This article examines five Chinese translations of Dante’s Divina Commedia, the latest version by the author of this article, which was published in 2003. After making some general observations, the article goes on to discuss translation problems relating to medium, rhythm, metre, rhyme, etc. with reference to Chinese and English versions of Dante’s masterpiece. By substantiating its argument with plenty of examples as well as with a comprehensive survey of the prosodic characteristics of the Chinese and European languages, it establishes criteria by which translations of the Divina Commedia should be judged.
Cet article examine cinq traductions chinoises de la Divina Commedia de Dante, dont la dernière version, publiée en 2003, est de la plume de l’auteur de cet article. Après avoir fait des remarques générales, l’auteur discute des problèmes de traduction relatifs à l’expression, au rythme, au mètre, à la rime, etc., faisant allusion aux traductions en chinois et en anglais du chef-d’oeuvre de Dante. En appuyant le raisonnement sur plusieurs exemples ainsi que sur une étude détaillée des caractéristiques prosodiques de la langue chinoise et des langues européennes, il établit des critères selon lesquels les traductions de la Divina Commedia devraient être jugées.
- moyen d’expression,
Corps de l’article
Translating the Divina Commedia into Chinese: General Observations
From the 1950s to 2000, four complete Chinese translations of Dante’s Divina Commedia were published, two in prose and two in verse. In going through these translations, one becomes aware of a number of problems that prevent them from meeting the needs of a more sophisticated, more demanding Chinese reading public in the twenty-first century. In this paper, I shall discuss these problems and describe what I have done to tackle them in my own, also the latest Chinese translation of the Commedia, entitled Shenqu 神曲 ‘The Divine Comedy’ and published under my Chinese name Huang Guobin in 2003 by Chiuko Press in Taipei, Taiwan in three volumes, subtitled Diyupian 地獄篇 ‘Inferno’, Lianyupian 煉獄篇 ‘Purgatory,’ and Tiantangpian 天堂篇 ‘Paradise’, respectively, corresponding to the three cantiche of the Commedia, namely, the Inferno, the Purgatorio, and the Paradiso. In my discussion, I shall try to show that verse is a more adequate medium for coping with the source text than prose, and that, in respect of terza rima, modern Chinese enjoys an advantage over modern English. As the focus of my paper is on practice and on translation as a craft and art, aimed at sharing the translation experience with practitioners and practitioner-theorists, abstract discussion in purely theoretical terms will be avoided.
Of all the requirements for a credible translation of the Commedia, the first is related to medium: to preserve as many poetic qualities of the original as possible, the translation has to be done in verse. The reason for this will become clear if we examine the two prose versions (Wang, 1954; Tian, 1997) alongside the original. Deprived of the resources of verse, prose translations have difficulty reproducing the effects achieved by Dante through poetic devices, especially those on the syntactic and phonological levels. Take the mounting tension and sense of expectancy suggested by the rhythm in the following lines:
Divina Commedia: Purgatorio, II, 13—18
Mandelbaum, 1982-84, Purgatorio, p. 13
In Wang’s version (Wang, 1954, p. 176), only the semantic level is taken care of; what comes through the language of poetry, such as the mounting tension and sense of expectancy mentioned above, has been reduced to a minimum:
The sentence “Na hongguang you haishang xiang women lai, bi niaofei hai yao kuai 那紅光由海上向我們來, 比鳥飛還要快” ‘The red light comes towards us from the sea, swifter than a bird’, because of its unvaried rhythm resulting from an unduly large number of three-syllable pauses in a row (Na / hong/guang// you / hai / shang// xiang / wo / men// lai,// bi / niao/ fei // hai / yao / kuai”), sounds monotonous, and fails to suggest the swiftness of the approaching angel.
Published more than forty years later, Tian’s version (Lianyupian 煉獄篇 ‘Purgatory’, 1997, p. 12) is more meticulous, able to attend to the phrase “sul presso del matino” ‘on the approach of morning’ (Sinclair, 1971, Purgatorio, p. 33), which Wang has left untranslated:
However, denied the resources of verse for approximating Dante’s poetic devices and hampered by a sluggish rhythm resulting from the unskillful arrangement of one-character, two-character, and three-character pauses in “ren/he // niao/fei/de // su/du // dou // bi/bu/shang // ta/de // yun/dong” ‘no speed of birds in flight is comparable to its movement’, which is hardly in keeping with the swiftness described in the original, it is ill-equipped to re-create the original poetic qualities.
With verse as my medium, I find it easier to suggest the original’s mounting tension and sense of expectancy. Through the functional arrangement of pauses in lines 4-6 and the interplay of the four tones in the rhyme scheme, I believe that I have also been able to emphasize the speed of the angel and re-create the musicality of terza rima:
Huang Guobin [Laurence Wong], 2003, Lianyupian, pp. 25-26
Using the same medium, Zhu should, in theory, have an advantage over Wang and Tian; in practice, however, he has a different limitation, which leads my discussion to the second requirement for a credible Chinese version of the Commedia: that it should be a direct translation from the Italian. As an indirect recipient of the message conveyed by Dante, Zhu often has difficulty gauging its impact with precision, as is the case with the following scene:
Divina Commedia: Inferno, III, 130-136
Mandelbaum, 1982, Inferno, p. 27
The sonorous “o” in “Finito”, “questo”, “tremò”, “forte”, “lo”, “spavento”, “sudore”, “ancor”, “lagrimosa”, “vento”, “balenò”, “sentimento”, “come”, “uom”, and “sonno” and the emphatic “en” in “spavento”, “mente”, “vento”, and “sentimento” work together like hammer-strokes to drive home the sense of shock and violence.
