Besides all this, an author has the choice of his own thoughts and words, which a translator has not; he is confined by the sense of the inventor to those expressions which are the nearest to it: so that Virgil, studying brevity, and having the command of his own language, could bring those words into a narrow compass, which a translator cannot render without circumlocutions. In short, they, who have called him the torture of grammarians, might also have called him the plague of translators; for he seems to have studied not to be translated. I own that, endeavouring to turn his "Nisus and Euryalus" as close as I was able, I have performed that episode too literally; that, giving more scope to "Mezentius and Lausus," that version, which has more of the majesty of Virgil, has less of his conciseness; and all that I can promise for myself, is only, that I have done both better than Ogleby, and perhaps as well as Caro; so that, methinks, I come like a malefactor, to make a speech upon the gallows, and to warn all other poets, by my sad example, from the sacrilege of translating Virgil. Yet, by considering him so carefully as I did before my attempt, I have made some faint resemblance of him; and, had I taken more time, might possibly have succeeded better; but never so well as to have satisfied myself.
He who excels all other poets in his own language, were it possible to do him right, must appear above them in our tongue, which, as my Lord Roscommon justly observes, approaches nearest to the Roman in its majesty; nearest indeed, but with a vast interval betwixt them. There is an inimitable grace in Virgil's words, and in them principally consists that beauty, which gives so inexpressible a pleasure to him who best understands their force. This diction of his (I must once again say) is never to be copied; and, since it cannot, he will appear but lame in the best translation. The turns of his verse, his breakings, his propriety, his numbers, and his gravity, I have as far imitated, as the poverty of our language, and the hastiness of my performance, would allow. I may seem sometimes to have varied from his sense; but I think the greatest variations may be fairly deduced from him; and where I leave his commentators, it may be I understand him better: at least I writ without consulting them in many places. But two particular lines in Mezentius and Lausus, I cannot so easily excuse. They are indeed remotely allied to Virgil's sense; but they are too like the trifling tenderness of Ovid, and were printed before I had considered them enough to alter them. The first of them I have forgotten, and cannot easily retrieve, because the copy is at the press. The second is this: