Home  »  The Oxford Book of American Essays  »  XXXII On Translating the Odes of Horace

Matthews, Brander, ed. (1852–1929). The Oxford Book of American Essays. 1914.

William Peterfield Trent (1862– )

XXXII On Translating the Odes of Horace

IN a letter written on August 21, 1703, to Robert Harley, afterward Earl of Oxford and Prime Minister, by Dr. George Hickes, the famous scholar and non-juror, there is a reference to “old Dr. Biram Eaton who has read Horace over, as they tell me, many hundred times, oftener, I fear than he has read the Gospels.” Dr. Biram Eaton has escaped an article in the Dictionary of National Biography, and, so far as I know, he has never been reckoned by Horatians among their patron saints. In view of the slur cast upon him by Dr. Hickes I should like to propose his canonization, but I should much prefer to lay a wager that he found time between his readings to try to turn some of the odes of his favorite writer into English verses, probably into couplets resembling those of Dryden. And I should also be willing to wager that before and after making each of his versions, he gave expression, in some form or other, to the proverbial statement that to attempt to translate Horace is to attempt the impossible.

Perhaps we owe to this proverbial impossibility the fact that the translator of Horace is always with us. A living antinomy, he writes a modest preface; then exclaiming in the words of his master, “Nil mortalibus ardui est,” he tries to scale very heaven in his folly, to rush blindly per vetitum nefas. But because he has loved much, therefore is much forgiven him. To love Horace and not attempt to translate him would be to flout that principle of altruism in which some modern thinkers have discovered, more poetically perhaps than philosophically, the motive force of civilization. “We love Horace, and hence we must try to set him forth in a way to make others love him,” is what all translators, it would seem, say to themselves, consciously or unconsciously, when they decide to publish their respective renditions. And who shall blame them? Where is the critic competent to judge their work, who has not himself listened to the Siren’s song, if but for a moment in his youth, who has not a version of some ode of Horace hid away among his papers, the memory of which will doubtless forever prevent him from flinging a stone at any fellow-offender?

It is not only impossible to translate Horace adequately, but it is impossible to explain satisfactorily the causes of his unbounded popularity—a popularity illustrated by the fact that when that well-known group of American booklovers, the Bibliophile Society, were seeking to determine what great man of letters they would first honor by issuing one or more of his works in sumptuous form, they chose—not an author of their own day or nation or language—but a writer dead nearly two thousand years, of alien race and tongue, spokesman of a civilization remote and strange, the Horace of the immortal Odes. Yet admirers of Lucretius and of Catullus tell us very plainly and insistently that this Horace of the Odes is not a great poet. We listen respectfully to the charge and somehow we do not seem greatly to resent it; we merely read the Odes, if possible, more diligently and affectionately—not, it is true, in the splendid Bibliophile volumes, but in some well-worn pocket edition that has accompanied us on our journeys, or, like one I own, has helped us to while away the hours on a deer stand, through which the deer, as shy as the fawn with which the poet compared Chloë, simply would not run. If we own such a pocket volume, we leave our critical faculties in abeyance when Dante, in the Inferno, introduces Horace to us along with Homer and Ovid and Lucan; for do not our hearts tell us that in the truest sense of the phrase, he is worthy to walk with the greatest of this mediævally assorted company? We feel sure that Virgil must have loved him as a man; we have proof that Milton admired him as a poet. We deny to him “the grand manner,” but we attribute to him every charm. When we seek to analyze this charm, we are left with the suspicion that, after we have pointed out many of its elements, such as humor, vivacity, kindliness, sententiousness, and the like, there are as many others, equally potent but more subtle, that escape us altogether. So we turn the hackneyed saying into “the charm is the man,” and contentedly exchange analysis for enjoyment. And yet we are persuaded that no author is more worthy of the painstaking, detailed study characteristic of modern scholarship than is this same Epicurean poet, who so utterly defies analysis and would be the first, were he not but “dust and a shade,” to smile at our ponderous erudition. We feel that the scholar who shall devote the best years of his life to studying the influence of Horace upon subsequent writers in the chief literatures and to collecting the tributes that have been paid to his genius by the great and worthy of all lands and ages, will deserve sincere benedictions. We conclude, in short, that that exquisite epithet, “the well-beloved,” so inappropriately bestowed upon the worthless and flippant French King, belongs to Horace, and to Horace alone, jure divino.

