THE task of a translator is a thankless one at best. Be he never so skilful and accurate, be he never so amply endowed with the divine qualifications of the poet, it is still questionable if he can ever succeed in saying satisfactorily with new words that which has once been inimitably said—said for all time—with the old words. "Psychologically, there is perhaps nothing more complex than an elaborate poem. The sources of its effect upon our minds may be likened to a system of forces which is in the highest degree unstable; and the slightest displacement of phrases, by disturbing the delicate rhythmical equilibrium of the whole, must inevitably awaken a jarring sensation." Matthew Arnold has given us an excellent series of lectures upon translating Homer, in which he doubtless succeeds in showing that some methods of translation are preferable to others, but in which he proves nothing so forcibly as that the simplicity and grace, the rapidity, dignity, and fire, of Homer are quite incommunicable, save by the very words in which they first found expression. And what is thus said of Homer will apply to Dante with perhaps even greater force. With nearly all of Homer's grandeur and rapidity, though not with nearly all his simplicity, the poem of Dante manifests a peculiar intensity of subjective feeling which was foreign to the age of Homer, as indeed to all pre-Christian antiquity. But concerning this we need not dilate, as it has often been duly remarked upon, and notably by Carlyle, in his "Lectures on Hero-Worship." Who that has once heard the wail of unutterable despair sounding in the line

can rest satisfied with the interpretation

yet this rendering is literally exact.

A second obstacle, hardly less formidable, hardly less fatal to a satisfactory translation, is presented by the highly complicated system of triple rhyme upon which Dante's poem is constructed. This, which must ever be a stumbling-block to the translator, seems rarely to interfere with the free and graceful movement of the original work. The mighty thought of the master felt no impediment from the elaborate artistic panoply which must needs obstruct and harass the interpretation of the disciple. Dante's terza rima is a bow of Odysseus which weaker mortals cannot bend with any amount of tugging, and which Mr. Longfellow has judiciously refrained from trying to bend. Yet no one can fail to remark the prodigious loss entailed by this necessary sacrifice of one of the most striking characteristics of the original poem. Let any one who has duly reflected upon the strange and subtle effect produced on him by the peculiar rhyme of Tennyson's "In Memoriam," endeavour to realize the very different effect which would be produced if the verses were to be alternated or coupled in successive pairs, or if rhyme were to be abandoned for blank verse. The exquisite melody of the poem would be silenced. The rhyme-system of the "Divine Comedy" refuses equally to be tampered with or ignored. Its effect upon the ear and the mind is quite as remarkable as that of the rhyme-system of "In Memoriam"; and the impossibility of reproducing it is one good reason why Dante must always suffer even more from translation than most poets. 34

Something, too, must be said of the difficulties inevitably arising from the diverse structure and genius of the Italian and English languages. None will deny that many of them are insurmountable. Take the third line of the first canto,—

which Mr. Longfellow translates

Perhaps there is no better word than "lost" by which to translate smarrita in this place; yet the two words are far from equivalent in force. About the word smarrita there is thrown a wide penumbra of meaning which does not belong to the word lost. 35 By its diffuse connotations the word smarrita calls up in our minds an adequate picture of the bewilderment and perplexity of one who is lost in a trackless forest. The high-road with out, beaten hard by incessant overpassing of men and beasts and wheeled vehicles, gradually becomes metamorphosed into the shady lane, where grass sprouts up rankly between the ruts, where bushes encroach upon the roadside, where fallen trunks now and then intercept the traveller; and this in turn is lost in crooked by-ways, amid brambles and underbrush and tangled vines, growing fantastically athwart the path, shooting up on all sides of the bewildered wanderer, and rendering advance and retreat alike hopeless. No one who in childhood has wandered alone in the woods can help feeling all this suggested by the word smarrita in this passage. How bald in comparison is the word lost, which might equally be applied to a pathway, a reputation, and a pocket-book! 36 The English is no doubt the most copious and variously expressive of all living languages, yet I doubt if it can furnish any word capable by itself of calling up the complex images here suggested by smarrita. 37 And this is but one example, out of many that might be cited, in which the lack of exact parallelism between the two languages employed causes every translation to suffer.

All these, however, are difficulties which lie in the nature of things,—difficulties for which the translator is not responsible; of which he must try to make the best that can be made, but which he can never expect wholly to surmount. We have now to inquire whether there are not other difficulties, avoidable by one method of translation, though not by another; and in criticizing Mr. Longfellow, we have chiefly to ask whether he has chosen the best method of translation,—that which most surely and readily awakens in the reader's mind the ideas and feelings awakened by the original.

The translator of a poem may proceed upon either of two distinct principles. In the first case, he may render the text of his original into English, line for line and word for word, preserving as far as possible its exact verbal sequences, and translating each individual word into an English word as nearly as possible equivalent in its etymological force. In the second case, disregarding mere syntactic and etymologic equivalence, his aim will be to reproduce the inner meaning and power of the original, so far as the constitutional difference of the two languages will permit him.

It is the first of these methods that Mr. Longfellow has followed in his translation of Dante. Fidelity to the text of the original has been his guiding principle; and every one must admit that, in carrying out that principle, he has achieved a degree of success alike delightful and surprising. The method of literal translation is not likely to receive any more splendid illustration. It is indeed put to the test in such a way that the shortcomings now to be noticed bear not upon Mr. Longfellow's own style of work so much as upon the method itself with which they are necessarily implicated. These defects are, first, the too frequent use of syntactic inversion, and secondly, the too manifest preference extended to words of Romanic over words of Saxon origin.

To illustrate the first point, let me give a few examples. In Canto I. we have:—

which is thus rendered by Mr. Cary,—

and by Dr. Parsons,—

Inferno, I. 7-10.

Again in Canto X. we find:—

an inversion which is perhaps not more unidiomatic than Mr. Cary's,—

but which is advantageously avoided by Mr. Wright,—

and is still better rendered by Dr. Parsons,—

And here my eyes, reverting to the end of Canto IX.,

fall upon a similar contrast between Mr. Longfellow's lines,—

and those of Dr. Parsons,—

Does it not seem that in all these cases Mr. Longfellow, and to a slightly less extent Mr. Cary, by their strict adherence to the letter, transgress the ordinary rules of English construction; and that Dr. Parsons, by his comparative freedom of movement, produces better poetry as well as better English? In the last example especially, Mr. Longfellow's inversions are so violent that to a reader ignorant of the original Italian, his sentence might be hardly intelligible. In Italian such inversions are permissible; in English they are not; and Mr. Longfellow, by transplanting them into English, sacrifices the spirit to the letter, and creates an obscurity in the translation where all is lucidity in the original. Does not this show that the theory of absolute literality, in the case of two languages so widely different as English and Italian, is not the true one?

Secondly, Mr. Longfellow's theory of translation leads him in most cases to choose words of Romanic origin in preference to those of Saxon descent, and in many cases to choose an unfamiliar instead of a familiar Romanic word, because the former happens to be etymologically identical with the word in the original. Let me cite as an example the opening of Canto III.:—

Here are three lines which, in their matchless simplicity and grandeur, might well excite despair in the breast of any translator. Let us contrast Mr. Longfellow's version.—

with that of Dr. Parsons,—,

I do not think any one will deny that Dr. Parsons's version, while far more remote than Mr. Longfellow's from the diction of the original, is somewhat nearer its spirit. It remains to seek the explanation of this phenomenon. It remains to be seen why words the exact counterpart of Dante's are unfit to call up in our minds the feelings which Dante's own words call up in the mind of an Italian. And this inquiry leads to some general considerations respecting the relation of English to other European languages.

