instead of being, like Lucius and the Golden Ass, a tissue of stories essentially unconnected and little more than framed by the main tale, it is, though it may have a few episodes, an example of at least romantic unity throughout, with definite hero and definite heroine, the prominence and importance of the latter being specially noteworthy. It is in fact the first division of literature in which the heroine assumes the position of a protagonist. If it falls short in character, so do even later romances to a great extent: if dialogue is not very accomplished, that also was hardly to be thoroughly developed till the novel proper came into being. In the other two great divisions, incident and description, it is abundantly furnished. And, above all, the two great Romantic motives, Adventure and Love, are quite maturely present in it.

To pass to the deluge, and beyond it, and to come to close quarters with our proper division, the origin of Romance itself is a very debatable subject, or rather it is a subject which the wiser mind will hardly care to debate much. The opinion of the present writer—the result, at least, of many years' reading and thought—is that it is a result of the marriage of the older East and the newer (non-classical) West through the agency of the spread of Christianity and the growth and diffusion of the "Saint's Life." The beginnings of Hagiology itself are very uncertain: but what is certain is that they are very early: and that as the amalgamation or leavening of the Roman world with barbarian material proceeded, the spread of Christianity proceeded likewise. The Vision of St. Paul—one of the earliest examples and the starter it would seem, if not of the whole class of sacred Romances, at any rate of the large subsection devoted to Things after Death—has been put as early as "before 400 A.D." It would probably be difficult to date such legends as

those of St. Margaret and St. Catherine too early, having regard to their intrinsic indications: and the vast cycle of Our Lady, though probably later, must have begun long before the modern languages were ready for it, while that of the Cross should be earlier still. And let it be remembered that these Saints' Lives, which are still infinitely good reading, are not in the least confined to homiletic necessities. The jejuneness and woodenness from which the modern religious story too often suffers are in no way chargeable upon all, or even many, of them. They have the widest range of incident—natural as well as supernatural: their touches of nature are indeed extended far beyond mere incident. Purely comic episodes are by no means wanting: and these, like the parallel passages in the dramatising of these very legends, were sure to lead to isolation of them, and to a secular continuation.

But, once more, we must contract the sweep, and quicken the pace to deal not with possible origins, but with actual results—not with Ancient or Transition literature, but with the literature of English in the department first of fiction generally and then, with a third and last narrowing, to the main subject of English fiction in prose.

The very small surviving amount, and the almost completely second-hand character, of Anglo-Saxon literature have combined to frustrate what might have been expected from another characteristic of it—the unusual equality of its verse and prose departments. We have only one—not quite entire but substantive—prose tale in Anglo-Saxon, the version of the famous story of Apollonius of Tyre, which was to be afterwards declined by Chaucer, but attempted by his friend and contemporary Gower, and to be enshrined in the most certain of the Shakespearean "doubtfuls," Pericles. It most honestly gives itself out as a translation (no doubt from the Latin though there was an early Greek original) and it deals briefly with the subject. But as an example of narrative style it is very far indeed from being contemptible: and in passages such as Apollonius' escape from shipwreck, and his wooing of the daughter of Arcestrates, there is something which is different from style, and with which style is not always found in company—that faculty

of telling a story which has been already referred to. Nor does this fail in the narrative portions of the prose Saints' Lives and Homilies, especially Aelfric's, which we possess; in fact it is in these last distinctly remarkable—as where Aelfric tells the tale of the monk who spied on St. Cuthbert's seaside devotions. The same faculty is observable in Latin work, not least in Bede's still more famous telling of the Caedmon story, and of the vision of the other world.

But these faculties have better chance of exhibiting themselves in the verse division of our Anglo-Saxon wreckage. Beowulf itself consists of one first-rate story and one second-rate but not despicable tale, hitched together more or less anyhow. The second, with good points, is, for us, negligible: the first is a "yarn" of the primest character. One may look back to the Odyssey itself without finding anything so good, except the adventures of the Golden Ass which had all the story-work of two mightiest literatures behind them. As literature on the other hand, Beowulf may be overpraised: it has been so frequently. But let anybody with the slightest faculty of "conveyance" tell the first part of the story to a tolerably receptive audience, and he will not doubt (unless he is fool enough to set the effect down to his own gifts and graces) about its excellence as such. There is character—not much, but enough to make it more than a mere story of adventure—and adventure enough for anything; there is by no means ineffectual speech—even

dialogue—of a kind: and there is some effective and picturesque description. The same faculties reappear in such mere fragments as that of Waldhere and the "Finnsburgh" fight: but they are shown much more fully in the Saints' Lives—best of all in the Andreas, no doubt, but remarkably also (especially considering the slender amount of "happenings") in the Guthlac and the Juliana. In fact the very fragments of Anglo-Saxon poetry, by a sort of approximation which they show to dramatic narrative and which with a few exceptions is far less present in the classics, foretell much more clearly and certainly than in the case of some other foretellings which have been detected in them, the future achievements of English literature in the department of fiction. The Ruin (the finest thing perhaps in all Anglo-Saxon) is a sort of background study for something that might have been much better than The Last Days of Pompeii: and The Complaint of Deor, in its allusion to the adventures of the smith Weland and others, makes one sorry that some one more like the historian of a later and decadent

though agreeable Wayland the Smith, had not told us the tale that is now left untold. A crowd of fantastic imaginings or additions, to supply the main substance, and a certain common-sense grasp of actual conditions and circumstances to set them upon, and contrast them with—these are the great requirements of Fiction in life and character. You must mix prose and poetry to get a good romance or even novel. The consciences of the ancients revolted from this mixture of kinds; but there was no such revolt in the earlier moderns, and least of all in our own mediæval forefathers.

