As an example of this process I may mention a single sentence which occurs in my little book on Cyprus. It is a sentence belonging to a description of certain morning scenes—of dewy plains, with peasants moving across them, and here and there a smoke wreath arising from burning weeds. The effect of these scenes in some poignant way was primitive, and I was able at once to reproduce it by saying that the peasants were moving like figures out of the Book of Genesis. I felt, however, that this effect was not produced by the groups of peasants only. I felt that somehow—I could not at first tell how—some part in producing it was played by the smoke wreaths also. At last I managed to capture the suggestion, at first subconscious only, which had so far been eluding me. I finished my original description by adding the following words, "The smoke-wreaths were going up like the smoke of the first sacrifice."
It may be objected that prose built up in this elaborate way loses as much as it gains, because it is bound to lose the charm and the convincing force of spontaneity. This may be so in some cases, but it is not so in all. I have found myself that, so far as my own works are concerned, the passages which are easiest to read are precisely those which it has been most laborious to write. And for this, it seems to me, there is a very intelligible reason. Half of the interests and emotions which make up the substance of life are more or less subconscious, and are, for most men, difficult to identify. One of the functions of pure literature is to make the subconscious reveal itself. It is to make men know what they are, in addition to what spontaneously they feel themselves to be, but feel only, without clear comprehension of it. As soon as a writer, at the cost of whatever labor, manages to make these spontaneities, otherwise subconscious, intelligible, the spontaneity of the processes described by him adds itself at last to his description.
A signal example of this fact may be found, not in prose, but in love poems. Most people can fall in love. It takes no trouble to do so, whatever trouble it may bring them. If any human processes are spontaneous, falling in love is one of them. Most lovers feel more than they know until great love poetry explains it to them what they are; but great love poems are great, not because they are composed spontaneously, but because they express spontaneities which are essentially external to themselves. In other words, the achievement of perfection, whether in prose or poetry, is comparable to the task of a piano tuner, who may spend a whole morning in tightening or relaxing the strings, but who knows at once, when he gets them, the minutely precise tones which the laws of music demand.
Whether every reader will agree with me as to these questions or not, they are, at all events, examples of questions purely literary, which are in themselves captivating for large numbers of people, without any reference to ulterior, or what are called practical, objects. To these questions I may add a few others, which have been specially captivating to myself.
One of them is the use of metaphor as an immemorial literary device, especially in the case of poetry. What is the psychology of metaphor? Let us take an instance from Tennyson, who in one of his poems speaks, with very vivid effect, of Mediterranean bays as colored like "the peacock's neck." The color of the bay is at once made present to the reader's mind. But why? A discussion of this question occurs in a dialogue between two of the characters in my novel The Old Order Changes. The poet, urges one of them, might, if describing a peacock, have said with equal effect that the peacock's neck was colored like a Mediterranean bay. How is it that we gain anything by comparing one equally familiar thing to another? The secret of the use of metaphor in the poet's art is, says the speaker, this. When the mind is at rest its surface is alive with vivid images which have settled on it like sea birds on a rock, but the moment any one of these detects an approach on our part, in order that we may examine it carefully, its wings are spread, and in a flash it is gone. When, however, we use a simile in order to describe something which is obviously our main concern (say the color of a Mediterranean bay), the thing which we are anxious to describe acts as a kind of stalking-horse, which enables us to approach and capture the thing which we use as an illustration (say the neck of a peacock) before the peacock so much as suspects our neighborhood. We have it alive before us, with all its feathers glittering, and these throw a new light on objects which our direct touch might have frightened away beyond the confines of our field of vision. The more vivid of the two objects communicates its color to the less vivid.
