For instance, in the second chapter (the dance on the lawn in front of the castle) there are at least sixteen such lines, including one hendecasyllable (J’étais le seul garçon dans cette ronde), alexandrines (like Je ne pus m’empêcher / de lui presser la main) and various hemistichs (like La belle devait chanter, or Les longs anneaux roulés). In the twelfth chapter, we find Je jugeais que j’étais / perdu dans son esprit.
I wanted to respect those rhythms that gave the text its dreamy charm, but I felt involved in a process of losses and compensations, since Italian did not always allow me to produce a line of verse in the same position in the syntactic string as Nerval’s. Thus, when I lost the possibility of reproducing a verse where Nerval put it, I did my best to invent a verse immediately after, or before, to salvage the general effect, and to get an equivalent number of verses for every paragraph.
Let me quote only three examples, comparing the English translations of Halévy, Aldington and Sieburth. In the following quotations I shall put these lines in bold type to make them evident.
In chapter 3, when the Narrator (half asleep) evokes the vision of Adrienne we find:
Fantôme rose et blond / glissant sur l’herbe verte, à demi baignée de blanches vapeurs.
One can miss every opportunity to save the rhythm, as Halévy did with his A rosy and blond phantom gliding over the green grass that lay buried in white vapour. One can miss the first verse and recuperate later, as Aldington did: A rose and gold phantom gliding over the green grass, / half bathed in white mists. Or one can give the translation an additional verse, as I did (with a double line of six syllables) and Sieburth also:
Eco: Fantasma rosa e biondo / lambente l’erba verde, / appena bagnata / di bianchi vapori.
Sieburth: A phantom fair and rosy / gliding over the green grass / half bathed in white mist.
They are not two cases of excessive generosity. Translators are frequently aware of having missed a chance or being on the verge of missing one, and thus try to recover their losses.
In chapter 3 we read:
Aimer une religieuse / sous la forme d’une actrice! . . . / et si c’était la même? – Il y a de quoi devenir fou! c’est un entrainement fatal où l’inconnu vous attire comme le feu follet – fuyant sur les joncs d’une eau morte.
There are two alexandrines and two hemistichs. Halévy does not seem to have been sensitive to the rhythm, since he inserts lines, I think, at random, and not at strategic points. Probably his lines exemplify the hazards of a literal translation:
To love a nun in the form of an actress! – and suppose it was one and the same! It was enough to drive one mad! It is a fatal attraction when the Unknown leads you on, like the will-o’-the-wisp that hovers over the rushes of a standing pool.
Aldington does not make any effort and his only hemistich is due to the fact – as for the other translators – that English probably has nothing better for feu follet:
To love a nun in the shape of an actress . . . and suppose it was the same woman? It is maddening! It is fatal fascination where the unknown attracts you like the will-o’-the-wisp moving over the reeds of still water.
Sieburth, however, capitalises with two alexandrines and two hemistichs:
To be in love with a nun / in the guise of an actress! . . . and what if they were one and the same! It is enough to drive one mad – the fatal lure of the unknown drawing one ever onward / like a will o’ the wisp / flitting over the rushes of a stagnant pool.
I went further on, finding three complete alexandrines:
Amare una religiosa sotto le spoglie d’una attrice! . . . e se fosse la stessa? / C’è da perderne il senno! / è un vortice fatale / a cui vi trae l’ignoto, / fuoco fatuo che fugge / su giunchi d’acqua morta . . .
Perhaps I did too much, but I loved this ‘singing’ feel.
But Sieburth took his revenge in chapter 14, where we find these splendid closing lines:
Telles sont les chimères / qui charment et égarent / au matin de la vie. J’ai essayé de les fixer sans beaucoup d’ordre, mais bien des cæurs me comprendront. Les illusions tombent l’une après l’autre, / comme les écorces d’un fruit, et le fruit, c’est l’experience. Sa saveur est amère: elle a pourtant quelque chose d’âcre qui fortifie.