Every one is aware that French poetry, as compared with German poetry, seems to the English reader very tame and insipid; but the cause of this fact is by no means so apparent as the fact itself. That the poetry of Germany is actually and intrinsically superior to that of France, may readily be admitted; but this is not enough to account for all the circumstances of the case. It does not explain why some of the very passages in Corneille and Racine, which to us appear dull and prosaic, are to the Frenchman's apprehension instinct with poetic fervour. It does not explain the undoubted fact that we, who speak English, are prone to underrate French poetry, while we are equally disposed to render to German poetry even more than its due share of merit. The reason is to be sought in the verbal associations established in our minds by the peculiar composition of the English language. Our vocabulary is chiefly made up on the one hand of indigenous Saxon words, and on the other hand of words derived from Latin or French. It is mostly words of the first class that we learn in childhood, and that are associated with our homeliest and deepest emotions; while words of the second class—usually acquired somewhat later in life and employed in sedate abstract discourse—have an intellectual rather than an emotional function to fulfil. Their original significations, the physical metaphors involved in them, which are perhaps still somewhat apparent to the Frenchman, are to us wholly non-existent. Nothing but the derivative or metaphysical signification remains. No physical image of a man stepping over a boundary is presented to our minds by the word transgress, nor in using the word comprehension do we picture to ourselves any manual act of grasping. It is to this double structure of the English language that it owes its superiority over every other tongue, ancient or modern, for philosophical and scientific purposes. Albeit there are numerous exceptions, it may still be safely said, in a general way, that we possess and habitually use two kinds of language,—one that is physical, for our ordinary purposes, and one that is metaphysical, for purposes of abstract reasoning and discussion. We do not say like the Germans, that we "begripe" (begreifen) an idea, but we say that we "conceive" it. We use a word which once had the very same material meaning as begreifen, but which has in our language utterly lost it. We are accordingly able to carry on philosophical inquiries by means of words which are nearly or quite free from those shadows of original concrete meaning which, in German, too often obscure the acquired abstract signification. Whoever has dealt in English and German metaphysics will not fail to recognize the prodigious superiority of English in force and perspicuity, arising mainly from the causes here stated. But while this homogeneity of structure in German injures it for philosophical purposes, it is the very thing which makes it so excellent as an organ for poetical expression, in the opinion of those who speak English. German being nearly allied to Anglo-Saxon, not only do its simple words strike us with all the force of our own homely Saxon terms, but its compounds also, preserving their physical significations almost unimpaired, call up in our minds concrete images of the greatest definiteness and liveliness. It is thus that German seems to us pre-eminently a poetical language, and it is thus that we are naturally inclined to overrate rather than to depreciate the poetry that is written in it.

With regard to French, the case is just the reverse. The Frenchman has no Saxon words, but he has, on the other hand, an indigenous stock of Latin words, which he learns in early childhood, which give outlet to his most intimate feelings, and which retain to some extent their primitive concrete picturesqueness. They are to him just as good as our Saxon words are to us. Though cold and merely intellectual to us, they are to him warm with emotion; and this is one reason why we cannot do justice to his poetry, or appreciate it as he appreciates it. To make this perfectly clear, let us take two or three lines from Shakespeare:—

which I have somewhere seen thus rendered into French:

Why are we inclined to laugh as we read this? Because it excites in us an undercurrent of consciousness which, if put into words, might run something like this:—

No such effect would be produced upon a Frenchman. The translation would strike him as excellent, which it really is. The last line in particular would seem poetical to us, did we not happen to have in our language words closely akin to dent and penetrante, and familiarly employed in senses that are not poetical.

Applying these considerations to Mr. Longfellow's choice of words in his translation of Dante, we see at once the unsoundness of the principle that Italian words should be rendered by their Romanic equivalents in English. Words that are etymologically identical with those in the original are often, for that very reason, the worst words that could be used. They are harsh and foreign to the English ear, however homelike and musical they may be to the ear of an Italian. Their connotations are unlike in the two languages; and the translation which is made literally exact by using them is at the same time made actually inaccurate, or at least inadequate. Dole and dolent are doubtless the exact counterparts of dolore and dolente, so far as mere etymology can go. But when we consider the effect that is to be produced upon the mind of the reader, wretchedness and despairing are fat better equivalents. The former may compel our intellectual assent, but the latter awaken our emotional sympathy.