Two other purely literary questions are discussed in The Veil of the Temple, the first of these being as follows. One of the speakers calls attention to a criticism which is often and justly made with reference to many, and even to the best of novels, that, while the minor characters are drawn with the utmost skill, the heroes (such as most of Scott's) have often no characters at all. The reason, he says, is that, in most cases, the hero is not so much an individual, with characteristics peculiar to himself, as a certain point of view, from which all the other characters and incidents of the story are drawn. Or else, if some of these are, as very often happens, not drawn from the point of view of the hero, they are drawn from the point of view of some other ideal spectator, on whose position, moral or local, the whole perspective of the story, mental or ocular, depends. Let us take, for example, a typical opening scene of a kind proverbially frequent in the novels of G. P. R. James. Such scenes were proverbially described very much as follows: "To the right lay a gray wall, which formed, to all appearance, the boundary of some great sheep tract. To the left was a wood of larches. Between these was a road, showing so few signs of use that it might have been a relic of some almost forgotten world. Proceeding along this road on a late October evening might have been seen three horsemen, of imperfectly distinguishable, yet vaguely sinister, aspect." In the absence of an ideal spectator, who is tacitly identified with the novelist, his hero, or his reader, such a description would mean very little more than nothing. There would be no left or right unless for a supposed spectator standing in a particular place and looking in a particular direction. The aspect of the horsemen could not be sinister or indistinguishable unless there were an assumed man whose eyes were unable to distinguish it.
The argument here in question will carry us on to certain kindred problems, connected likewise with the novelist's art, which are these: The necessary assumption of the author as ideal spectator being given, a question arises with regard to the range of vision which, in his capacity of spectator, the novelist professes to possess. Many novelists mar the effect of their work—and among these Thackeray is notable—by adopting an attitude which in this respect is constantly vacillating. Sometimes it is one of omniscience, sometimes of blind perplexity. At one time he describes the inmost thoughts of his characters which are suffered or pursued in secret, as though he could see through everything. At another time he will startle the reader with some such question as this: "Who shall dare to say—I certainly cannot—what at that solemn moment the lad's real reflections were?" A partial escape from the sense of unreality which alternations like these produce is to be found in the method which many novelists have adopted—namely, that of dividing the story into so many separate parts, these being told in succession by so many different characters, each recording events as wholly seen from the point of his own unchanged perspective. Such is the method adopted by Wilkie Collins in The Woman in White, for example. The danger of this artifice is that it tends to be too apparent. The most logically complete escape from the difficulties which we are here glancing at is to be found, no doubt, in the method of autobiography in a single and undivided form; unless indeed the assumption of absolute omniscience on the author's part can be used with a rigid consistency which it very rarely exhibits.
Another question of a purely literary kind, reflection on which is to me, at least, pleasurable (though many persons of literary taste may, perhaps, regard it as a bore), is the relation of modern prosody to ancient, and more particularly to Latin. It has always seemed to me that the lengthening and shortening of syllables according to their position, as happens in classical Latin, with regard to the syllables that follow them, must always have corresponded with the stresses or absence of stress which would naturally be made apparent by the voice of an ideal reciter; and to me, as to some other people, the question has proved amusing of how far in English verse Latin prosody could be reproduced. Many attempts have been made at deciding this question by experiments. The most remarkable of these are two which were made by Tennyson. One of them, called "Hendecasyllabics," is little more than a trick played with extreme skill, and in no serious sense does it merit the name of poetry. The other, "An Ode to Milton," is no less charming as a poem than as a conquest over technical difficulties. Let us take the first stanza:
Oh, mighty-mouthed inventor of harmonies,
Oh, skilled to sing of time and eternity,
God-gifted organ voice of England,
Milton a name to resound for ages.
Here the stresses which the meaning of the English verse demands fall exclusively on syllables which would, according to Latin prosody, be long; but there are one or two syllables which in Latin verse would be long (such as "of" in the second line) which invite no stress in the English—which do not, indeed, admit of it—and must for that reason be treated by an English reader as short. Aiming at greater completeness, but otherwise in a manner very much less ambitious, I attempted an experiment of a similar kind myself, consisting of a few hexameters, in which not only do the natural stresses fall, and fall exclusively, on syllables which in Latin would be long, but in which also every syllable would be emphasized by an English reciter with a natural stress corresponding to it. These hexameters were a metrical amplification of an advertisement which figures prominently in the carriages of the Tube Railway, proclaiming the charms of a suburb called Sudbury Town, and remarkable for its surrounding pine woods. The moment I read the words "Sudbury Town" I recognized in them the beginning of a hexameter classically pure; and after many abortive attempts I worked out a sequel—a very short one—as follows: