But then the sufferings of Ulyffes, which are a confiderable part of that fentence, are omitted:

[Ὃς μάλα πολλὰ πλά[χθη:]

The confideration of thefe difficulties, in a fervile, literal, tranflation, not long fince made two of our famous wits, Sir John Denham, and Mr. Cowley, to contrive another way of turning authors into our tongue, called, by the latter of them, Imitation.As they were friends, I fuppofe they communicated their thoughts on this fubject to each other; and, therefore, their reafons for it are little different.Though the practice of one is much more moderate. I take imitation of an author, in their fenfe, to be an endeavour of a later Poet to write like one, who has written before him, on the same subject: that is, not to tranflate his words, or to be confined to his fenfe, but only to fet him as a pattern, and to write, as he fuppofes that author would have done, had he lived in our age, and in our country. Yet I dare not fay that either of them have carried this libertine way of rendering authors (as Mr. Cowley calls it) fo far as my definition reaches. For in the Pindaric Odes, the customs and ceremonies of ancient Greece are ftill preferved. But I know not what mifchief may arife hereafter from the example of fuch an innovation, when writers of unequal parts to him shall imitate fo bold an undertaking. To add and to diminish what we please, which is the way avowed by him, ought only to be granted to Mr. Cowley, and that too only in his tranflation of Pindar; because he alone was able to make him amends, by giving him

better of his own, whenever he refufed his author's thoughts. Pindar is generally known to be a dark writer, to want connexion, (I mean as to our understanding) to foar out of fight, and leave his reader at a gaze. So wild and ungovernable a Poet cannot be tranflated literally; his genius is too strong to bear a chain, and, Samfon-like, he shakes it off. A genius fo elevated and unconfined as Mr. Cowley's was but neceffary to make Pindar speak English, and that was to be performed by no other way than imitation. But if Virgil, or Ovid, or any regular intelligible authors be thus ufed, it is no longer to be called their work, when neither the thoughts nor words are drawn from the original: but instead of them there is foinething new produced, which is almost the creation of another hand. By this way, it is true, fomewhat that is excellent may be invented, perhaps more excellent than the firft defign; though Virgil muft be ftill excepted, when that perhaps takes place. Yet he who is inquifitive to know an author's thoughts, will be disappointed in his expectation. And it is not always that a man will be contented to have a prefent made him, when he expects the payment of a debt. To state it fairly: imitation of an author is the most advantageous way for a tranflator to fhew himself, but the greatest wrong which can be done to the memory and reputation of the dead. Sir John Denham (who advised more liberty than he took himfelf) gives his reafon for his innovation, in his admirable preface before the tranflation of the fecond Eneid. Poetry is of fo fubtle a fpirit, that, in

pouring out of one language into another, it will all evaporate; and, if a new fpirit be not added in the transfufion, there will remain nothing but a Caput Mortuum." I confefs this argument holds good against a literal tranflation; but who defends it? Imitation and verbal verfion are, in my opinion, the two extremes, which ought to be avoided: and therefore, when I have propofed the mean betwixt them, it will be feen how far his argument will reach.

No man is capable of tranflating Poetry, who, befides a genius to that art, is not a master both of his author's language, and of his own: nor must we underftand the language only of the Poet, but his particular turn of thoughts and expreffion, which are the characters that diftinguish, and as it were individuate him from all other writers. When we are come thus far, it is time to look into ourselves, to conform our genius to his, to give his thought either the fame turn, if our tongue will bear it, or, if not, to vary but the drefs, not to alter or deftroy the fubftance. The like care must be taken of the more outward ornaments, the words. When they appear (which is but feldom) literally graceful, it were an injury to the author that they should be changed: but fince every language is fo full of its own proprieties, that what is beautiful in one, is often barbarous, nay fometimes nonfenfe, in another, it would be unreasonable to limit a tranflator to the narrow compafs of his author's words. It is enough if he choose out fome expreffion which does not vitiate the fenfe. I fuppofe he may stretch his chain to fuch a latitude; but, by innovation of thoughts, methinks, he breaks it. By this

means the spirit of an author may be transfused, and yet not loft and thus it is plain, that the reafon alledged by Sir John Denham has no farther force than to expreffion for thought, if it be tranflated truly, cannot be loft in another language; but the words that convey it to our apprehenfion (which are the image and ornament of that thought) may be fo ill chofen, as to make it appear in an unhandfome dress, and rob it of its native luftre. There is, therefore, a liberty to be allowed for the expreffion; neither is it neceffary that words and lines fhould be confined to the measure of their original. The fenfe of an author, generally speaking, is to be facred and inviolable.If the fancy of Ovid be luxuriant, it is his character to be fo; and if I retrench it, he is no longer Ovid. It will be replied, that he receives advantage by this lopping of his fuperfluous branches; but I rejoin, that a tranflator has no fuch right. When a painter copies from the life, I fuppofe he has no privilege to alter features, and lineaments, under pretence that his picture will look better: perhaps the face, which he has drawn, would be more exact, if the eyes or nofe were altered; but it is his bufinefs to make it refemble the original. In two cafes only there may a feeming difficulty arife; that is, if the thought be notoriously trivial, or difhoneft: but the fame anfwer will ferve for both, that then they ought not to be tranflated: Et quæ

Defperes tractata nitefcere poffe, relinquas.

Thus I have ventured to give my opinion on this fubject against the authority of two great men, but I

hope without offence to either of their memories; for I both loved them living, and reverence them now they are dead. But, if, after what I have urged, it be thought by better judges, that the praife of a tranflation confifts in adding new beauties to the piece, thereby to recompenfe the lofs which it fuftains by change of language, I fhall be willing to be taught better, and to recant. In the mean time, it seems to me, that the true reafon, why we have fo few verfions which are tolerable, is not from the too close purfuing of the author's fenfe, but because there are fo few, who have all the talents, which are requifite for translation, and that there is fo little praise, and fo fmall encouragement, for fo confiderable a part of learning.