Doubtless by long familiarity with the Romanic languages, the scholar becomes to a great degree emancipated from the conditions imposed upon him by the peculiar composition of his native English. The concrete significance of the Romanic words becomes apparent to him, and they acquire energy and vitality. The expression dolent may thus satisfy the student familiar with Italian, because it calls up in his mind, through the medium of its equivalent dolente, the same associations which the latter calls up in the mind of the Italian himself. 41 But this power of appreciating thoroughly the beauties of a foreign tongue is in the last degree an acquired taste,—as much so as the taste for olives and kirschenwasser to the carnal palate. It is only by long and profound study that we can thus temporarily vest ourselves, so to speak, with a French or Italian consciousness in exchange for our English one. The literary epicure may keenly relish such epithets as dolent; but the common English reader, who loves plain fare, can hardly fail to be startled by it. To him it savours of the grotesque; and if there is any one thing especially to be avoided in the interpretation of Dante, it is grotesqueness.

Those who have read over Dante without reading into him, and those who have derived their impressions of his poem from M. Dore's memorable illustrations, will here probably demur. What! Dante not grotesque! That tunnel-shaped structure of the infernal pit; Minos passing sentence on the damned by coiling his tail; Charon beating the lagging shades with his oar; Antaios picking up the poets with his fingers and lowering them in the hollow of his hand into the Ninth Circle; Satan crunching in his monstrous jaws the arch-traitors, Judas, Brutus and Cassius; Ugolino appeasing his famine upon the tough nape of Ruggieri; Bertrand de Born looking (if I may be allowed the expression) at his own dissevered head; the robbers exchanging form with serpents; the whole demoniac troop of Malebolge,—are not all these things grotesque beyond everything else in poetry? To us, nurtured in this scientific nineteenth century, they doubtless seem so; and by Leigh Hunt, who had the eighteenth-century way of appreciating other ages than his own, they were uniformly treated as such. To us they are at first sight grotesque, because they are no longer real to us. We have ceased to believe in such things, and they no longer awaken any feeling akin to terror. But in the thirteenth century, in the minds of Dante and his readers, they were living, terrible realities. That Dante believed literally in all this unearthly world, and described it with such wonderful minuteness because he believed in it, admits of little doubt. As he walked the streets of Verona the people whispered, "See, there is the man who has been in hell!" Truly, he had been in hell, and described it as he had seen it, with the keen eyes of imagination and faith. With all its weird unearthliness, there is hardly another book in the whole range of human literature which is marked with such unswerving veracity as the "Divine Comedy." Nothing is there set down arbitrarily, out of wanton caprice or for the sake of poetic effect, but because to Dante's imagination it had so imposingly shown itself that he could not but describe it as he saw it. In reading his cantos we forget the poet, and have before us only the veracious traveller in strange realms, from whom the shrewdest cross-examination can elicit but one consistent account. To his mind, and to the mediaeval mind generally, this outer kingdom, with its wards of Despair, Expiation, and Beatitude, was as real as the Holy Roman Empire itself. Its extraordinary phenomena were not to be looked on with critical eyes and called grotesque, but were to be seen with eyes of faith, and to be worshipped, loved, or shuddered at. Rightly viewed, therefore, the poem of Dante is not grotesque, but unspeakably awful and solemn; and the statement is justified that all grotesqueness and bizarrerie in its interpretation is to be sedulously avoided.

Therefore, while acknowledging the accuracy with which Mr. Longfellow has kept pace with his original through line after line, following the "footing of its feet," according to the motto quoted on his title-page, I cannot but think that his accuracy would have been of a somewhat higher kind if he had now and then allowed himself a little more liberty of choice between English and Romanic words and idioms.

A few examples will perhaps serve to strengthen as well as to elucidate still further this position.

"Inferno," Canto III., line 22, according to Longfellow:—

According to Cary:—

According to Parsons:—

Canto V., line 84:—

Canto XVII., line 42:—

Canto XVII., line 25:—

Canto V., line 51:—

Line 136:—

"Purgatorio," Canto XV., line 139:—

Mr. Cary's "bright vespertine ray" is only a trifle better; but Mr. Wright's "splendour of the evening ray" is, in its simplicity, far preferable.