Nerval conceals in the text two alexandrines, a hendecasyllable and two more hemistichs. Translations follow, and Sieburth does his best to put the verses more or less where Nerval put them. This time his performance is better than mine. Chapeau.
Such are the charms that fascinated and beguile us / in the morning of life. I have tried to depict them without much order, but many hearts will understand me. Illusions fall, like leaves, one after another, and the kernel that is left when they are stripped off is experience. The taste is bitter, but it has an acid flavor that acts as a tonic. (Halévy)
Such are the delusions which charm and lead us astray in the morning of life. / I have tried to set them down in no particular order, but there are many hearts / which will understand me. Illusions fall one by one, like the husks of a fruit, / and the fruit is experience. Its taste is bitter, yet there is something sharp about it which is tonic. (Aldington)
Such are the chimeras / that beguile and misguide us / in the morning of life. / I have tried to set them down without much order, but many hearts will understand me. Illusions fall away one after another like the husks of a fruit, / and that fruit is experience. It is bitter to the taste, but there is fortitude to be found in gall. (Sieburth)
Tali son le chimere / che ammaliano e sconvolgono / all’alba della vita. Ho cercato di fissarle senza badare all’ordine, ma molti cuori mi comprenderano. Le illusioni cadono l’una dopo l’altra, come scorze d’un frutto, / e il frutto è l’esperienza. Il suo sapore è amaro; e tuttavia esso ha qualcosa di aspro che tonifica. (Eco)
Which translation is more faithful to Nerval’s intention? I suggest the reader speak them aloud and decide which one ‘sings’ better – but reject those where the lines are too perceptible (I repeat: the verses should seduce the ear in a subliminal way without being immediately detectable).
Substance in poetic translations
A translation must preserve the textual rhythm. But that is not enough. As I have said in the previous chapters, many substances of expression are displayed on the expression plane and many of them are not specifically linguistic (such as metre and many phonosymbolic values). Metre is so independent from the structure of a given language that the same metric scheme can migrate from one language to another.
One of my favourite of Montale ’s poems is:
Addio, fischi nel buio, cenni, tosse
e sportelli abbassati: È l’ora. Forse
gli automi hanno ragione: Come appaiono
dai corridoi, murati! . . .
– Presti anche tu alla fioca
litania del tuo rapido quest’orrida
e fedele cadenza di carioca?
There are two translations of it, one in English by Katherine Jackson1 and the other in French by Patrice Dyerval Angelini:2
Goodbyes, whistles in the dark, nods, coughing,
and train windows down. It’s time. Perhaps
the robots are right. How they lean
from the corridors, walled in!
And do you too lend, to the dim
litany of your express train, this constant
fearful cadenza of a carioca? (Jackson)
Adieux, sifflets dans l’ombre, signes, toux
Et vitres fermées. C’est l’heure. Peut-être
Les automates ont-ils raison. Comme des couloirs
Il apparissent murés! . . .
Toi aussi, prêtes-tu à la sourde
Litanie de ton rapide cette affreuse
Et fidèle cadence de carioca? (Angelini)
A short comparison with the original tells us that in both translations the metre is not respected. Now metre is fundamental in this poem because (especially in the last tercet) the language reproduces both the rhythm of the train and the rhythm of the dance (carioca). I do not think that it is impossible to reproduce such a double rhythm in another language and I apologise for feeling obliged to quote a personal experience. I played many times with lipograms (a technique by which a given text is ‘retold’ once without using A, then without using E and so on – in fact one can make lipograms, as the old Greek poets did, for every letter of the alphabet). I also made monovocalic or one-vowel poems (in fact a monovocalic poem is a lipogram that avoids four vowels at the same time). Thus I tried to reproduce Montale’s poem first as lipograms avoiding A, E, I, O and U, then as five monovocalic poems.3
My problem was certainly not to ‘translate’ the poem, but to save, with my pseudo-paraphrase, the real thing. And the real (poetic) thing was that in the poem there are five hendecasyllables, of which two are proparoxitone, and two seven-syllable verses; the first four verses are without rhyme and the last
