Corps de l’article

1. Introduction

William Shakespeare, as a literary icon, arrived in Brazil during the 19th century by way of various adaptations of his plays from French, Spanish and European Portuguese translations. As English was known only by a restricted group of people, this lack of knowledge of the language helped build an aura of non-accessibility and erudition around the Bard’s works.

The first Brazilian Portuguese translations of some of Shakespeare’s plays appeared during the first half of the 20th century, such as Carlos Alberto Nunes’s translations of all the Shakespearian theatre. According to Gomes (1961: 31),[1] “a curious fact, but logical and understandable nevertheless, is that in Brazil it is the poets that have translated Shakespeare most often.” This statement reveals the common belief among translators and critics that only poets can properly transpose a poetic text into another language.

Shakespeare’s image in Brazilian territory has been constantly associated with nobility and refinement. This is because of the widespread notion that Shakespeare wrote to an ideal, scholarly Elizabethan public, instead of a diversified audience. Therefore, by placing the Bard and his home country in a superior cultural paradigm in relation to Brazil, contemporary Brazilian critics and readers will see themselves as privileged members of an elitist culture. This has strongly influenced Brazilian translators’ conceptions on the subject, which have in general produced scholarly and complex translations, and thus have excluded a large segment of readers (Vollet 1998).

Such Shakespeare plays as Hamlet, Macbeth and Romeo and Juliet have had a greater impact in Brazil than the Bard’s Sonnets, due mainly to stage and cinematographic productions based on his dramatic works. Originally published in England in 1609[2], the Sonnets were translated for the first time into Brazilian Portuguese during the 1950s and, since then, have been sporadically re-translated. These later translations indicate that the controversial nature of this poetic sequence – the poet’s love for a young man and his scorn for a dark lady who in no way resembles the sacred object of courtly love – may have compromised its public reception in Brazil in the early years of the 20th century. This attitude has left the Sonnets as a relatively neglected piece of work, on the part of Brazilian translators and literary critics alike.

In this article, I intend to identify the conceptions of translation, authority, and fidelity of three Brazilian translators of the Sonnets, Péricles E. da Silva Ramos, Oscar Mendes, and Jorge Wanderley, as well as briefly analyze their own versions of Sonnet 55. Each of their translations shows specific interpretations and reveals an ideology, even though their introductory notes all refer to a traditional idea of translation as an inevitable process of loss and compensation. Such traditional concepts have been linked by post-structuralist thinkers as Jacques Derrida to the notion of logocentrism. According to Derrida (1995), logocentrism, within traditional western philosophy, underlies man’s search for unique Truth through Reason, and this search for Truth influences conceptions of the subject, language, and translation.

Logocentrism’s search for the Truth, seen as existing independently of any subject, “its obsession with logic and rationality and its need to reject everything subjective, contingent, and context-bound” (Arrojo 1992: 74) forms the basis for an ahistorical notion of translation. This notion implies an obligation to transport form and content from one language to another without the least interference from the translator, whose task is mechanical and non-creative. This aim of complete equivalence demands, apart from identity between meanings, a balance between the values and concepts inherent in each language involved in the translation process. According to logocentric philosophy, the translator should never attempt to outshine the author – the true focus of a literary tradition –, but instead to carry out his task, a necessary evil, as effectively and invisibly as possible.

However, post-structuralist thought has sought to redefine and reflect upon such long-standing notions of language and translation, especially in the context of deconstruction. Jacques Derrida, “by denying the existence of Truth, Origin and Center” (Koskinen 1994: 446), also denies the possibility of semantic transport between languages, believing instead in a reorganization, a play of meanings linked to context and subject to space and time factors. According to Derrida (1995: 232), “in the absence of center or of origin, everything turns into discourse […] that is, a system in which the central, original or transcendental meaning is never absolutely present outside of a system of differences.” Such a notion calls into question the actual status of the original text, and the underlying idea that its equally original meanings – devised by the author’s genius – serves as an unquestionable basis for the translator’s work.

Therefore, the relation between the translated text and the original is redefined in terms of an intertextuality that presupposes mutual debt. In this new light, a translation helps the original to enjoy renewed life in later generations, and the never-ceasing production of meanings justifies the notion of the translation as a new text, a reading offered by a sociohistorical and cultural subject who is part of a given community. In Derrida’s own words:

[…] the translator must assure the survival, which is to say the growth, of the original. Translation augments and modifies the original, which, insofar as it is living on, never ceases to be transformed and to grow. It modifies the original even as it also modifies the translating language. This process – transforming the original as well as the translation – is the translation contract between the original and the translating text.

Derrida 1988: 122

Although this new status of translation has now been widely studied in scholarly research, the old, logocentric assumption still dictates many contemporary analyses of translations and their effects. This is especially true in the case of poetry translations, in which the combination of sound and meaning is considered solely responsible for the poeticity of a text. This is important for the present analysis, as the idea of a translation as a never-ending process of loss and compensation is still extremely vivid. This notion serves as a background for the translators’ own comments on modifications and insertions.

