The Penguin Book of French Poetry Read online

  There are many tightropes to be walked here. Experience has taught me that only the pragmatic approach works, for the conditions surrounding the editorial balancing act alter from poet to poet and period to period. Instinct and personal predilection must come into play, as well as an awareness of the judgements of posterity; enthusiasm and objectivity have to find a fresh working relationship in each individual case. To take as an example the question of representing all phases of a poet’s career: only a masochist (or a dedicated researcher) would want to read now the verse which Lamartine or Musset trundled out long after the emotional traumas that produced their best work, or share Verlaine’s descent into self-parodying mediocrity; it seems essential on the other hand to perceive the linear evolution of Laforgue or Rimbaud, and the imaginative breadth of Baudelaire or Apollinaire.

  Many fine poets, of course, have had to be excluded. The names of a good number of them are mentioned in the chapters on Romanticism, on the Parnassians, on Symbolism, on Cubism, and on the Surrealists. Choices and compromises have to be made: Lautréamont, for example, must do duty for the whole satnic, self-destructive fringe of Romanticism; Anna de Noailles and Catherine Pozzi, briefly, for the relative abundance of women poets in the early decades of this century; Cendrars for the cosmopolitan spirit of compulsive travelling (to the exclusion of Larbaud, a dilettante by comparison with Cendrars, whose work blazes like his name with love, anguish, irony, originality of perception and of technique). That dated and clumsy optimist of the age of technology, Jules Romains, is similarly eclipsed by the fire of Verhaeren’s vision. And no doubt certain inclusions will be controversial, but sometimes it is important to include a poet scorned by some of his successors, yet whose work nevertheless is symptomatic of something interesting in his times.

  In this comprehensive panorama of French poetry through a 130-year period, it has also been impossible to include some poets who had certainly established a reputation by 1950. This is the date at which I have drawn the line, allowing readers to see the emergence of Michaux, Ponge and Char as major and innovative post-war poets (1945, originally seen as an end-point, would have been less satisfactory, for it would have interrupted the linear development of poetry at a fairly low point in qualitative terms, however passionately felt much Resistance poetry undoubtedly is). There is, sadly, no room for Queneau, Follain, Tardieu, Guillevic, La Tour du Pin, Emmanuel, Albert-Birot, Bousquet, Mac Orlan, Daumal, Fombeure and others, in addition to those mentioned in my presentational chapters.

  And of course this anthology has to stop short of an impressive list of fine contemporary poets who first captured major attention in the 1950s and ’60s: Bonnefoy, Bosquet, Dadelsen, Deguy, du Bouchet, Dupin, Gaspar, Jabès, Jacottet, Mansour, Pleynet, Réda, Renard, Roubaud and others, whose work I hope readers will explore for themselves.1 For some of them, poetry remains a means, an existential or spiritual quest; for others, it is an end in itself, a self-referential, Mallarmé-haunted, self-analytical play of language.

  To avoid distorting the historical perception of the state of poetic composition, it also seemed logical to draw that 1950 line right across the board, even for those contemporary poets who have been included (the point is stretched only very slightly for Césaire’s revised version of Ode à la Guinée, and for Char’s L’inoffensif and Front de la rose).

  Much has been written on the theory of translation, and I do not propose to add anything weighty to it. It is important, nevertheless, to comment on the practical strategy used here. The main purpose is to provide access to meaning by removing semantic barriers, and thus to accelerate the penetration of the poem itself by the non-French reader. My translations are, I hope, neither woodenly literal nor pretentiously literary. They seek neither to efface themselves nor to blur criteria by competing with the original for attention. Above all, they are intended as a service, and to excite interest in the original as a creative event.