So few people are really acquainted with the whole range of Romance (even in English), or with any large part of it, that one may without undue presumption set down in part, if not in whole, to ignorance, a doctrine and position which we must now attack. This is that romance and novel are widely separated from each other; and that the historian of the novel is really straying out of his ground if he meddles with Romance. These are they who would make our proper subject begin with Marivaux and Richardson, or at earliest with Madame de La Fayette, who exclude Bunyan altogether, and sometimes go so far as to question the right of entry to Defoe. But the counter-arguments are numerous: and any one of them would almost suffice by itself. In the first place the idea of the novel arising so late is unnatural and unhistorical: these Melchisedecs without father or mother are not known in literature. In the second a pedantic insistence on the exclusive definition of the novel involves one practical inconvenience which no one, even among

those who believe in it, has yet dared to face. You must carry your wall of partition along the road as well as across it: and write separate histories of Novel and Romance for the last two centuries. The present writer can only say that, though he has dared some tough adventures in literary history, he would altogether decline this. Without the help of the ants that succoured Psyche against Venus that heap would indeed be ill to sort.

But there is a third argument, less practical in appearance but bolder and deeper, which is really decisive of the matter, though few seem to have seen it or at least taken it up. The separation of romance and novel—of the story of incident and the story of character and motive—is a mistake logically and psychologically. It is a very old mistake, and it has deceived some of the elect: but a mistake it is. It made even Dr. Johnson think Fielding shallower than Richardson; and it has made people very different from Dr. Johnson think that Count Tolstoi is a greater analyst and master of a more developed humanity than Fielding. As a matter of fact, when you have excogitated two or more human beings out of your own head and have set them to work in the narrative (not the dramatic) way, you have made the novel in posse, if not in esse, from its apparently simplest development, such as Daphnis and Chloe, to its apparently most complex, such as the Kreutzer Sonata or the triumphs of Mr. Meredith. You have started the "Imitation"—the "fiction"—and tout est là. The ancients could do

this in the dramatic way admirably, though on few patterns; in the poetical way as admirably, but again not on many. The Middle Ages lost the dramatic way almost entirely, but they actually improved the poetical on its narrative side, and the result was Romance. In every romance there is the germ of a novel and more; there is at least the suggestion and possibility of romance in every novel that deserves the name. In the Tristram story and the Lancelot cycle there are most of the things that the romancer of incident and the novelist of character and motive can want or can use, till the end of the world; and Malory (that "mere compiler" as some pleasantly call him) has put the possibilities of the latter and greater creation so that no one who has eyes can miss them. Nor in the beginning does it much or at all matter whether the vehicle was prose or verse. In fact they mostly wrote in verse because prose was not ready.

In the minor romances and tales (taking English versions only) from Havelok to Beryn there is a whole universe of situation, scenario, opportunity for "business." That they have the dress and the scene-backing of one particular period can matter to no one who has eyes for anything beyond dress and scene-backing. And when we are told that they are apt to run too much into grooves and families, it is sufficient to answer that it really does not lie in the mouth of an age which produces grime-novels, problem-novels, and so forth, as if they had been struck off on a hectograph, possessing the not very exalted gift of varying names and places—to reproach any other age on this score. But we have only limited room here for generalities and still less for controversy; let us turn to our proper work and survey the actual turn-out in fiction—mostly as a result of mere fashion, verse, but partly prose—which the Middle Ages has left us as a contribution to this department of English literature.

It has been said that few people know the treasures of English romance, yet there is little excuse for ignorance of them. It is some century since Ellis's extremely amusing, if sometimes rather prosaic, book put much of the matter before those who will not read originals; to be followed in the same path by Dunlop later, and much later still by the invaluable and delightful Catalogue of [British Museum] Romances by Mr. Ward. It is nearly as long since the collections of Ritson and Weber, soon supplemented by others, and enlarged for the last forty years by the publications of the Early English Text Society, put these originals themselves within the reach of everybody who is not so lazy or so timid as to be disgusted or daunted by a very few actually obsolete words and a rather large proportion of obsolete spellings, which will yield to even the minimum of intelligent attention. Only a very small number (not perhaps including a single one of importance) remain unprinted, though no doubt a few are out of print or difficult to

obtain. The quality and variety of the stories told in them are both very considerable, even without making allowance for what has been called the stock character of mediæval composition. That almost all are directly imitated from the French is probable enough, that most are is certain: but this matters, for our purpose, nothing at all. That the imitation was not haphazard or indiscriminate is obvious. Thus, though we have some, we have not very many representatives of the class which was the most numerous of all in France—the chansons de geste or stories of French legendary history, national or family. Except as far as the Saracens are concerned, they would naturally have less interest for English hearers. The Matière de Rome, again—the legends of antiquity—though represented, is not very abundant outside of the universally popular Tale of Troy; and the almost equally popular Alexander legend does not occupy a very large part of them. What is perhaps more remarkable is that until Malory exercised his genius upon