Canto XXXI., line 131:—

Here Mr. Longfellow has apparently followed the authority of the Crusca, reading

and translating carribo by saraband, a kind of Moorish dance. The best manuscripts, however, sanction M. Witte's reading:—

If this be correct, carribo cannot signify "a dance," but rather "the song which accompanies the dance"; and the true sense of the passage will have been best rendered by Mr. Cary. 49

Whenever Mr. Longfellow's translation is kept free from oddities of diction and construction, it is very animated and vigorous. Nothing can be finer than his rendering of "Purgatorio," Canto VI., lines 97-117:—

So, too, Canto III., lines 79-84:—

Francesca's exclamation to Dante is thus rendered by Mr. Longfellow:—

This is admirable,—full of the true poetic glow, which would have been utterly quenched if some Romanic equivalent of dolore had been used instead of our good Saxon sorrow. 53 So, too, the "Paradiso," Canto I., line 100:—

And, finally, the beginning of the eighth canto of the "Purgatorio":—

This passage affords an excellent example of what the method of literal translation can do at its best. Except in the second line, where "those who sail the sea" is wisely preferred to any Romanic equivalent of naviganti the version is utterly literal; as literal as the one the school-boy makes, when he opens his Virgil at the Fourth Eclogue, and lumberingly reads, "Sicilian Muses, let us sing things a little greater." But there is nothing clumsy, nothing which smacks of the recitation-room, in these lines of Mr. Longfellow. For easy grace and exquisite beauty it would be difficult to surpass them. They may well bear comparison with the beautiful lines into which Lord Byron has rendered the same thought:—

Setting aside the concluding sentimental generalization,—which is much more Byronic than Dantesque,—one hardly knows which version to call more truly poetical; but for a faithful rendering of the original conception one can hardly hesitate to give the palm to Mr. Longfellow.

Thus we see what may be achieved by the most highly gifted of translators who contents himself with passively reproducing the diction of his original, who constitutes himself, as it were, a conduit through which the meaning of the original may flow. Where the differences inherent in the languages employed do not intervene to alloy the result, the stream of the original may, as in the verses just cited, come out pure and unweakened. Too often, however, such is the subtle chemistry of thought, it will come out diminished in its integrity, or will appear, bereft of its primitive properties as a mere element in some new combination. Our channel is a trifle too alkaline perhaps; and that the transferred material may preserve its pleasant sharpness, we may need to throw in a little extra acid. Too often the mere differences between English and Italian prevent Dante's expressions from coming out in Mr. Longfellow's version so pure and unimpaired as in the instance just cited. But these differences cannot be ignored. They lie deep in the very structure of human speech, and are narrowly implicated with equally profound nuances in the composition of human thought. The causes which make dolente a solemn word to the Italian ear, and dolent a queer word to the English ear, are causes which have been slowly operating ever since the Italian and the Teuton parted company on their way from Central Asia. They have brought about a state of things which no cunning of the translator can essentially alter, but to the emergencies of which he must graciously conform his proceedings. Here, then, is the sole point on which we disagree with Mr. Longfellow, the sole reason we have for thinking that he has not attained the fullest possible measure of success. Not that he has made a "realistic" translation,—so far we conceive him to be entirely right; but that, by dint of pushing sheer literalism beyond its proper limits, he has too often failed to be truly realistic. Let us here explain what is meant by realistic translation.

Every thoroughly conceived and adequately executed translation of an ancient author must be founded upon some conscious theory or some unconscious instinct of literary criticism. As is the critical spirit of an age, so among other things will be its translations. Now the critical spirit of every age previous to our own has been characterized by its inability to appreciate sympathetically the spirit of past and bygone times. In the seventeenth century criticism made idols of its ancient models; it acknowledged no serious imperfections in them; it set them up as exemplars for the present and all future times to copy. Let the genial Epicurean henceforth write like Horace, let the epic narrator imitate the supreme elegance of Virgil,—that was the conspicuous idea, the conspicuous error, of seventeenth-century criticism. It overlooked the differences between one age and another. Conversely, when it brought Roman patricians and Greek oligarchs on to the stage, it made them behave like French courtiers or Castilian grandees or English peers. When it had to deal with ancient heroes, it clothed them in the garb and imputed to them the sentiments of knights-errant. Then came the revolutionary criticism of the eighteenth century, which assumed that everything old was wrong, while everything new was right. It recognized crudely the differences between one age and another, but it had a way of looking down upon all ages except the present. This intolerance shown toward the past was indeed a measure of the crudeness with which it was comprehended. Because Mohammed, if he had done what he did, in France and in the eighteenth century, would have been called an impostor, Voltaire, the great mouthpiece and representative of this style of criticism, portrays him as an impostor. Recognition of the fact that different ages are different, together with inability to perceive that they ought to be different, that their differences lie in the nature of progress,—this was the prominent characteristic of eighteenth-century criticism. Of all the great men of that century, Lessing was perhaps the only one who outgrew this narrow critical habit.