Inevitably, this analysis is also of a comparative nature. However, such comparison does not aim to detect equivalencies with the original Shakespearean text nor to approve or to condemn the translated text. The aim of this study is to verify how the three translators in question shape their interpretations of Sonnet 55 and, through their choices, reveal underlying ideologies. Calling attention to certain linguistic and literary details will aim to identify specific aspects of the translations which point toward historical, linguistic, and ideological factors which influenced the translators during their production.

2. Theoretical positions of the Sonnets’ translators

Péricles E. da Silva Ramos (1919-1992) first produced his translation of 33 sonnets in 1953,[3] and this is probably the most famous translation of the Sonnets in Brazil, given his wide literary fame as a poet. Ramos was considered, during his time, “one of the greatest, if not the greatest, specialist in Verse Theory in our country” (Chociay 1974: x). In all the editions of his translation, Ramos explicitly explains his metrical and rhyme choices in a long introduction in which he also includes, as well as an extensive bibliographical research on the Sonnets’ critical literature, his conceptions on poetry translation. Ramos uses the same rhyming scheme as the English text (abab-cdcd-efef-gg) and chooses the iambic hexameter or a rhyming verse of 12 syllables so as not to “sacrifice much of the text” (Ramos 1970: 12) in relation to the meaning, since, according to the poet, there are many monosyllabic words in the English language which would not fit into Portuguese in a shorter verse, such as the ten-syllable verse.

Therefore, Ramos believes that certain texts, especially some Shakespearean sonnets, require the translator to favour a given interpretation, upon which to build his/her text, since “to translate is, above all, to understand; […] recreate, embark on an adventure of comprehension and re-expression of a given text” (Ramos 1970: 9). By arguing for the absence or modification of structural and semantic elements in Shakespeare’s text, Ramos states the non-existence of faithful poetry translations and his belief that there are only re-creations, a position which frees him from the guilt he feels when effecting changes to the original text. This is clearly visible in the following excerpt:

[…] it is not possible to talk, if ever, of a translation. There are no completely faithful translations in poetry. There are, I repeat, re-creations from one language to another, all of which, quite often notable, will not, however, preserve all the values of the text. Poetry, therefore, is not only meaning, but meaning linked with rhythm, sonority, talent in the use of words and many other unpreservable factors at one time.

Ramos 1970: 15

This logocentric belief reduces the translation act to a process of additions, losses, and compensation of values on the part of a subject who, from the beginning, finds himself beaten by the enormous task of transporting form and content in untouchable containers. It can be equally observed in Oscar Mendes, in his introductory notes to his translation of the Sonnets. Mendes (1902-1983) was a famous Brazilian translator who produced translations of all of Shakespeare’s lyrical work, including the lyrical parts in the Bard’s plays (Mendes 1969), and who published his translation of the Sonnets in 1969.[4]

The apology to the reader for changes and adaptations in his (re)reading precedes the structural observations, when Mendes states “I owe the readers an explanation in terms of the discrepancies in meter and the absence of rhymes in relation to the original” (Mendes 1969: 737. The translator chooses to use the classic alexandrine verse (12 syllables) to ensure the preservation of Shakespearian images and concepts. However, Mendes does not present a rhyming scheme, as in his opinion the sonorous and musical effects are peculiar to each language and “the search for rhymes would lead to modifications […] which would disfigure the poet’s thought” (Mendes 1969: 737). Furthermore, it is the status of a translator and not of a poet which forces Mendes’ purpose to “only be faithful, as far as my understanding goes, to the author’s thought” (Mendes 1969: 737). It is clear once more that poetic license would not be under the constraints of a fidelity contract normally imposed on the regular translator.

The translator Jorge Wanderley[5] (1938-1999), a neurologist turned translator of poetry and prose alike (translations which made him famous in Brazil), in his turn, focuses on priorities which differ significantly from those of Ramos and Mendes. By stating that translation “is an act of force” which needs to flow like any appropriate text, “in such a way that its tense muscle does not show in the smoothness of the dance” (Wanderley 1994: 18-19), Wanderley does not believe in the power of extensive footnotes or introductory texts to help towards a better acceptance or understanding of a translation. In his opinion, there exists a relation of similarity woven between the author and the translator which allows this last to carry out his task in a dignified way and as intellectually and creatively as the author, as we see in the following excerpt:

The act of translating poetry is established against the intellectual impotence of conceiving the poem as untranslatable (all of it is, nothing is) and in some cases – maybe especially in this case – against the reverential fear of such an established and mythical author. However, at the end of it all, the author, whoever he may have been, wrote with a quill, on a piece of paper and on a smooth surface, exactly like the translator.

