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 "the French book," the more poetical parts of the "matter of Britain" itself do not seem to have been very much written about in English. The preliminary stuff about Merlin and Vortigern exists in several handlings; the foreign campaigns of Arthur seem always (perhaps from national vanity) to have been popular. The "off"-branches of Tristram and Percivale, and not a few of the still more episodic romances of adventures concerning Gawain, Iwain, and other knights, receive attention. The execrable Lonelich or Lovelich, who preceded Malory a little, had of course predecessors in handling the other parts of the Graal story. But the crown and flower of the whole—the inspiration which connected the Round Table and the Graal and the love of Lancelot and Guinevere—though, so far as the present writer's reading and opinion are of any weight, the recent attempts to deprive the Englishman, Walter Map, of the honour of conceiving it are of no force—seems to have waited till the fifteenth century—that is to say the last part of three hundred years—before Englishmen took it up. Most popular of all perhaps, on the principle that in novels the flock "likes the savour of fresh grass," seem to have been the pure romans d'aventures—quite unconnected or nearly so with each other or with any of the larger cycles. Those adventures of particular heroes have sometimes a sort of Arthurian link, but they really have no more to do with the main Arthurian story than if Arthur were not.

For the present purpose, however, filiation, origin, and such-like things are of much less importance than the actual stories that get themselves told to satisfy that demand which in due time is to produce the supply of the novel. Of these the two oldest, as regards the actual forms in which we have them, are capital examples of the more and less original handling of "common-form" stories or motives. They were not then, be it remembered, quite such common-form as now—the rightful heir kept out of his rights, the usurper of them, the princess gracious or scornful or both by turns, the quest, the adventure, the revolutions and discoveries and fights, the wedding bells and the poetical justice on the villain. Let it be remembered, too, if anybody is scornful of these as vieux jeu, that they have never been really improved upon