In Zhu’s indirect verse translation, the original images are retained, and the forceful words “feng風” ‘wind’ and “guang光” ‘light’ in lines 4 and 5 have re-created some sense of violence:
Zhu, 1984, Diyupian, p. 22
In terms of phonological effects, however, it is still a far cry from the original. Had Zhu translated the Inferno directly from the Italian, the construction “huixiangqi / wo nashide kongbu hai shi wo hunshen chuzhe lenghan 回想起 / 我那時的恐怖還使我渾身出著冷汗” ‘recalling my terror I experienced then, I am still sweating all over’ in lines 2 and 3, made limp by the clogged rhythm of “hui/xiang/qi //wo// na/shi/de// kongbu” ‘recalling my terror I experienced then’, the feeble form word “le了”, and the attributive adjectival structure “turan shuiqude 突然睡去的” ‘like…sleep has seized’, which has weakened the rhythm of line 7, would probably have been replaced by something more suggestive of violence.
Translating directly from the Italian, I find it easier to make accurate decisions as to what phonological effects to preserve:
Huang Guobin [Laurence Wong], 2003, Diyupian, p. 141
Apart from “guangmang 光芒” ‘flash’(line 5) and “shang上” ‘on’ (line 7), which are loud and heavy in phonological terms because of the open vowel “a”, “I have used “huoran 霍然” ‘suddenly’ (line 5) to convey the sense of abruptness suggested by “balenò” ‘flashed’. Conscious of the need to re-create a scene of violence and shock, I have avoided using feeble form words like le了 (an aspect word here) in rendering “e caddi come l’uom che’l sonno piglia” ‘and fell like one whom sleep has seized’; instead, I have tried to make my version as forceful as possible by tightening up the rhythm and using a loud-sounding word (“shang上” ‘on’) to bring up the rear: “xiang chenshuide ren dao zai dishang 像沉睡的人倒在地上” ‘fell on the ground like one whom sleep has seized’.
In comparing Zhu’s version with mine, one will also see a difference with respect to the translation of “lagrimosa” ‘tear-drenched’. In rendering “lagrimosa” as “yincancan 陰慘慘” ‘gloomy’, Zhu has left out one important message of the original, which is both literal and symbolic. According to the Italian commentator Chiappelli (1972, p. 40), “lagrimosa” means “triste, irrorata di lagrime” ‘gloomy, bedewed by tears’. As the scene describes damned souls entering hell to be punished, the tear image is highly symbolic, and should, for this reason, be preserved. Without direct access to “lagrimosa”, which derives from “lagrima” ‘tear’, a translator is unlikely to see how important it is to look for an equally specific, equally evocative, and equally symbolic word in the target language. In my version, I have taken the image into consideration and rendered the line as “Hanleide tudi ba yizhen yinfeng chuiqi 含淚的土地把一陣陰風吹起” ‘A whirlwind burst out of the tear-drenched earth,” which matches the original’s atmosphere on both the mimetic and symbolic levels.
Translating the Commedia from the Italian into Chinese verse, Huang Wenjie should be the best-qualified translator of the four, potentially able to do better than Wang, Zhu, and Tian. In practice, however, he has put on a rather mixed performance. With respect to the lines quoted above, his version is an improvement upon those of the other three:
Huang, Wenjie, 2000, Diyupian, p. 28
The rhythm is more precise, and the image of “lagrimosa” ‘tear-drenched’ is preserved in “leishui jintou de 淚水浸透的” ‘bedewed by tears’; yet, he seems unable to exercise control over line-length, so that the first line gets too long and unwieldy, thereby undermining the effectiveness of metre, which works on the basis of variety in regularity. Properly manipulated, metre can enable a poet to create contrapuntal effects by gliding back and forth within a line, as well as by allowing free play to pauses, feet, or, in the case of Greek and Latin poetry, quantity. When line-length gets out of control, as is the case with Huang Wenjie’s version, metre as a means of creating phonological effects will be vitiated; in extreme cases, it may even cease to function. In Huang Wenjie’s translation, whether with regard to the Inferno, the Purgatorio, or to the Paradiso, many lengthy lines simply sprawl into shapeless prose.
In his translation of the climax of the Commedia, Huang Wenjie’s inadequacies become most obvious:
Divina Commedia: Paradiso, XXXIII, 133-145
Inspired, as it were, by the Holy Spirit, Dante’s imagination in these lines soars from pinnacle to pinnacle without showing any sign of fatigue. The reader, rhythm-drunk and in transports of joy, is hurled higher and higher until he reaches the limit of all human experience—the experience of sharing the Beatific Vision with Dante the Pilgrim, an experience beyond words and understanding. To appreciate how formidable the task of translating the above quotation can be, one has only to read the following comment by Sinclair:
Nowhere else does Dante attain to the greatness of the last canto of the Paradiso, and in it more than any other it must be remembered that a canto is a song. Here his reach most exceeds his grasp, and nothing in all his work better demonstrates the consistency of his imagination and the integrity of his genius. In the culmination of his story he reports his experience with such intensity of conviction, in a mood so docile and so uplifted, and in terms so significant of a vision at once cosmic and profoundly personal, that we are persuaded and sustained to the end.Sinclair, 1974, Paradiso, p. 487
Words like “Nowhere else”, “greatness”, “his reach most exceeds his grasp”, “the consistency of his imagination”, “the integrity of his genius”, “the culmination of his story”, “such intensity of conviction”, “uplifted”, “so significant of a vision”, “at once cosmic and profoundly personal”, and “sustained to the end” are sufficient warning to the translator that, in rendering these lines, he should be prepared to find his abilities taxed to the utmost.