Now nineteenth-century criticism not only knows that in no preceding age have men thought and behaved as they now think and behave, but it also understands that old-fashioned thinking and behaviour was in its way just as natural and sensible as that which is now new-fashioned. It does not flippantly sneer at an ancient custom because we no longer cherish it; but with an enlightened regard for everything human, it inquires into its origin, traces its effects, and endeavours to explain its decay. It is slow to characterize Mohammed as an impostor, because it has come to feel that Arabia in the seventh century is one thing and Europe in the nineteenth another. It is scrupulous about branding Caesar as an usurper, because it has discovered that what Mr. Mill calls republican liberty and what Cicero called republican liberty are widely different notions. It does not tell us to bow down before Lucretius and Virgil as unapproachable models, while lamenting our own hopeless inferiority; nor does it tell us to set them down as half-skilled apprentices, while congratulating ourselves on our own comfortable superiority; but it tells us to study them as the exponents of an age forever gone, from which we have still many lessons to learn, though we no longer think as it thought or feel as it felt. The eighteenth century, as represented by the characteristic passage from Voltaire, cited by Mr. Longfellow, failed utterly to understand Dante. To the minds of Voltaire and his contemporaries the great mediaeval poet was little else than a Titanic monstrosity,—a maniac, whose ravings found rhythmical expression; his poem a grotesque medley, wherein a few beautiful verses were buried under the weight of whole cantos of nonsensical scholastic quibbling. This view, somewhat softened, we find also in Leigh Hunt, whose whole account of Dante is an excellent specimen of this sort of criticism. Mr. Hunt's fine moral nature was shocked and horrified by the terrible punishments described in the "Inferno." He did not duly consider that in Dante's time these fearful things were an indispensable part of every man's theory of the world; and, blinded by his kindly prejudices, he does not seem to have perceived that Dante, in accepting eternal torments as part and parcel of the system of nature, was nevertheless, in describing them, inspired with that ineffable tenderness of pity which, in the episodes of Francesca and of Brunetto Latini, has melted the hearts of men in past times, and will continue to do so in times to come. "Infinite pity, yet infinite rigour of law! It is so Nature is made: it is so Dante discerned that she was made." 57 This remark of the great seer of our time is what the eighteenth century could in no wise comprehend. The men of that day failed to appreciate Dante, just as they were oppressed or disgusted at the sight of Gothic architecture; just as they pronounced the scholastic philosophy an unmeaning jargon; just as they considered mediaeval Christianity a gigantic system of charlatanry, and were wont unreservedly to characterize the Papacy as a blighting despotism. In our time cultivated men think differently. We have learned that the interminable hair-splitting of Aquinas and Abelard has added precision to modern thinking. 58 We do not curse Gregory VII. and Innocent III. as enemies of the human race, but revere them as benefactors. We can spare a morsel of hearty admiration for Becket, however strongly we may sympathize with the stalwart king who did penance for his foul murder; and we can appreciate Dante's poor opinion of Philip the Fair no less than his denunciation of Boniface VIII. The contemplation of Gothic architecture, as we stand entranced in the sublime cathedrals of York or Rouen, awakens in our breasts a genuine response to the mighty aspirations which thus became incarnate in enduring stone. And the poem of Dante—which has been well likened to a great cathedral—we reverently accept, with all its quaint carvings and hieroglyphic symbols, as the authentic utterance of feelings which still exist, though they no longer choose the same form of expression.