  The prose format implies a discipline in the rendering of sense which inhibits the verse-to-verse translator rather less, particularly if he is reshaping the original impulse into a new metrical and rhyming structure (reading of existing work in this field has convinced me that only a small number of gifted individuals can do this without producing distortion or bathos). Even so, if my translations are successful they are a step above the literal. They have decided rhythms and sound-patterns that please the ear, or at least do not displease it. They may occasionally even have a certain equivalent musicality. At times they interpret cautiously (Fargue’s ‘le trottoir tout gras de bouges’ is rendered as ‘the pavement swilling with brothels’); and in places they modify syntax and grammar or translate one word by two (Baudelaire’s smoking log sings ‘in shrill discord’ for ‘en fausset’, with its falsetto and out-of-tune connotations), or two words by one. Every small liberty thus taken has emerged as the best solution after thorough componential analysis and consideration of more literal options, and I do not think any major liberties have been taken. I have tried at all times to balance my responsibilities to the reader with those I have to the poet, to find an intuitive blend of communicative and formal levels of equivalence, for a poem, as Claudel pointed out, is not just a bag of words.

  As with selection policy, pragmatism dictates the outcome. Only with Mallarmé does a mainly literal approach often seem the most productive, and even there an occasional communicative adjustment is called for. In Petit Air I, for example, I have judged it right to add a word to clarify ‘Au regard que j’abdiquai / Ici’ as ‘in the gaze which I withdrew Down here’. In the same poem, despite the third-person endings of the verbs ‘longe’ and ‘plonge’, some translators have settled on ‘toi’ as their subject. If, as I believe, ‘ta jubilation nue’ is the subject, then ‘devenue’ qualifies ‘onde’. Thus ‘your naked jubilation’ in English must come earlier in the sentence structure, and the waters subsequently mentioned ‘(have) become you’ rather than any question of ‘you having become your naked jubilation’.

  It is of course much harder to take account of the semantic potential of rhythm and sound-patterns, but the effort must be made whenever possible. In Jammes’ Il va neiger, for example, a subtle change of line-order at the end produces a soft, serene, feminine ending that contrasts with the clipped aggression in the opening quatrain, and in itself expresses the personal therapy that the poet has performed. It would be wrong to translate ‘Qu’est-ce?’ by the crisp ‘t’ sounds, for example, of ‘What is it?’ It may be possible to capture at least intermittently the railway-train rhythms of Cendrars or the kaleidoscopic effects of Apollinaire and Reverdy, and the supple rhythmic plasticity of the best Parnassian poetry, such as the movement of our eyes down the creepers or the progress of the animal through the jungle in Leconte de Lisle’s Le Rêve du Jaguar. But when an experimental metre is itself part of the message, as it is in Je ne sais pourquoi… by Verlaine or in Apollinaire’s La jolie rousse, the translator is probably defeated – as he is, in a different way, by Hugo’s Demain, dès l’aube… In this beautifully understated poem, grief is projected concretely on to a lonely figure walking compulsively and rhythmically along a road. No close translation could do justice to the reinforcement of the dominant future tense by an insistent sound-pattern (the ‘ai’ sound, in various spellings, recurs fourteen times with further supportive assonance in a 12-line poem), and to the internal rhymes at rhythmically regular points. The sense that a destiny is operating here comes as much through musical versification as through images.

  Ambiguities of sense have also been assessed individually: in some cases a happy equivalent can be found in English, in some I have expanded a word or phrase slightly to incorporate the ambiguity, in some I have resorted to an explanatory footnote, and in some I have simply decided to opt for one sense. Once again, the method here is pragmatic, and I hope renderings will be judged on their individual merits.

  Except when absolutely necessary, I have not taken liberties with punctuation where the original poet has used it. For th
e translator into prose, however, a difficulty arises with the invention of free verse, and above all with the discovery by Cendrars, Apollinaire, Aragon and many others that a poem’s spatial presence on the page can be enhanced and its resonance greatly expanded through the abolition of formal punctuation, and through a reliance on the internal dynamics of rhythm and pause to shape the phrases. It seems to me that translations which supply punctuation of their own for such material are intrusive, presumptuous and disrespectful of authorial intention. A partially satisfactory alternative is to use as a marker the capital letter with which many poets continue to begin each line, though of course proper names and ‘I’ cause a small degree of confusion that cannot be helped. Where the capital letter has also been abandoned, the oblique stroke has to serve as divider. These are, I hope, acceptable conventions.