To his readers’ disappointment, however, Huang Wenjie has failed to rise to the challenge. Not only has he misinterpreted two crucial lines (lines 139-40), in which “se non” means “if not”, not “chufei 除非” ‘unless’, as is understood by Huang, but he has also failed to convey Dante the Pilgrim’s ecstasy, which the original imagery and rhythm, working in perfect unison, have so powerfully expressed. Although meant to be poetry, the translation sounds like prose, for no poet with an ear for rhythm would, at such a crucial moment, allow himself to produce limp, monotonous lines with seven two-character pauses coming in a row as Huang has done, seriously retarding the crescendo of the original movement: “Ru/tong // yi/wei // ji/he // xue/jia // qing/zhu // quan/bu // xin/xue如同 / 一位 /幾何 / 學家 / 傾注 / 全部 / 心血” ‘As the geometer intently seeks’ . Nor would such a poet tolerate the jumble of ineffective syllables in lines 3-5, “ta baiban sicun, ye wufa ba ta suo xuyaode nage yuanli tanxun, / wo cike miandui na xinqide jingxiang ye shi zheizhong qingxing: /wo xiang kanqing: na renxing ruhe yu na guangquan xiang shiying 他百般思忖, 也無法把他所需要的那個原理探尋, / 我此刻面對那新奇的景象也是這種情形: / 我想看清: 那人形如何與那光圈相適應”  ‘but he cannot reach, / through thought on thought, the principle he needs, / so I searched that strange sight: I wished to see / the way in which our human effigy / suited the circle’, which drag on with little vigour. Towards the end of the poem, the translation (lines 9-11) continues to flag, until it ends in an anticlimax: “zhengshi zhei ai tuidong taiyang he qita qunxing正是這愛推動太陽和其他群星” ‘the Love that moves the sun and the other stars’. As the Chinese counterpart of the last—perhaps also the most powerful—line of the entire Commedia, the rendering is a total disaster. Whereas the original line “l’amor che move il sole e l’altre stelle” ‘the Love that moves the sun and the other stars’ is worthy of the lofty task assigned to it, functioning as the last notes of a perfect coda to a poem of 14,233 lines, with the emphatic “o” and “e” in “amor”, “che”, “move”, “sole”, “altre”, and “stelle” resonating and rhyming across the line, the Chinese translation can hardly match the original; impeded by four two-character pauses coming in a row (“zheng/shi // zhei / ai // tui/dong // tai/yang正是 / 這愛 / 推動 / 太陽” ‘the Love that moves the sun and the other stars’, it just dies out in a whimper.
In a way, one is not even sure whether Huang Wenjie’s rendering of the last lines of Canto 33 of the Paradiso is comparable with Wang’s prose version:
Wang, 1954, p. 546
Though Wang has made the same mistake as Huang Wenjie in interpreting lines 140-41 of the original, his rendering of the last line (“shi / ai / ye,// dong / tai/yang // er // yi / qun/xing 是愛也, 動太陽而移群星” ‘the Love that moves the sun and the other stars’), with a more forceful, more varied rhythm made possible by the one-character pause “//er//” inserted between the two three-character pauses, “//dong / tai/yang // and // yi / qun/xing//”, conveying the same clinching effect as the original, is far superior to its counterpart in Huang Wenjie’s (2000) version.
Translating Dante’s Metre in a Different Linguistic Medium
To capture the rhythm, the emotional intensity and, above all, the dizzy height the poet’s art has attained, one has to rely on verse:
Huang Guobin [Laurence Wong], 2003, Tiantangpian, pp. 510-511
In my version, I have used the five-pause line to translate Dante’s hendecasyllable (eleven-syllable line). My choice was made with reference to Western poetry and in the light of actual practice. In long narrative poems—especially epics—written in European languages such as English, Greek, and Latin, the average line-length normally falls within the range of the English pentameter, the Greek dactylic hexameter, or the Latin hexameter. The English pentameter as used in long poems is best exemplified by Milton’s decasyllabic line in Paradise Lost:
Paradise Lost, VI, 1-5
All night | the dread|less an|gel un|pursued
Through heav’n’s | wide cham|paign held | his way, | till Morn,
Waked by | the cir|cling Hours, | with ros|y hand
Unbarred | the gates | of light. | There is | a cave
Within | the mount | of God, | fast by | his throne…,
while the “standard” Greek line is to be found in the opening of Homer’s Iliad:
Murray, 1924, p. 2
Murray, 1924, p. 3
With regard to Latin poetry, the representative line is provided by Virgil’s Aeneid:
Aeneid, I, 418-20
Fairclough, 1999, p. 290
In addition to what I have observed in the works of the masters of epics, I have also discovered, during the actual translation process, that the average hendecasyllabic line in Dante’s Divina Commedia is best matched by the five-pause (five-foot) line in modern Chinese. Take the first five lines of my translation quoted above:
One can see that they have accommodated the original sense-units quite comfortably, and that the need to compress content unduly or to pad out a line for metrical reasons has been reduced to a minimum. If I had opted for a longer or shorter line, that is, a line with more or fewer pauses, I would have been harder put to avoid padding or to cope adequately with the original sense-units. In general, I find that, within five pauses, there is just about the right amount of space for me to reproduce the music of the original by moving the caesura back and forth, by introducing variation, by deploying one-character, two-character, and three-character pauses as required, and by creating contrapuntal effects as well as tension between similarity and variety.
Terza Rima in Modern Chinese versus Terza Rima in English
As for rhyming, I have followed Dante’s terza rima. Assigned prosodic as well as symbolic functions, terza rima in the Commedia is no mere embellishment. Prosodically, it gives the poem an interlocking unity in which lines, while resonating across the page, echo each other within each tercet. Take the following lines, for example:
Paradiso, XXXIII, 115-20
Mandelbaum, 1986, p. 303
Together with the whole complex of vowels (“a”, “e”, “i” ,”o”, “u”) and the consonant “n” (“profonda”, “d’una”, “l’un”, “quinci”, “quindi”), the rhyme scheme (“sussistenza”, “giri”, “contenenza”, “iri”, “foco”, “spiri”) weaves an intricate and highly effective pattern of sounds to suggest the ever increasing rapture of the Pilgrim.
By giving up terza rima, English translators like Mark Musa can convey only part of the emotional intensity:
Paradise, XXXIII, 115-20
Within Its depthless clarity of substance
I saw the Great Light shine into three circles
in three clear colors bound in one same space;
the first seemed to reflect the next like rainbow
on rainbow, and the third was like a flame
equally breathed forth by the other two.
