greater clearness, grace, and vigour might easily have been attained by transferring a word or phrase from this line to that, or even altering the sequence of a whole group of lines. In no case have we yielded to such temptation, feeling that, our rule once relaxed, we should insensibly but inevitably lapse into mere paraphrase. Temptation beset us with especial force in the less vital passages of the poem. In these places it would have been easy to give our rendering some approach to grace and point by disregarding inversions and other defects of expression, justified in the original by the wit and spirit of the rhymes, but of course deprived in our transcript of any such excuse. Here, as elsewhere, we were proof against temptation; it is for our readers to decide whether our constancy was heroic or pedantic.
It would be folly to pretend either that we have reproduced every word of the original, or that we have avoided all necessity for "padding." The chief drawback of our line-for-line principle is that it has debarred us from eking out the deficiency of one line with the superfluity of the next. We trust, however, that few essential ideas, or even words, of the original will be found quite unaccounted for; while with regard to padding, we have tried, where we found it absolutely forced upon us, to use only such mechanical parts of speech as introduced no new idea into the context. We have found by experiment that the fact of writing in measure has frequently enabled us to keep much closer to the original than would have been possible in prose. This is not in reality so strange as it may at first sight appear. A prose translation of verse can avoid paraphrase only at the cost of grotesque inelegance; whereas in rendering metre into metre, we are working under the same laws which