A century ago, therefore, a translation of Dante such as Mr. Longfellow's would have been impossible. The criticism of that time was in no mood for realistic reproductions of the antique. It either superciliously neglected the antique, or else dressed it up to suit its own notions of propriety. It was not like a seven-league boot which could fit everybody, but it was like a Procrustes-bed which everybody must be made to fit. Its great exponent was not a Sainte-Beuve, but a Boileau. Its typical sample of a reproduction of the antique was Pope's translation of the Iliad. That book, we presume, everybody has read; and many of those who have read it know that, though an excellent and spirited poem, it is no more Homer than the age of Queen Anne was the age of Peisistratos. Of the translations of Dante made during this period, the chief was unquestionably Mr. Cary's. 59 For a man born and brought up in the most unpoetical of centuries, Mr. Cary certainly made a very good poem, though not so good as Pope's. But it fell far short of being a reproduction of Dante. The eighteenth-century note rings out loudly on every page of it. Like much other poetry of the time, it is laboured and artificial. Its sentences are often involved and occasionally obscure. Take, for instance, Canto IV. 25-36 of the "Paradiso":

Here Mr. Cary not only fails to catch Dante's grand style; he does not even write a style at all. It is too constrained and awkward to be dignified, and dignity is an indispensable element of style. Without dignity we may write clearly, or nervously, or racily, but we have not attained to a style. This is the second shortcoming of Mr. Cary's translation. Like Pope's, it fails to catch the grand style of its original. Unlike Pope's, it frequently fails to exhibit any style.

It is hardly necessary to spend much time in proving that Mr. Longfellow's version is far superior to Mr. Cary's. It is usually easy and flowing, and save in the occasional use of violent inversions, always dignified. Sometimes, as in the episode of Ugolino, it even rises to something like the grandeur of the original:

That is in the grand style, and so is the following, which describes those sinners locked in the frozen lake below Malebolge:—

And the exclamation of one of these poor "wretches of the frozen crust" is an exclamation that Shakespeare might have written:—

There is nothing in Mr. Cary's translation which can stand a comparison with that. The eighteenth century could not translate like that. For here at last we have a real reproduction of the antique. In the Shakespearian ring of these lines we recognize the authentic rendering of the tones of the only man since the Christian era who could speak like Shakespeare.

In this way Mr. Longfellow's translation is, to an eminent degree, realistic. It is a work conceived and executed in entire accordance with the spirit of our time. Mr. Longfellow has set about making a reconstructive translation, and he has succeeded in the attempt. In view of what he has done, no one can ever wish to see the old methods of Pope and Cary again resorted to. It is only where he fails to be truly realistic that he comes short of success. And, as already hinted, it is oftenest through sheer excess of LITERALISM that he ceases to be realistic, and departs from the spirit of his author instead of coming nearer to it. In the "Paradiso," Canto X. 1-6, his method leads him into awkwardness:—

This seems clumsy and halting, yet it is an extremely literal paraphrase of a graceful and flowing original:—

Now to turn a graceful and flowing sentence into one that is clumsy and halting is certainly not to reproduce it, no matter how exactly the separate words are rendered, or how closely the syntactic constructions match each other. And this consideration seems conclusive as against the adequacy of the literalist method. That method is inadequate, not because it is too REALISTIC, but because it runs continual risk of being too VERBALISTIC. It has recently been applied to the translation of Dante by Mr. Rossetti, and it has sometimes led him to write curious verses. For instance, he makes Francesca say to Dante,—

for

Mr. Longfellow's good taste has prevented his doing anything like this, yet Mr. Rossetti's extravagance is due to an unswerving adherence to the very rules by which Mr. Longfellow has been guided.

Good taste and poetic genius are, however, better than the best of rules, and so, after all said and done, we can only conclude that Mr. Longfellow has given us a great and noble work not likely soon to be equalled. Leopardi somewhere, in speaking of the early Italian translators of the classics and their well-earned popularity, says, who knows but Caro will live in men's remembrance as long as Virgil? "La belie destinee," adds Sainte-Beuve, "de ne pouvoir plus mourir, sinon avec un immortel!" Apart from Mr. Longfellow's other titles to undying fame, such a destiny is surely marked out for him, and throughout the English portions of the world his name will always be associated with that of the great Florentine.