phrase can hardly be too homely for the true Epic style, if it be but energetic and graphic. Those words only are to be rejected as mean which are also weak and petty.
The problem presented to a translator is highly complicated, and he is constantly obliged to compromise. Even after he has chosen all his principles correctly, he is liable perpetually to miss in detail, from the delicacy of applying them: but if he chose his first principles wrongly, all is over: no skill can bring his work right. He may produce a splendid piece of varnish, as Pope has done; or a vigorous poem, as Chapman; but it will not so represent the original as the translation of a great poem ought. And here, the first matter of all, is, to select the metre; with which the style is intimately connected. The moral qualities of Homer's style being like to those of the English ballad, we need a metre of the same genius. It must be fundamentally musical and popular. Only those metres which, by the very possession of these qualities, are liable to degenerate into doggerel, are suitable to reproduce the ancient Epic. To say this, is to say, that our metre must be composed of systems of either four or three beats; for it is of such lines that English ballads or ditties are composed. Indeed, musicians tell us that all simple melodies are formed in eight bars,—even what is called "the subject" in the most complicated pieces of Mozart or Beethoven. I imagine that the "Long Metre" of our Hymn Books,—(the metre of Walter Scott, by far the most Homeric of our poets,)—is in fact founded on this musical principle; while our "Common Metre" is the same, with merely "a rest" at the end. How naturally one generates the other, is seen in Scott's own practice, who intermingles lines of three beats as a sort of close to those of four. The same thing appears in Greek anapæsts, which close with a "parœmiac" verse. Indeed the Homeric line itself is composed of two shorter lines, with
