lately been published an unusually large number of books dealing with poetic theory. Some of these have been merely superficial journalism, written without any wide knowledge of English poetry or its history, and showing no natural gifts of judgment. But there have been others which are real contributions to the subject, and are likely in time to take a place with the considerable poetry itself of this age. But I cannot believe that any honest reader can pretend that these books are easy to understand. I have lately read, for example, Mr. Lascelles Abercrombie's essay "Towards a Theory of Art,' Mr. Robert Graves's 'On English Poetry,' and Dr. Strachan's "The Soul of Modern Poetry.' All three are responsible, well-informed, and acute work. It is doubtful whether the metaphysical nature of poetry has ever been so subtly stated before in English as it is by Mr. Abercrombie; a patient reading of the work will, I think, convince any competent reader that this is not an extravagant claim. Mr. Graves's essay is a much more mercurial affair, just, as it were, the personal notes made by a poet in-between-whiles upon the pro

cesses of his own art. It is more often than not unconvincing, but it was not written to convince, and it remains a very charming record full of independence and personality. Just as Mr. Abercrombie's work is important as a contribution towards the metaphysics of poetry, so Dr. Strachan's book has

real originality as a study of the moral philosophy of poetry. And yet though I know all these books to be admirable, there is a great deal in each of them that I cannot understand at all. Dr. Strachan is relatively plain sailing, although even he often persuades us in his gentle way to take things on faith instead of showing any good reason why we should do so, but Mr. Graves says things at times that seem to qualify for a place in Mr. Lear's 'Book of Nonsense,' where the joke always is not that something silly is being said, but that something is being said that isn't anything, while Mr. Abercrombie, although he never comes within a thousand miles of saying anything that is nonsense, says a great many things that leave me painfully aware that I am but a very poor simpleton. But the fun of the whole thing is, if I may by way of illustration be personal for a moment, that I, who never say anything that is not as plain as the turnpike, am told by Mr. Abercrombie that I have precisely the same effect upon him. Nobody I know will accuse me of presuming to place myself in Mr. Abercrombie's company either as a poet or critic, but we are old friends and that is how we affect one another in these matters. The whole truth of the thing is that critical theory is nearly always less intelligible than the art with which it is dealing, because, while the essential condition of art is that completeness of form which makes the thing

created easy to perceive, critical theory is always incomplete, full of loose ends, and largely dependent upon the definition of terms about which no two people in the world are in agreement. In the course of time any particular poetic theory of great distinction is sanctioned by a common consent as to what it means. We may disagree with Wordsworth's view, for example, but we to-day do know what that view really was, or at least we all of us agree that we know this. But in Wordsworth's own time even informed opinion was not sure what he meant. It is not even vital that the interpretation which we make to-day of Wordsworth's theory may conceivably be one from which Wordsworth himself would dissent a fact which we are apt to overlook. The important thing is that we have made out of his statement a very significant piece of poetic theory as to the nature of which, though not necessarily as to the importance of which, we are now of one mind. It may be suggested that the same thing happens about poetry itself, but I do not think this is so. It is perfectly true that any great poem, 'Paradise Lost' let us say, has a more obvious significance to us now than it could have had to its readers when it was published. That is to say, we approach it now with all the assurance bred of two hundred and fifty years of habit, and our minds, because of our ancestry, are able much more readily to perceive the full beauty of the

poem. But this perception of beauty is a different thing from the understanding of a meaning, and while it is easier for the new reader of 'Paradise Lost' to appreciate its grandeur as poetry than it was for the original reader, it is no easier for him to understand its meaning, and he has to apply himself with as much individual intelligence to that task as was necessary in the beginning. But with poetic theory the case is different, it might almost be said that it is reversed. An instructed reader, coming across Matthew Arnold's 'Preface' in 1853, must have known at once that here was somebody speaking with the voice of authority, but he might very well have been excused for not exactly understanding what the voice was saying. We to-day can only confirm the first impression as to the authority, but we can very much more clearly catch the purport and implication of what was being said. In short, it is the profounder quality in any work that most profits by the revealing processes of time. In 'Paradise Lost' the doctrine is of less fundamental importance than the poetry, and it is the poetry that has grown in dominion with the passing years; in the famous 'Preface' the art was little and the doctrine much, and it is the doctrine which has gained in definition.

Much, therefore, that we may perhaps now find difficult in such a book as Mr. Abercrombie's will no doubt be very easy going to readers two genera

tions hence. But he raises one question with his conclusions about which I find myself in positive disagreement at once. It so happens that Dr. Strachan, by an independent process, confirms Mr. Abercrombie's opinion, and since the matter is one which must very profoundly affect any theory of art, and particularly from the artists' own point of view, I should at least like to try to state the other side of the question. My admiration for the two books in question I have already recorded, and anything I may have to say is said with a due sense of obligation. Mr. Abercrombie, then, in his essay "Towards a Theory of Art,' writes:

Several theorists having assumed, as they must, that art is expression, go on to point out that expression is not communication, and conclude from that that communication is a mere accident in art, as though the artist in his work were just talking to himself, and we happen along and overhear what he is saying. This is mere confusion.... What happens when an artist makes a work of art? He makes his experience communicable: and in order to make it exactly and perfectly so he will spend the whole force of his spirit.... If æsthetic experience is the condition of art's activity, the essence of its activity is communication.

This passage is, I am aware, in a very elaborate context, but I do not think that Mr. Abercrombie would consider it unfair to set it apart. The argument seems to me to be fallacious in this respect, that while we may agree that art is expression, I