only two lines of narration, the other two lines being refrains. For example, the second line might be

So fair upriseth the rim of the sun,

and the fourth,

So grey is the sea when the day is done.

In such a case the second and the fourth lines of every stanza from beginning to end would be the same, and in printing ballads we usually print the burthen in italics. Here you may be naturally inclined to ask what relation does the burthen bear to the meaning of ballad. In the old ballads it seldom bears any relation at all to the subjecthas nothing to do with it. Then you may ask why not. There is only one explanation that I can give you, and it is this new ballads were generally composed to be sung to the tune of older ballads, and although the main part of the older ballad in such a case would be forgotten as it ceased to be popular, the old refrain would be preserved by the liking of the people for it-having been accustomed to sing it in a great chorus, they would persist in singing it even with the new ballad. Then, again, popular songwriters, having observed this fact, would presently begin to compose new songs with old refrains, knowing that the old refrains would "catch the people." If you bear these possibilities in mind, you will easily perceive that there is nothing extraordinary about the fact of a refrain having nothing to do with its ballad. In my opinion most of our ballad burthens represent the only extant portions of hundreds of older compositions that have been forever lost. And my theory is supported by the existence of a number of different ballads, undoubtedly written at different times and all having the same refrain, or a part of it. Yet in some cases we find that the refrain, as in modern ballads, is made to bear a relation to the story. Such is the case

mind. But the Frenchman is an infinitely finer artist. He also gives you a description of Tokyo seen as a wilderness of roofs; but he first chooses a beautiful place from which to look and a beautiful time of the day in which to see it. Let me translate a few sentences for you:

Uyeno, a very large park, wide avenues, well gravelled,bordered with magnificent old trees, and tufts of bamboos.

I halt upon an elevation at a point overlooking the Lotos-lake, which reflects the evening, like a slightly tarnished mirror, all the gold of sunset. Yedo is beyond those still waters; Yedo is over there, half-lost in the reddish mist of the Autumn evening; a myriad of infinite little greyish roofs all alike,—the furthest, almost indistinguishable in the vague horizon, giving nevertheless an impression that that is not all, that there are more of them, much more, in distances beyond the view. You can distinguish, amidst the uniformity of the low small houses, certain larger buildings with the angles of their roofs turned up. These are the temples. If it were not for them, you might imagine that you were looking at almost any great city quite as well as you could imagine that you were looking at Yedo. Indeed, it requires the effects of distance and of a particular light to make Yedo appear charming; at this moment, for example, I must confess that it is exquisite

to see.

It is dimly outlined in the faintest colours; it has the look of not really existing, of being only a mirage. Then it seems as if long bands of pink cotton were slowly unrolling over the world,drawing this chimerical city in their soft undulations. Now one can no longer distinguish the interval between the lake and the further high land upon which all those myriads of far-away shapes are built. One even doubts whether that really is a lake, or only a very smooth level, reflecting the diffused light of the sky, or simply a stretch of vapour; nevertheless, some few long rosy gleams, still showing upon the surface, almost suffice to assure you that it is really water, and that Lotos-beds here and there make black patches against the reflecting surface.

Although this rapid translation does not give you the colour and charm of the original French, you must be able to see even through it how very accurate and fine the description is an effect of evening sunlight and rosy mist.

I think that most of you have enjoyed the same view, and have noticed how black the lotos leaves really do seem, when the surface of the water is turned to gold by sunset. And then the description of the coming of the mists like long cloud bands of pink cotton is surely as beautiful as it is true. That is the way that a Japanese painter would paint "a picture of Tokyo as seen from the same place at the same time. The Englishman would not have noticed all those delicate and dreamy colours, or if he did, would not trouble himself to try to paint them. Really it is a most difficult thing to do.

Now after this little digression let me come back to the subject of variety in style. Loti knows the art of it; so does many another French writer; but very few Englishmen do. What I am going to say is this, that an author ought to be able to choose a different style for different kinds of work,—that is, a great author. But it is so much trouble to master even one style perfectly well, that very few authors attempt this. However, I think it can be laid down as a true axiom that the style ought to vary with the subject in certain cases; and I think that the great writers of the future will so vary it. The poetical prose, of which I gave you an example from Loti, is admirably suited for particular kinds of composition-short and dreamy things. It is very exhausting to write much in such a style; it is quite as much labour as to write the same thing in verse. But a whole book upon one subject could not be written in this way. The simple naked style, on the other hand, is particularly adapted to story telling, to narrative, even to certain forms of history. The rhetorical style, ornamental without being exactly poetical, has also a special value; it is in such a style that logical argument and philosophical work in the form of essays can perhaps be most effectively presented. I think that some day this will be generally done. But once it becomes a fashion to do it, there will be danger ahead,—the danger of the custom hardening into

conventionalism. Conventionalism kills style. The best way, I think, to meet the difficulty suggested will be to persuade oneself that sentiment, artistic feeling, absolute sincerity of the emotion and of the thought will guide the writer better than any rules as to what style ought to be used. If you try to imitate a model, you will probably go wrong. All literary imitation means weakness. But if you simply follow your own feeling and tastes, trying to be true to them, and to develop them as much as you can— then I think that your style will form itself and will naturally, without direction, take at last the particular form and tone best adapted to the subject.

CHAPTER V

THE VALUE OF THE SUPERNATURAL IN FICTION

Young men of your age are nor inclined to consider the

THE subject of this lecture is much more serious than may appear to you from this title. not likely to believe in ghosts, subject as worthy of attention. The first things necessary to understand are the philosophical and literary relations of the topic. Let me tell you that it would be a mistake to suppose that the stories of the supernatural have had their day in fine literature. On the contrary, wherever fine literature is being produced, either in poetry or in prose, you will find the supernatural element very much alive. Scientific knowledge has not at all diminished the pleasure of mankind in this field of imagination, though it may have considerably changed the methods of treatment. The success of writers of to-day like Maeterlinck is chiefly explained by their skill in the treatment of the ghostly, and of subjects related to supernatural fear. But without citing other living writers, let me observe that there is scarcely any really great author in European literature, old or new, who has not distinguished himself in the treatment of the supernatural. In English literature, I believe there is no exception-even from the time of the Anglo-Saxon poets to Shakespeare, and from Shakespeare to our own day. And this introduces us to the consideration of a general and remarkable fact, a fact that I do not remember to have seen in any books, but which is of very great philosophical importance; there is something ghostly in all great art, whether of literature, music, sculpture, or architecture.

But now let me speak to you about this word "ghostly"; it is a much bigger word, perhaps, than some of you imagine. The old English had no other word for "spiritual" or