This difference between the original and the translation will become audible if one reads the Italian and English quotations aloud: in the original, there is an exciting symphony of interlocking sound patterns echoing one another, delighting the ear and reinforcing the mounting emotional intensity at the same time; in the translation, the semantic content is retained, but the symphony can no longer be heard. In the light of this comparison, then, Musa’s defence of his not using terza rima merits detailed discussion:
My desire to be faithful to Dante, however, has not led me to adopt his metrical scheme. I do not use terza rima, as, for example, Dorothy Sayers does, or even the “dummy” terza rima of John Ciardi. My medium is rhymeless iambic pentameter, that is, blank verse. I have chosen this, first, because blank verse has been the preferred form for long narrative poetry from the time of Milton on. It cannot be proved that rhyme necessarily makes verse better: Milton declared rhyme to be a barbaric device, and many modern poets resolutely avoid it.Musa, 1986, Paradise, p. 61
At first sight, the defence appears to be an excuse made by one who has not done his job properly. As one reads on, however, one will see that he has put his finger on an obstacle that English translators cannot easily surmount:
But my main reason for avoiding rhyme has been the results achieved by all those who have used rhyme in translating The Divine Comedy: they have shown that the price paid was disastrously high (…).There are two reasons for the crippling effects of rhyme in translating a lengthy poem. First of all, it is apparently impossible always to find perfect rhymes in English for a long stretch of lines—and if good rhyme gives a musical effect, bad rhyme is cacophonous; it is a reminder (…) that the search for rhyme has failed (…). One can be more faithful to Dante (…) by avoiding rhyme than by introducing imperfect rhyme into the rendition of his lines, whose rhymes are always acoustically perfect.Musa, 1986, Inferno, pp. 61-62
In going through the translations of the Commedia that employ rhyme, such as the two mentioned by Musa (one by Sayers and the other by Ciardi), one does see “the crippling effects of rhyme in translating a lengthy poem”, and tends to agree that “it is apparently impossible always to find perfect rhymes in English for a long stretch of lines”. In using terza rima to translate the Commedia, an English translator will, much to his chagrin, find that he cannot be faithful to the rhyme scheme without doing severe violence to the meaning, so that his gain will be outweighed by his loss. Perhaps for this reason, no English translations of the Commedia, whether partial or complete, published in recent years have used rhyme in general, much less the very “crippling” terza rima. When I say “English translations of the Commedia, whether partial or complete, published in recent years”, I have in mind the versions by Sisson, Durling, and Zappulla respectively. Of the three translators, Durling’s case is especially noteworthy. In the Introduction to his translation, Durling is keenly aware of the symbolic significance of Dante’s terza rima:
Terza rima is an extremely supple and flexible medium. In the Divine Comedy, there is no set number of lines in a canto; the cantos range in length from 115 to 160 lines. It is clear that Dante associated the triplicities of the form (groups of three lines, interlocking chains of three rhymes) with the idea of the Creator as triune and with the idea of the chain of being. In the wake of Saint Augustine’s De Trinitate, he saw the marks of the Creator’s triple unity everywhere in creation—in the structure of time (past, present, and future), in the triple structure of man’s nature (rational, appetitive, and vegetative), and in the three “first things” (form and matter, separate and conjoined)—and regarded his verse medium, terza rima, as one of the ways his creation of the poem imitated God’s creation of the universe: others are, of course, that the poem has three parts and that it consists of a “perfect” number of cantos, 100—or, after the prologue of the first canto, three parts of thirty-three cantos each.Inferno, 1996, p. 23
Yet, in his translation, he has avoided it, probably because, like Musa, he knows how helpless the English language is when confronted with this intractable rhyme scheme.
Not so with modern Chinese. To be sure, translating terza rima into modern Chinese is an arduous task, but I have found that modern Chinese has more affinities with Dante’s Italian than English and classical Chinese. This is because unique syllable-endings in modern Chinese, called by Chinese linguists “yunmu韻母” ‘simple or compound vowels of a Chinese syllable, sometimes with a terminal n or ng’ (Wu, 1983, p. 860), make up only a relatively small number. Compared with the rich and diverse syllable-endings of English words and with the larger variety of consonant endings of classical Chinese, which make the task of finding matching rhymes formidable, they are more “rhyme-friendly” when it comes to using terza rima. Thus, in my translation of the Commedia, I have chosen to retain this prosodic feature, a feature that is anathema to English translators. Because of the linguistic “advantage” I enjoy, I believe that, in my version of the lines quoted above, I have generally succeeded in achieving my goal: to convey the original’s emotional intensity and re-create its phonological features, features that include the resonance between vowels within the lines, the echoing between vowels and sound patterns, and the orchestration of vowels, sound patterns, and the rhyme scheme:
Words like “guang光” (“lume” ‘light’), “shen深” (“profonda” ‘profound’), “huan環” (“giri” ‘circles’), “tong同”(“d’una” ‘of one’), “can燦” (‘bright’, translation of the meaning implied in the original), “hong虹”(“iri” ‘rainbow’), “ying映”(“reflesso” ‘reflected’), “san三”(“tre” ‘three’), “hun渾”(“igualmente” ‘all in one’), “ran然” (Chinese function word), and “yan焰”(“foco” ‘flame’) echo one another with their vowels, as well as with their “n”s or “ng”s, creating an intricate phonological pattern that suggests the emotional intensity of the original; reinforced by the rhymes (“jiao皦”—“xiao小”, “huan環”—“ran然”—“zhuan轉”), some of which in turn set off sub-resonance, as it were, in the diezi疊字 ‘reduplication’ “jiaojiao皦皦” ‘bright’, as well as in the dieyun疊韻 ‘rhyming collocation’ “canran燦然” ‘brightly,’ this phonological pattern becomes all the more powerful, creating effects which are, to use Nida’s terminology, dynamically equivalent to those in the original.
In an article entitled “What is Minor Poetry?”, Eliot has made the following point about Crabbe:
I think that George Crabbe was a very good poet, but you do not go to him for magic: if you like realistic accounts of village life in Suffolk a hundred and twenty years ago, in verse so well written that it convinces you that the same thing could not be said in prose, you will like Crabbe.Eliot, 1957, p. 49
If there are “accounts of village life” that “could not be said in prose”, how much more is this dictum true of accounts of Hell, Purgatory, and Paradise! In using verse as a medium of translation to re-create the effects produced by the prosodic devices used by Dante, I hope that I have minimized the inevitable refraction of “la profonda e chiara sussistenza / de l’alto lume” ‘the profound and clear / essence of that lofty Light’.