  Almost all drafts of translations could be reworked for ever, unlike finished poems. Those privileged moments when the translator feels entirely satisfied with a piece of work are rare, and are more than counterbalanced by the usual awareness that, like Gautier, he is chipping away doggedly at a resistant marble block, gaining intermittent fulfilment in pursuit of something ideal. But decisions have to be made, final versions have to be typed to meet the deadline, though he would love to go on moulding them ad infinitum. As Ernest Hemingway pointed out, the typewriter sometimes fixes words before they’re ready to be fixed, but if these translations are offered now as a vehicle to carry the reader into the marvellous domain of French poetry, I hope that the journey will not be an uncomfortable one.

  W. H. Rees

  TECHNICALITIES

  Invented in the twelfth century and named after Alexandre de Bernay, one of its first exponents, the 12-syllable Alexandrine verse line came to dominate French poetry from the middle of the seventeenth century, and classically educated Frenchmen have it in their bloodstream still today. Indeed, despite the proliferation of other metrical forms and the breakdown of regular structure into vers libres and the Claudelian verset at the end of the nineteenth century, the Alexandrine has a remarkably tenacious hold on French poets. It often appears still, sometimes openly, sometimes in semi-disguise within the rhythms that give modern verse composition and prose-poetry their shape and dynamics.

  In its classical form, the Alexandrine is end-stopped by punctuation, and divided in half by a caesura, which is a pause both formal and natural, whether punctuated or not. Each half is called a hemistich. This line is an ideal expression of intellectual balance, symmetry and wholeness, of thesis, antithesis and implied synthesis. In the hands of Racine, probably its greatest exponent, it is the perfect form for the expression of tragic dilemma. Established in his time as the prescriptive model, it became in the eighteenth century (a bleak period in the history of French poetry) a hollow shell, which the Romantics were to crack, if not break, in the post-Napoleonic period. They experimented with new positions for the caesura and were particularly successful with their more musical Alexandrin trimètre, or threefold division of the line, as in: ‘Je marcherai les yeux fixés sur mes pensées’ (Hugo). Nevertheless, Hugo continued to make memorable use of the Alexandrin classique for balanced antithesis, often incorporating syntactic inversion or chiasmus: ‘Ces murs maudits par Dieu, par Satan profanés’.

  There have always been lines of fewer syllables than 12, of course, but rarely more than 12 until Laforgue effectively invented free verse, and rarely an uneven number of syllables. The haunting, lyrical Impair line of 5, 7, 9 or 11 units, extolled and demonstrated beautifully by Verlaine above all, suggests to the French intellect and instinct even today a frustrating lack of shape, balance, precision and finality, and it is interesting that the most fastidious and self-conscious perfectionists among the poets, Mallarmé and Valéry, rarely used it.

  Lamartine brings to the Alexandrine some fluidity and elegiac musicality, with soft consonants and mellow assonance, gently modifying old conventions. In line two of Le Lac, for example, he makes a very smooth elision immediately after the central caesura:

  Dans la nuit éternelle emportés sans retour

  This sliding of the last vowel of ‘éternelle’ into the first vowel of ‘emportés’ effectively dilutes the stress that would normally come at the sixth syllable, before the pause, and creates a mellifluous 12-syllable unit to be spoken in one breath. He follows it with an 18-syllable unit that brings added force and poignancy to his metaphor:

  Ne pourrons-nous jamais sur l’océan des ages

  Jeter l’ancre un seul jour?

  To create an effect of emotion welling up, he is using here the technique of enjambement (‘striding over’ or ‘encroaching’). Hugo and the other Romantics, and then the Parnassians, were to find in this a major instrument of rhythmic and metrical flexibility. There are two types of enjambement, which is the overflowing of a phrase from one line to the next in a suppression of end-stopping punctuation. Its effect is often to delay or hasten stress, and to attenuate the force of rhyme.