But this praise of Horace and this defense of his translators fails to justify or explain the writing of this paper. An honest confession being good for the soul, I will confess that the remarks that follow were first employed to introduce some versions of selected Odes I was once rash enough to publish. It is not a good sportsman that shuts his eyes and bangs away with both barrels at a flock of birds, and I now doubt whether I was wise in trying to bring down readers, if not with my verse-barrel, at least with my prose-barrel. Being older, I use at present only one barrel at a time and, perhaps for the same reason, I am wont to try the prose-barrel. And fortunately I can apply to the comments I intend to make on Horatian translators the quotation I used in order to mollify irate readers of my own verse renderings. It came from a once popular, now forgotten poet, the Rev. John Pomfret, and it ran as follows:—“It will be to little purpose, the Author presumes, to offer any reasons why the following POEMS appear in public; for it is ten to one whether he gives the true, and if he does, it is much greater odds whether the gentle reader is so courteous as to believe him.”

So much has been written on the methods of Horace’s translators, and so much remains to be written, that it is hard to determine where to being; but perhaps the preface of the late Professor Conington to his well-known translation of the Odes will furnish a proper point of departure. Few persons, whether translators or readers, are likely to object to Conington’s first premise that the translator ought to aim at “some kind of metrical conformity to his original.” To reproduce an original Sapphic or Alcaic stanza in blank verse, or in the couplets of Pope, is at once to repel the reader who knows Horace well, and to give the reader who is unacquainted with Latin lyric poetry a totally erroneous conception of the metrical and rhythmical methods of the poet. To render a compressed Latin verse by a diffuse English one is to do injustice, as Conington observes, to the sententiousness for which Horace is justly celebrated, although the English scholar, had he written after the appearance of Mr. Gladstone’s attempt to render the Odes, might with propriety have added that the translator should not, in his avoidance of diffuseness, be seduced by the facility of the octosyllabic couplet. To translate Horace’s odes into anything but quatrains, except on occasions, is also to offend the meticulous Horatian and to mislead any reader who seeks to know the poet through an English rendering. It would seem, however, that when Professor Conington insisted that an English measure once adopted for the Alcaic must be used for every ode in which Horace employed the stanza just named, he went far toward hampering the translator, who, despite his proneness to offend, has his rights. That such uniformity ought to be aimed at, and that it will, as a rule, be aimed at, is doubtless true; but there is an element of the problem with which Conington does not seem sufficiently to have reckoned.

This is rhyme, which he assumed to be necessary to a successful rendition of an ode of Horace. A particular stanza not employing rhyme may probably be used without resulting loss in translating every ode written in a special form. Yet this may not be the case with a stanza employing rhymes, if the translator aim, as he should, at a fairly, though not an awkwardly literal rendering of the language of his original. There will necessarily be coincidences of sound in a literal prose version of a Latin stanza that will suggest a definite and advantageous arrangement of rhymes for a poetical version. To adopt a certain English stanza in which to render a certain Latin stanza wherever it occurs, is to do away with this natural advantage, which presents itself oftener than might at first be supposed.

Concrete examples will serve to make my meaning clear. The third ode of the first book, the admirable “Sic te diva potens Cypri,” is written in what is called the Second Asclepiad meter; so is the delightful ninth ode of the third book, the “Donec gratus eram.” We will assume that for the first of these odes the translator has chosen a quatrain with alternating rhymes (a, b, a, b). Following Professor Conington’s rule of uniformity, he must employ the same stanza for the second of the two odes, which, by the way, Conington himself did not do, for reasons which he gave at length. Now the fifth stanza of the “Donec gratus eram” runs as follows:—

  • “Quid si prisca redit Venus
  • Diductosque jugo cogit aëneo,
  • Si flava excutitur Chloë
  • Rejectaeque patet janua Lydiae?”
  • This may be rendered in prose:—
  • “What if the former Love return and join with brazen yoke the parted ones, if yellow-haired Cholë be shaken off, and the door stand open for rejected Lydia?”
  • If my memory does not deceive me, it was this stanza, and especially one word in its last verse, that determined the arrangement of rhymes in a version I attempted years ago, “Consule Planco.” This verse seemed to run inevitably into

  • “And open stand for Lydia the door.”
  • It needed but a moment to detect in the first verse of the stanza a possible rhyme-word. The syllable re of redit furnished more, not the most apt of rhymes with door, but still sufficient, as things go with amateur translators, and with a perhaps pardonable tautology I wrote

  • “What if the former Love once more
  • Return—”
  • Two others rhymes were found with little difficulty in the di of diductos and in excutitur, which suggested wide and cast aside, and the whole stanza, omitting strictly metrical considerations, appeared, or rather might have appeared, for I have changed it slightly, as follows:—
  • “What if the former Love once more
  • Return and yoke the sweethearts parted wide,
  • If fair-haired Chloë be cast aside,
  • And open stand for Lydia the door?”
  • This stanza seemed to have the merit of almost complete literalness, since it omitted only two epithets, and I thought it had no unpardonable defects of rhythm and diction. So I took it as a model, and with little difficulty translated the entire ode—with what success I should not say and others need not inquire.

    That rhymes and their position in the stanza are often determined for the translator by his original or else by a prose rendering of that original seems also to be shown by the following version of the closing ode of the first book (Carm. xxxviii)—the graceful “Persicos odi”:—

  • “I hate your Persian trappings, boy,
  • Your linden-woven crowns annoy,
  • Cease searching for the spot where blows
  • The lingering rose.
  • “To simple myrtle nothing add;
  • The myrtle misbecomes, my lad,
  • Nor thee nor me drinking my wine
  • ’Neath close-grown vine.”
  • Here “puer,” boy, and “Displicent,” displease or annoy, seem to determine, not merely the first rhyme, but the rhyme arrangement (a, a), and it needs but a glance at the close of the first stanza of the original to show that another word rhyming with “boy” would be hard to obtain. It follows that, if we are to have a quatrain, the third and fourth verses should probably be made to rhyme (b, b), and it is not difficult to comply with this requirement, or to cast the second stanza in the mold of the first. It is, alas! too true that no equivalent to “blows” will be found in Horace, that “Sedulus curo” has been unceremoniously thrown aside, that the poet does not specifically mention “wine” as the beverage he liked to drink in his rustic arbor. But a “rose,” which Horace does mention, certainly “blows” or blooms very often in English verse; it is not too far-fetched to get “nothing add” and “lad” out of “nihil allabores” and “ministrum”; and “vine” (“vite”) has suggested “wine” to many generations of poets. But it is rhyme suggestions and their influence upon the choice of stanzaic form that have occasioned this mild protest against Professor Conington’s precepts of rigid stanzaic conformity. I am convinced, from the above examples and from many more, not only that uniformity of stanza is not to be strictly insisted upon when one is employing rhymes, but also that translators should search more diligently than they appear to do for the rhyme suggestions implicit in so many Horatian stanzas.

    Upon other points it is easier to agree with Conington. For most of the odes the iambic movement natural to English is preferable, as Milton may be held to have perceived. He abandoned rhyme in his celebrated version of the “Quis multa gracilis” (i., v.), and hence he had an excellent opportunity to indulge in experiments in so-called logaœdic verse. But he clung to the iambic movement, and the fact is significant, although not to be pressed, since he gave us no other rendering of an entire ode. Here too, however, I must plead for a careful study of each ode by the would-be translator, for there seem to be cases in which it would be almost disastrous to attempt a version in iambics. Such a case is presented by the beautiful “Diffugere nives” (iv., vii.). The iambic renderings of Professor Conington and Sir Theodore Martin seem to stray far from the original movement—as far as the former’s “‘No ’scaping death’ proclaims the year” does from the diction of Horace or of any other good poet. It is true that English dactyls are dangerous things, especially in translations, where the padding or packing which is natural to the measure when employed in English, is increased by the padding inevitably introduced into a translation from a synthetic into an analytic language. Yet the dactylic movement of the First Archilochian, in which the “Diffugere nives” is written, is hardly without great loss to be represented by any use of English iambics. It presents more difficulty than the introduction of something resembling the movement of dactylic hexameters proper into our blank verse.

    When the translator makes up his mind to attempt a close approximation to the Horatian meter, it would seem that he should eschew the use of rhyme as likely to operate against that effect of likeness to the original which he is striving to secure. But, since the use of rhyme in lyric poetry appears, as Conington held, to be essential at present if the English version is to be acceptable as poetry, this close approximation can be desirable in a few special cases only. It will not do to dogmatize on such matters, but it may be safely said that no poet, not even Milton or Whitman, has yet accustomed the English or the American ear to the use of rhymeless verse in lyrical poetry. Here and there a successful rhymeless lyric, such as Collins’s “Ode to Evening” and Tennyson’s “Alcaics” on Milton, shows us that rhymeless stanzas may occasionally be used for lyric purposes with good effect; but thus far those translators of Horace who have made a practice of eschewing rhyme have failed, as a rule, like the first Lord Lytton, to give us versions that charm. Yet charm is what the translator of Horace should chiefly endeavor to convey.