At the beginning of the twentieth century, when translation from the major European languages like English, French, Italian, German, and Spanish into Chinese was a relatively new activity, it was not uncommon to see someone who did not know any English or French become a well-known “translator” of English and French novels. Of this class of “translators”, the most famous was Lin Shu林紓 (1852-1924), who, by rewriting in classical Chinese what he was told by someone who knew the original, had “translated” some 170 works written in European languages. In the twenty-first century, when translation from the major European languages has developed for nearly a hundred years, and when the Chinese reading public, with its taste made more sophisticated by a host of fine translations, has become more demanding, this kind of “story-retelling” is no longer acceptable. In translating a masterpiece like the Divina Commedia for the Chinese reading public in the twenty-first century, one has to go many steps further: not only must one translate from the original Italian, but one must use a medium which can reproduce as many qualities of the original as possible. By this, I mean verse in modern Chinese that can cope with all the prosodic devices used by Dante, including his terza rima.
Laurence K. P. Wong (Huang Guobin) is Professor of Translation in the Department of Translation, The Chinese University of Hong Kong, Hong Kong. He translates from Chinese into English as well as from English, French, Italian, German, and Spanish into Chinese. He has published 12 books of poems, six collections of lyrical essays, seven collections of critical essays, two collections of essays in translation studies, an English translation of a book of Chinese poems, and an annotated Chinese terza rima translation (in three volumes) of Dante’s Divina Commedia.
These are Wang Weike 王維克 (1954); Zhu Weiji 朱維基 (1984); Tian Dewang 田德望 (1997); Huang Wenjie 黃文捷 (2000).
Since the May Fourth Movement, which took place in 1919 in China, Chinese translations of works of Western literature, particularly European literature, have multiplied. Brought up over the past eighty years on these translations, some of which are of high quality, the reading public is likely to use a rigorous yardstick to measure any new translations of the Divine Comedy; hence one is justified to say that the Chinese reading public has become “more sophisticated, more demanding.”
To contextualize my argument, a brief account of the metre and form of modern Chinese verse may be helpful. When modern Chinese poets began to depart from classical Chinese verse around the first and second decades of the twentieth century, they abandoned its metre, which was based on tone (level or oblique), and which had been used by traditional poets for more than a thousand years. Writing in vernacular Chinese (which is largely polysyllabic) instead of classical Chinese (which is largely monosyllabic), they used a metre based on syllables, each syllable being equivalent to one Chinese character. Thus a poem may consist of lines made up of four, five, six, seven, eight, nine… syllables; lines consisting of less than four syllables were rare. For a detailed discussion of this kind of metre, see Wang, 1979, pp. 822-950. Later, poets found that this syllable-based metre did not work, and that verse written in this kind of metre often sounded unnatural and mechanical. This is because units of rhythm in modern Chinese verse are, most of the time, based on groups of syllables (characters) rather than single syllables (characters), as is the case with classical Chinese verse. Such a difference is due largely to a difference between classical Chinese and vernacular Chinese: whereas characters in classical Chinese are all stressed, characters in vernacular Chinese can be stressed or unstressed. As a result of their new awareness, poets began to experiment with a new kind of metre, which has become the most prevalent metre today among modern Chinese poets who care to tap the resources of Chinese prosody. According to the requirements of this new metre, a line is divided into pauses or, to use the terminology of English prosody, feet, with each foot consisting of one, two, three, or four syllables (characters). Within the same line, there may be one-character feet, two-character feet, or three-character feet; as the four-character foot is often regarded as two two-character feet, modern Chinese metre can practically be considered to be a metre based on three types of feet, which, when exploited by a skilful poet, can yield various rhythmic effects. As the foot is the basic unit of rhythm, in a stanza of four lines, for example, the number of syllables (characters) in each line may vary, but as long as each line consists of the same number of feet, the four lines are considered to be written in the same metre. For a detailed discussion of this kind of metre, see Wen Yiduo’s 聞一多article, “Shi de gelü 詩的格律” ‘The Metre of Poetry’ (Wen, 1985, pp. 81-87). In the following paragraphs, the metre and form of modern Chinese verse will be further discussed in connection with textual analysis.
In quoting the Divina Commedia, I have relied on the Società Dantesca Italiana edition rather than Petrocchi’s edition, despite the fact that, in recent years, the latter has become more popular with scholars. However, in translating the poem into Chinese, which requires close reading and comparison of different source texts, I have found that, generally speaking, the Società Dantesca Italiana edition is superior to the Petrocchi edition in terms of poetic effects. Take, for example, lines 43-44 of Canto 2 of the Purgatorio, which describes the angel from Heaven approaching, on a boat, Virgil and Dante the Pilgrim, the Società Dantesca Italiana edition reads: “Da poppa stava il celestial nocchiero, / tal che parea beato per iscripto” ‘On the poop stood the heavenly steersman, such that blessedness seemed written upon him’ (Sinclair, 1971, Purgatorio, p. 35); the Petrocchi edition reads: “Da poppa stava il celestial nocchiero, /tal che faria beato pur descripto” ‘On the poop stood the heavenly steersman, such that he, even when merely described, would make one blessed’. Judging by the context, it would be more dramatic, more to the purpose to focus at this moment on the angel instead of on the angel’s influence on the beholder; by focusing on the effect of looking, Petrocchi’s reading has less immediacy and is, as a result, less effective in presenting a moment of high drama. The Italian Dante scholar Sapegno (1997, Purgatorio, p. 18) is certainly right when he observes: “Il Petrocchi legge invece faria beato pur descripto; che è meno espressivo” ‘Instead, the Petrocchi edition reads would make one blessed even if merely described, which is less expressive’. Because of this literary consideration, I have voted for the Società Dantesca Italiana edition.
For the convenience of those who do not read Italian, I shall include Mandelbaum’s English version after I have quoted Dante’s original. However, as Mandelbaum’s English version does not always follow the original as closely as Sinclair’s prose version, probably because of the need to satisfy prosodic requirements, to highlight certain semantic units in the original, I shall, whenever necessary, rely on Sinclair’s or my own version for English glosses in this article.
For the convenience of those who do not read Chinese, I have romanized all Chinese quotations in this article. In the main text, a Chinese quotation and its romanized version are accompanied by a gloss in English.
Modern Chinese has four tones: the first tone, the second tone, the third tone, and the fourth tone. If the light tone is also counted, there are altogether five tones. As it is not possible to convey, without a tape-recorder, to non-speakers of Chinese what the five tones sound like, the tones are not indicated in the romanized transcriptions.
Zhu’s version is translated from the Carlyle-Wicksteed English version of 1919, London, J. M. Dent and Sons, which was published again in 1932 and 1950 by Vintage Books, a division of Random House; the Carlyle-Wicksteed quotation in this paper is from the 1950 Vintage Books edition.
Zhu’s translation is based on the following Carlyle-Wicksteed English version:
Carlyle, 1950, p. 25
When he had ended, the dusky champaign trembled so violently, that
the remembrance of my terror bathes me still with sweat.
The tearful ground gave out wind, which flashed forth a crimson light
that conquered all my senses; and I fell, like one who is seized with sleep.
According to Huang Wenjie (Diyupian地獄篇 ‘Inferno’: acknowledgment page; Tiantangpian天堂篇 ‘Paradise’, p. 488), his Chinese version is translated from two editions of the Divina Commedia: Natalino Sapegno (ed.) (1985), La Divina Commedia. Firenze, La Nuova Italia Scandicci; Umberto Bosco and Giovanni Reggio (ed.) (1994), La Divina Commedia. Firenze: Le Monnier.
In modern Chinese verse, when more than three two-character pauses appear in a row, the rhythm will sound unvaried and monotonous, lacking the vigour and force produced by the accurate spacing of one- two- and three-character pauses. By “accurate spacing of one- two- and three-character pauses” is meant the spacing of the three kinds of pauses in keeping with the sense. In the hands of a poet practised in the art of poetry-writing or, for that matter, prose-writing, the rhythm can accelerate and decelerate, moving with a speed that reinforces what the poet wants to express semantically. For a detailed discussion of metre in Chinese verse, which is based on feet made up of one- two- and three-character pauses, see Wen Yiduo’s 聞一多article “Shi de gelü 詩的格律” ‘The Metre of Poetry’ (Wen, 1985, pp. 81-87). In this article, Wen shows how two-character and three-character pauses (feet) are spaced out to give the rhythm vigour and force. In the lines quoted by Wen, three two-character pauses (much less four) never appear in a row. When more than three two-character pauses appear in a row, the same rhythm will keep repeating itself, thus resulting in monotony.
The words “jumble” and “ineffective” are used to describe the syllables for two reasons. First, the metre of the three lines consists largely of two-character feet, such as “baiban / sicun./… / wufa /… / ba ta /… / nage / yuanli/ tanxun /… / cike / miandui /… / jingxiang / yeshi / zheizhong / qingxing /… /wo xiang / kanqing /…/ renxing / ruhe /… 百般 / 思忖 /…/ 無法 / 把他 /… / 那個/ 原理 / 探尋 /… / 此刻 / 面對 / ⋯ / 景象 / 也是 / 這種 / 情形: / 我想 / 看清: / … /人形 / 如何 / … ” Secondly, the rhythm is often repetitive, lacking variation even when non-two-character feet are used, as can be seen in “ye wufa/ ba ta / suo xuyaode/ / 也無法 / 把他 / 所需要的 /” and “/ wo cike / miandui / na xinqide / / 我此刻 / 面對 / 那新奇的”, as well as in the four three-character feet appearing in a row in “na renxing / ruhe yu / na guangquan / xiang shiying / /那人形 / 如何與 / 那光圈 / 相適應 /”. A translator who is more sensitive to rhythm would have avoided this kind of repetition and introduced more variation; as it is, the Chinese translation under discussion gives one the impression that it is made up of two- and three-character feet accidentally thrown together, forming a “jumble” of syllables that fails to suggest the sense of the original effectively.
Footnotes 11 and 12 have already explained how metre in modern Chinese verse works. In connection with the phrases “drag on” and “continues to flag”, further explanation is necessary, so as to avoid giving readers the impression that the observations are subjective. In modern Chinese verse, one foot--whether consisting of one character (one syllable), two characters (two syllables), or three characters (three syllables)--occupies the time-span of one beat; when a line is read aloud, every beat has the same time-value. In terms of rhythm, therefore, a one-character foot, in which one beat is assigned to one syllable, sounds slower than a two-character foot, in which one beat is shared by two syllables, or a three-character foot, in which one beat is shared by three syllables; for the same reason, a two-character foot sounds faster than a one-character foot but slower than a three-character foot, and a three-character foot sounds the fastest, for, in actual reading, the reader has to finish reading three syllables within the time-span of one beat. When too many two-character feet appear in a row, the rhythm tends to slow down, going on in a monotonous sing-song manner with no element of surprise; when the three major types of feet are skilfully arranged, the rhythm of a line or of the whole poem can quicken or slow down in accordance with the meaning. As footnote 12 has shown, in the lines under discussion, there are too many two-character feet appearing in a row, resulting in a slow rhythm which fails to accelerate with the heightening tension signified by the lines on the semantic level. Hence the phrases “drags on” and “continues to flag”. From the above discussion and from the discussion in footnotes 11 and 12, it can be seen that adjectives like “brisk”, “sluggish”, “lingering”, “clogged”, etc., which are used by critics to describe rhythm of various kinds, can all be verified objectively and scientifically.