  (a) Enjambement with a rejet: where the extra element is in the line following the main bulk of the whole unit, as in:

  En un creux du bois sombre interdit au soleil

  Il s’affaisse,…

  (Leconte de Lisle)

  Here the poet represents rhythmically the sinking down of a jaguar at the end of its journey back to its lair.

  (b) Enjambement with a contre-rejet: where the extra element is in the line preceding the main bulk of the whole unit, as in:

  Le jeune Cellini, sans rien voir, ciselait

  Le combat des Titans au pommeau d’une dague.

  (Heredia)

  Or in:

  Pour n’être pas changés en bêtes, ils s’enivrent

  D’espace et de lumière et de cieux embrasés.

  (Baudelaire)

  Musset brings a more intimate and colloquial tone to the 12-syllable line, but Nerval constructs his evocative private mythology using alexandrins classiques almost exclusively. Hugo experiments with other metrical forms, but the Alexandrine reimposes itself continually on his work. Fenêtres ouvertes is an attempt by him at literary Impressionism but, despite the apparent fragmentation, it is revealed, when scanned, to be in Alexandrines. The same will apply, very much more powerfully, to Aragon’s magnificent 1940 poem, Les lilas et les roses, which filters the poet’s overwhelming emotion at the fall of France through a ‘whirlwind’ of surreal visual correlatives, in a superficially chaotic, unpunctuated, torrential stream of consciousness that ends arbitrarily. The rhythm breathes in and out as emotion surges in excitement or panic or subsides in unhappy peace. But not only does the poem rhyme, it is also written in Alexandrines and most of them are classical. Thus the poem is what it celebrates and pledges to continue: the spirit and historical continuity of French culture.

  Baudelaire uses a variety of forms, but his finest talent probably lies in the sonnet in Alexandrines. In his hands it is a vibrant object, bursting with compressed imaginative and emotional energy, with which he works a sensuous magic and gives to the poem the status of an event. The best Parnassian poetry continues to operate within the Alexandrine, enhancing its three-dimensional concrete images with supple rhythms. In Le Rêve du Jaguar, again, Leconte de Lisle makes the jungle creepers hang, suspended on the isolated conjunction ‘et’, before their 9-syllable spiralling plunge. The jaguar’s motion is given majesty by the isolated two-stress phrase ‘Il va’, before the powerfully rhythmical and internally rhyming 10-syllable unit, ‘frottant ses reins musculeux qu’il bossue’, which is almost English in its stress pattern.

  It is important for the inexperienced reader to grasp the essential facts that French scansion is a matter of syllable-counting, and that stress in the English sense is very intermittent and very much attenuated in French verse. There are points in the line where a syllable will certainly be accented: in the Alexandrin classique these are the 6th and 12th, in such a way that the final word of each hemistich is likely to be a key element in what is being communicated. In th
e Alexandrin trimètre the 4th, 8th and 12th syllables will be most heavily accented (in so far as stress in French verse could ever be described as ‘heavy’), though the displaced caesuras are weaker than their classical counterparts.

  It is in a sense more ‘monotonous’ than English poetry, lacking a thumping regular dynamic, but it achieves tonal variety through subtle action of sound-patterns (mainly assonance) and subtle interplay through pitch, tempo and syntactic position of the subordinate stresses that certainly do exist and do bring into relief other syllables in the line.

  In the decasyllabic line (as used, for example, by Valéry in Le Cimetière marin), the caesura comes most often after the 4th syllable, though a 6/4 division also occurs, and there are a few poems with a 5/5 division. The octosyllable usually divides 4/4 if it divides at all, and in some of Mallarmé’s most enigmatic verse the caesura could be at any one of several points in the line.

  In the absence of a strong metrical pulse, it is perhaps not surprising that in the evolution of poetry the French have found it harder than the English to dispense with rhyme, a strong unifying factor.

  It is also essential to know that a mute ‘e’ in an orthodox line of French verse is counted as a syllable, except: