Page:R L Stevenson 1917 Familiar studies of men and books.djvu/40

 that art is working far ahead of language as well as of science, realising for us, by all manner of suggestions and exaggerations, effects for which as yet we have no direct name; nay, for which we may never perhaps have a direct name, for the reason that these effects do not enter very largely into the necessities of life. Hence alone is that suspicion of vagueness that often hangs about the purpose of a romance: it is clear enough to us in thought; but we are not used to consider anything clear until we are able to formulate it in words, and analytical language has not been sufficiently shaped to that end. We all know this difficulty in the case of a picture, simple and strong as may be the impression that it has left with us; and it is only because language is the medium of romance, that we are prevented from seeing that the two cases are the same. It is not that there is anything blurred or indefinite in the impression left with us, it is just because the impression is so very definite after its own kind, that we find it hard to fit it exactly with the expressions of our philosophical speech.

It is this idea which underlies and issues from a romance, this something which it is the function of that form of art to create, this epical value, that I propose chiefly to seek and, as far as may be, to throw into relief, in the present study. It is thus,