Lines 140-41 in the original (“se non che la mia mente fu percossa / da un fulgore in che sua voglia venne”) means “had not my mind been smitten by a flash wherein came its wish” (Sinclair, 1971, Paradiso, p. 484), which can be translated into Chinese as “幸虧我的心神獲靈光 火霍 然 / 一擊, 願望就這樣垂手而得”. In the translations by Huang Wenjie (2000) and Wang (1954), the first line is translated as “chufei wode xinling bei yidao shanguang jizhong除非我的心靈被一道閃光擊中” ‘unless my mind is struck by a flash’ and “chufei wode xinling bei na shanguang suo ji除非我的心靈被那閃光所擊” ‘unless my mind is struck by that flash” respectively, thereby deviating from the meaning of the source text.
Wang’s (1954) version is not, of course, without flaws. In my article, “The Translation of Poetry”, I have discussed the weaknesses of Wang’s version in detail.
In re-creating the terza rima alongside the five-pause line in my Chinese translation of the Divina Commedia, I have used what Holmes describes as a “mimetic” verse form (quoted from Crisafulli, 2000, p. 340). While the arguments made by Lefevere about “patronage” (1992, pp. 11-25) and “ideology” (1992, pp. 59-72) are relevant to the texts he discusses in his Translation, Rewriting, and the Manipulation of Literary Fame, my strategies in translating the Divina Commedia into Chinese are mainly dictated by poetic considerations.
The metrical scheme of the two lines is as follows:
— ∪ ∪ | — ∪ ∪ | — — | — ∪ ∪ | — ∪ ∪ | — —
— ∪ ∪ | — — | — ∪ ∪ | — — | — ∪ ∪ | — —
in which “—” stands for a long syllable and “∪” for a short. For a detailed description of Greek prosody and the Greek hexameter, see Pharr, 1985, pp. 29-31; pp. 309-312. Unscanned, that is, with the syllables undivided, the two lines from the Iliad are as follows:
Μῆνιν ἄϵιδϵ, θϵά, Πηληϊάδϵω ᾿Αχιλῆοϛ
οὐλομένην ἣ μυρί᾿ ᾿Αχαιοῖϛ ἄλγϵ᾿ ἔθηκϵν…
Metre in Greek and Latin verse is based on quantity, metre in English verse on stress, metre in classical Chinese verse on tone (level or oblique), and metre in modern Chinese verse on pause (one-character pause, two-character pause, and three-character pause, which, as has been shown earlier, are also described as “one-character foot”, “two-character foot”, and “three-character foot”). However, as I am concerned only with line-length here, I shall not discuss Chinese and Western prosody in detail.
The metrical scheme of the three lines is as follows:
— ∪ ∪ | — ∪ ∪ | — ∪ ∪ | — — | — ∪ ∪ | — ∪
— — | — — | — — | — — | — ∪ ∪ | — —
— ∪ ∪ | — — | — — | — — | — ∪ ∪ | — —
where, in terms of syllable quantity, “—” stands for heavy, and “∪” for light. The metre is hexameter, consisting of six feet, which are either dactyls (— ∪ ∪) or spondees ( — — ). For a detailed description of Latin metre and Virgil’s hexameter, see Jones, 1986, pp. 318-320.
The Divina Commedia is regarded both as an allegory and as an epic. See Cuddon, 1992, pp. 284-293.
One might add Anderson’s translation as well.
In three papers in Chinese, I have discussed how formidable the task is if one decides to use terza rima in a modern Chinese translation of the Divina Commedia. The papers are entitled “Zi tao ku chi: Tan Shenqu yunge de fanyi自討苦吃—談《神曲》韻格的翻譯” ‘You Asked for It: On the Translation of the Rhyme Scheme of the Divina Commedia’; “Zai tan Shenqu yunge de fanyi再談《神曲》韻格的翻譯” ‘A Second Discussion on the Translation of the Rhyme Scheme of the Divina Commedia’; “Bing fen liu lu qin xianyin: Shenqu changju de fanyi 兵分六路擒仙音—《神曲》長句的翻譯” ‘Dispatching Troops in Six Directions to Capture the Music of Heaven: On the Translation of Long Sentences in the Divina Commedia’. The three papers have been collected in my book Yuyan yu fanyi語言與翻譯‘Language and Translation’, published by Chiuko Press in Taiwan in 2001.
Spanish, with most words ending in vowels, can accommodate terza rima even more comfortably than modern Chinese does. For a detailed discussion, see the Introduction to my Chinese translation of the Divina Commedia (Huang Guobin [Laurence Wong], 2003, Shenqu: Diyupian, pp. 17-59).
In modern Chinese, each syllable has one of five tones, including the light tone. As each syllable has a tone, the phonological orchestration in my Chinese translation is more complex than has been represented here, since I have not indicated the tone of each syllable.
In his The Theory and Practice of Translation, co-authored with Charles Taber, Nida (1969, p. 22) emphasizes “the priority of dynamic equivalence over formal correspondence.”
For Lin Shu’s story-retelling as translation, see Wong, 1998, pp. 208-233.
- ALIGHIERI, Dante (1960). Le Opere di Dante. M. Barbi et al., ed. Firenze, Società Dantesca Italiana.
- ANDERSON, Melville B. (1932). The Divine Comedy of Dante Alighieri. By Dante Alighieri. 3 vols. Vol. 1: Inferno; vol. 2: The Purgatorio; vol. 3: The Paradiso. “The World’s Classics” Series. London, Oxford University Press.
- CARLYLE, John Aitken, Thomas OKEY, and P. H. Wicksteed (trans.) (1950 ). The Divine Comedy by Dante Alighieri. New York, Vintage Books.
- CIARDI, John (trans.) (1961-1982). The Divine Comedy. By Dante Alighieri. 3 vols. Vol. 1: The Inferno; vol. 2: The Purgatorio; vol. 3: The Paradiso. New York, Mentor.
- CHIAPELLI, Fredi, ed. (1972). La Divina Commedia. By Dante Alighieri. Milano, U. Mursia & C.
- CRISAFULLI, Edoardo (2000). “Dante. The Divine Comedy”. Encyclopedia of Literary Translation into English. Olive Classe, ed. London, Fitzroy Dearborn Publishers. 2 vols., pp. 339-44.
- CUDDON, J.A. (1992). The Penguin Dictionary of Literary Terms and Literary Theory. London, Penguin Books.
- DURLING, Robert M. (trans.) (1996). The Divine Comedy of Dante Alighieri. By Dante Alighieri. Vol. 1: Inferno. New York, Oxford University Press.
- ELIOT, T. S. (1957). On Poetry and Poets. London, Faber and Faber.
- FAIRCLOUGH, H. Rushton (trans.) (1999). Virgil. Eclogues. Georgics. Aeneid I-VI. Cambridge, Massachusetts, Harvard University.
- HUANG, Guobin 黃國彬 [Laurence Wong] (trans.) (2003). Shenqu 神曲 ‘La Divina Commedia’. 3 vols. By Dante Alighieri. Vol. 1: Diyupian 地獄篇 ‘Inferno’; vol. 2: Lianyupian 煉獄篇 ‘Purgatorio’; vol. 3: Tiantangpian 天堂篇 ‘Paradiso’. Taipei, Chiuko Press.
- HUANG, Wenjie 黃文捷 (trans.) (2000). Shenqu神曲 ‘La Divina Commedia’. By Dante Alighieri. 3 vols. Vol. 1: Diyupian 地獄篇 ‘Inferno’ ; vol. 2: Lianyupian 煉獄篇 ‘Purgatorio’; vol. 3: Tiantangpian 天堂篇 ‘Paradiso’. Guangzhou, Huacheng Chubanshe.
- JONES, Peter V., and Keith C. Sidwell (1986). Reading Latin: Grammar, Vocabulary, and Exercises. Cambridge, Cambridge University Press.
- LEFEVERE, André (1992). Translation, Rewriting, and the Manipulation of Fame. London, Routledge.
- MANDELBAUM, Allen (trans.) (1980-1984). Dante Alighieri. The Divine Comedy of Dante Alighieri: A Verse Translation. 3 vols. 1: Inferno; 2: Purgatorio; 3: Paradiso. New York, Bantam Books.
- MILTON, John (1966). Poetical Works. Douglas Bush, ed. London, Oxford University Press.
- MURRAY, A. T. (trans.) (1924). Homer. The Iliad. Vol. 1. Cambridge, Massachusetts, Harvard University Press. 2 vols.
- MUSA, Mark (trans.) (1986). The Divine Comedy. By Dante Alighieri. 3 vols. Vol. 1: Inferno; vol. 2: Purgatory; vol. 3: Paradise. New York, Penguin Books.
- NIDA, Eugene, and Charles R. Taber (1969). The Theory and Practice of Translation. Leiden, E. J. Brill.
- PHARR, Clyde. (1985). Homeric Greek: A Book for Beginners. Revised by John Wright. Norman, University of Oklahoma Press.
- SAPEGNO, Natalino, a cura di (1997). Dante Alighieri. La Divina Commedia. 3 vols. 1: Inferno; 2: Purgatorio; 3: Paradiso. Firenze, La Nuova Italia.
- SAYERS, Dorothy, and Barbara Reynolds (trans.) (1949-62). The Comedy of Dante Alighieri the Florentine. By Dante Alighieri. 3 vols. Vol. 1: Hell; vol. 2: Purgatory; vol. 3: Paradise. Harmondsworth, Penguin Books.
- SINCLAIR, John D. (trans.) (1971). The Divine Comedy of Dante Alighieri. By Dante Alighieri. 3 vols. Vol. 1: Inferno; vol. 2: Purgatorio; vol. 3: Paradiso. London, Oxford University Press.
- SISSON, C. H. (1980). The Divine Comedy. By Dante Alighieri. Oxford, Oxford University Press.
- TIAN, Dewang 田德望 (trans.) (1997). Shenqu 神曲 ‘La Divina Commedia’. By Dante Alighieri. 3 vols. Vol. 1: Diyupian 地獄篇 ‘Inferno’ ; vol. 2: Lianyupian 煉獄篇 ‘Purgatorio’; vol. 3: Tianguopian 天國篇 ‘Paradiso’. Beijing, Renmin Wenxue Chubanshe.
- WANG, Li 王力 (1979). Hanyu shilü xue 漢語詩律學 ‘Chinese Prosody’. 1st ed. 1958. Shanghai, Shanghai Jiaoyu Chubanshe.
- WANG, Weike王維克 (trans.) (1954). Shenqu 神曲 ‘La Divina Commedia’. By Dante Alighieri. Beijing, Renmin Wenxue Chubanshe.
- WEN, Yiduo 聞一多 (1985). Wen Yiduo lun xinshi聞一多論新詩 ‘Wen Yiduo on New Poetry’. Research Unit for Wen Yiduo Studies, ed. Wuhan University. Wuhan, Wuhan University Press.
- WONG, Laurence 黃國彬 (1997). “The Translation of Poetry”. Translation Quarterly, 3-4, pp. 1-40.
- WONG, Laurence 黃國彬 (1998). “Lin Shu’s Story-retelling as Shown in His Chinese Translation of La Dame aux Camélias”. Babel, 44, 3, pp. 208-233.
- WU, Jingrong 吳景榮 et al., ed. (1983). The Pinyin Chinese-English Dictionary. Hong Kong, Commercial Press.
- ZAPPULLA, Elio (trans.) (1998). Inferno. By Dante Alighieri. New York, Pantheon Books.
- ZHU, Weiji 朱維基 (trans. from the 1919 English Carlyle-Wicksteed version) (1984). Shenqu 神曲. By Dante Alighieri. 3 vols. Vol. 1: Diyupian 地獄篇 ‘Inferno’; vol. 2: Lianyupian 煉獄篇 ‘Purgatorio’; vol. 3: Tiantangpian 天堂篇 ‘Paradiso’. Shanghai, Shanghai Yiwen Chubanshe.