principles of criticism, and, only in a secondary degree, with the application of those principles to the examination of particular works. At the same time, in the analyses and extracts which I have placed before the reader, I have endeavoured to indicate the relationship of one method to another, and to some extent to harmonize the different points of view from which master minds have regarded literature. Indeed, one conspicuous result of the comparison afforded by these pages will, I trust, be to establish the fact that validity of judgment is not to be assigned to any single test; but that a work of literature must often be approached from more than one side, and that a true account of its merits or deficiencies can only be given by applying several tests, and these tests in degrees which vary with the character of the given work.

In the case of one book it may be best to apply the test of truth, and ask with Plato, 'Is the body of information which this book conveys consistent with the facts of life?' In the case of another it may be best to apply the test of symmetry, and ask with Aristotle, 'Is this work constructed in the best possible manner, having regard both to the form of literature to which it belongs, and the common purpose of art, which is to give pleasure?' And, moreover, if we adopt the former of these two central principles, we shall find that we must go further than Plato takes us; for we must distinguish between the truth of art and the truth of logic. In

the case of a work of poetry and of prose-fiction we require a more subtle test. We must ask, 'Does it convey not merely truth as being consistent with the facts of life, but truth as consistent with the mental rendering of those facts-the general conceptions that are based not upon the here and now of every-day experience, but on the generalized experience of more than one country and of more than one age?' ?' That is the truth of art. And because such truth can only be attained by minds that unite a wide range of experience with a resolute determination to know the best, Matthew Arnold finds the 'high seriousness of absolute sincerity,' to be the test of the highest order of poetic merit. When the poet and the artist embody this wider purview in their works-idealize, in a word—poetry, by virtue of the contrast between this ideal rendering of the facts of life, and these facts as they are presented to us in every-day experience, becomes a criticism' of life, and art a criticism' of nature.

6

Similarly, in adopting the artistic standpoint and in applying the test of symmetry, we have to notice that the symmetry required of each art, and of each form of the several arts, is different. 'Symmetry,' from this closer standpoint, becomes composition'; for constructive excellence is resolved into an adaptation of the special means of the special art to the particular purpose which it seeks to fulfil. And in order to judge of this excellence-excellence of technique-we must have a knowledge of these

means.

Here the analyses of the processes of literary and artistic production given by Lessing and Victor Cousin, and their inquiries into the limits of the several arts, provide us with at least the 'bare necessaries' for judgments based upon such artistic considerations.

Let me give an example of a critical difficulty which arises solely from the neglect to discriminate between the different purposes of different forms of literature. Popular judgment is in favour of a 'happy ending' in a novel, but it is commonly maintained that such a termination of the plot is at variance with the canons of criticism, because Aristotle (and other critics after him) has declared the plot ending in 'disaster' to be the most effective and the most artistic. As a matter of fact, the plot terminating in a disaster is required for the one form of literature, the tragedy, where intense pathos, the appeal to the emotions of fear and pity, is the central effect sought to be produced. But in the case of novels in general the principle of 'poetic justice' applies; since only a small class depend entirely, or mainly, upon pathos for their effect. In other words, the 'disaster' is essential to the tragedy, but only incidental to the novel. In prose-fiction, as the vehicle which gives the widest and most varied picture of life, the natural tendency of the human mind towards optimism coincides with the idealizing process of art, and we are justified on both philosophic and critical grounds in expecting

to find in prose-fiction 'a more exact goodness and a more absolute variety than can be found in the nature of things.'1

1

Once more to take a third standpoint. The enlarged psychological knowledge which modern science has put at our disposal, and the expanded range of the imagination due to the quickening of the pulse of life by modern inventions, and by the increased and increasing facilities of inter-communication, together form a notable feature in the progress of man. In view of this, Addison's proposal to measure literature by the degree in which it possesses the power to affect the imagination, acquires a new significance. For here we have a test which is applicable to every species of literature, and one which measures a work of literature by reference to what experience now shows to be a merit of increasing importance. It is only by the application of this test that the experience of the booksellers can be harmonized with our belief in the validity of critical principles. By the application of this test we can discern merit in works of literature, which, being entirely defective on the side of construction, yield no reply to any test of formal or artistic canons. Defective in all else, they have yet this one virtue of stimulating thought by appealing to the imagination of the reader.

But neither this, nor any other single test, is sufficient in itself to measure the merit of a work of

1 The point is fully discussed in chapter iv. p. 80.

literature. In order to obtain a valid judgment, a work of literature must be approached from more than one side, and our verdict must be based upon a balance of the results so obtained. This last test, however, serves one purpose which is worthy of notice. It enables us to discriminate between science and literature. Science is thought embodied in writing; literature is thought first moulded into form by the idealizing process of the human mind, and then, when so moulded, expressed in writing. Where the idealizing process has been employed by the author, there, in whatever branch of literature it may be, will be the appeal to the imagination of the reader. For this power to affect the imagination reveals the presence of the 'something more' added by the writer-the presence, that is, of the personal element which raises history or biography to the rank of literature, and lends a new value to the work of the philosopher or the man of science.

II. THE RELATION OF ESTHETICS TO CRITICISM

Man discriminates a certain aspect in his surroundings, material and moral, which he designates as 'beautiful' and refers to the quality of 'beauty.' In conventional language beauty is said to produce sensations and emotions which differ from those produced by the pleasures of sense, while, on the other hand, they do not coincide with the satisfac