    I am still more confident that Conington was right when he insisted that the English rendering houls be confined “within the same number of lines as the Latin.” He was surely right when he taxed Sir Theodore Martin, who so frequently violated this rule, with an exuberance that is totally at variance with the severity of the classics. Such exuberance is almost certain to result if the translator abandon the strict number of the lines into which the Roman poet compressed his thought. It results, too, from the use of stanzas of over four verses each. There is no other rule of translating that will so effectively insure a successful retention of the diction of the original as this of the line for line rendering, whenever such rendering is possible. And that the diction and the thought of the poet should be more closely followed than is usually the case, admits of no manner of doubt. We have already seen that a close scrutiny of the Latin will often suggest an almost literal rendering of the thought and diction. Such a rendering is more desired by the reader who is familiar with Horace than by the reader who is not, but it will be both pleasing and serviceable to the latter, if the quality of literalness be not too slavishly obtained. Metrical considerations and general smoothness ought, as a matter of course, to weigh with every translator, but surely they ought not to outweigh accurate rendering of diction and thought, especially the diction and thought of a poet so felicitous as Horace in his phrasing, and so just and happy in his observation of life.

    In this connection I am not sure but that Conington went too far when he recommended the Horatian translator to hold by the diction of our own Augustan period. That the Age of Pope corresponds in many ways with that of Horace is true enough, and the student of the poetry of the eighteenth century who cares at all for the poets he studies is almost sure to be an admirer of the “Roman bard” whom Pope imitated. But the diction of Horace does not strike one as stilted, while that of Pope often does; and for a translator of our own days to employ a diction that seems in any way stilted is fatal not merely to the popularity and hence to the present effectiveness of his work, but also, in all probability, to its intrinsic value. There is a good deal of the commonplace also in the poetry produced in the eighteenth century; but commonplace the translator of Horace can least afford to be. Horace himself may approach dangerously near the commonplace, yet he seems always to miss it by a dexterous and graceful turn. The translator, running after, will miss this turn sufficiently often, as it is; he cannot, therefore, afford to steep himself in a literature that has a tendency to the commonplace. But just as little can he afford to steep himself in the Romantic Poets from Shelley to Swinburne. A translation, whether from the Greek or the Latin, imbibing the luxuriance of imagination and phrasing characteristic of these modern poets, may satisfy a reader still in his intellectual teens, but the reader who makes use of a translation of Horace is likely to have passed out of that period of immaturity. It may be heretical, but I fancy that the translator of Horace who steeps himself in Keats or Tennyson, will be even less likely to give us the ideal rendering than the translator who steeps himself in Pope. Luxuriance and elegance may at times be more displeasing than excessive polish and point.

    To mention the eighteenth century is to bring up the thought of Horatian paraphrases. A successful paraphrase is sometimes better as poetry than a good poetical translation, and it not infrequently conveys a juster idea of the spirit of Horace. It is almost needless to praise the work in this kind of Mr. Austin Dobson and of the late Eugene Field. But a paraphrase, however good, can never be entirely statisfying either to the reader that knows Horace or to the reader that desires to know him. Nor can a prose version be thoroughly satisfactory. What is wanted is not merely the drift of the poet’s thought, but, as near as may be, what he actually sang. The paraphrase may sing, and the prose version may give us the thought in nearly equivalent words, which may carry along with them not a little of the poet’s feeling; but neither answers all our requirements as well as a good rendering in verse may do—such a rendering, for example, as that which the late Goldwin Smith gave of the “Cœlo tonantem” (iii., v.)—yet there is surely room for all these forms of approach to a poet who is, paradoxically enough, at one and the same time, the most approachable and the most unapproachable of writers.

    But one could write forever upon the topic of poetical translation in general, and of the translation of Horace’s odes in particular. It is a subject about which people will differ to the end of time; a subject the principles of which will never be thoroughly exemplified in practice. Still, it always seems to fascinate those who discuss it, and they have a way of hoping that what they have said about it will not be without value to those who want to read about it. “Hope springs eternal in the human brest,” said the poet who also wrote of his great master lines that have not been surpassed in their kind:—

  • “Horace still charms with graceful negligence,
  • And without method talks us into sense,
  • Will like a friend familiarly convey,
  • The truest notions in the easiest way.”