Wanderley 1994: 19

As far as the metrical and rhyming choices are concerned, Wanderley (1994: 19) conceives Shakespeare’s sonnet as “a basically musical […] being,” and he also believes that a translation in ten-syllable lines – an emblematic verse type in the Portuguese language – must adopt this musicality. This is more important “than the losses it may have suffered during the negotiation” (Wanderley 1994: 20). Once more, the notion of the process of loss and compensation is referred to, in which the translator needs to recognize the most relevant elements that must be part “of the translation’s boarding list” (Wanderley 1994: 20) into the target language, guided by the safety of common sense. Therefore, Wanderley uses the same rhyming scheme as the Bard, expressed in twelve cross-rhymes and two parallel rhymes.

These brief discussions of choices of the translators in question serve to emphasize both the differences between their motivations (that are due to aesthetic-literary affiliations which are peculiar to each one) and the points they have in common. The latter are related to the idea that a poetic translation makes up a strictly compensatory process, which implies guilt at the absence of total equivalence between the original and the translations. This can also be inferred through the translations themselves, as will be observed in the following sections.

3. Sonnet 55

Sonnet 55 reveals the theme of the immortalization of the beauty of the fair youth, so future generations may equally be admirers, an idea reiterated throughout the Sonnets’ sequence of 154 poems. Somewhat hyperbolically, the poet believes his powerful rhyme will outlive marble or the monuments of monarchs, a superior source of light to that of time-weathered stone.

Apart from surviving the corrosive action of time, the poet’s verse will be able to resist the destruction that war brings. Its living record, the paper on which the verses were written, would only survive through the publication of the manuscript, “to ensure the proliferation of copies,” as states Duncan-Jones (1997). In summary, Shakespeare focuses in this sonnet on the downfall of human kind, devastated by a succession of conflicts and the merciless action of time, a process which, though it leaves only destruction and suffering in its wake, is unable to erase such intense feelings as those offered to the fair youth.

3.1. Ramos’ translation

Ramos was one of the most distinguished poets and poetry critics of the Generation of 1945, the third modernistic phase in Brazil. The poets in this group, gathering in São Paulo and Rio de Janeiro, had a negative opinion of the free verse used by the first-phase modernists of 1922, preferring to adopt “other forms of verse based on the rules and disciplines of versification” (Coutinho 1986: 195). This position imposed on this generation the pejorative label of neo-Parnassian.[6]

Ramos’ translation offers a clear view of the translator’s link with the Generation of 1945 because of his preference for the 12-syllable verse (Parnassianism’s favourite) and the maintenance of the rhyming scheme of the English poem. To replace the rhetorical-formal aspect of the sonnet, which, according to Ramos, “the translation from English to Portuguese cannot keep [and] at the same time [maintain] strict fidelity to form, meter, rhythm, and rhyme” (Ramos 1970: 12), he inserts some additions, even though, in his introductory text, he defends the permanence of meaning and poetic images. In the first verse, there is a rhetorical insertion of the expression não sei (I don’t know),[7] aiming at filling the extension of the chosen meter, “De mármore não sei, nem de áureos monumentos” (Of marble I don’t know, nor of golden monuments) (Ramos 1970: 75).

On the layout of the page showing the translation, Ramos presents two possible ways of translating Shakespeare’s text: in the text itself the voice of the poet/translator and, in the long footnotes at the bottom, the voice of the scholar/translator. In the first case, as in lines 7 and 8, “Não poderão espada ou fogo destruir/O vivo memorial que te há de relembrar” (Nor sword nor fire will destroy/The living memorial which will remember thee) (Ramos 1970: 75), Ramos aims for poetic effect and the establishment of a rhyme in derruir-destruir (fall-destroy, lines 5 and 7). In the second case, in footnotes 7 and 8, the poet presents to the reader what he considers a literal translation, with no links to requirements of a poetry translation such as limitations of meter and rhyme and use of a less colloquial, more poetic vocabulary: “Nem a espada de Marte nem o fogo rápido da Guerra queimarão/O memorial vivo de tua lembrança” (Neither Mar’s sword nor war’s quick fire will burn/The living record of your memory) (Ramos 1970: 75).

As for lexical choices, Ramos’ attempt to apply a refined and complex style to the poem is quite clear in his use of archaic Portuguese terms. As far as syntax is concerned, Ramos tries to follow the tendency of Shakespeare’s text to leave the verb form at the end of the line, which is foreign to a Brazilian reader, who is accustomed to the verb immediately following the subject, at the beginning of the sentence.

3.2. Mendes’ translation

Oscar Mendes also adopts alexandrine verse, but without rhyme. This choice is related to his own idea that, “not being a poet” (Mendes 1969: 737), he as a mere translator cannot deal with rhyming schemes combining sound and meaning. His translation is thus restricted to images and poetic concepts in a style close to prose.

He uses a vocabulary more accessible to the public of his time. There is an interesting example in line 9, where he translates “all-oblivious enmity” as “a aversão que de tudo se esquece” (the aversion that forgets all) (Mendes 1969: 837) because, as he does not consider himself under the constraints of rhyme, he can use longer and more explanatory clauses. On the other hand, the syntax used by Mendes is archaic, as it is an approximation to the order used by the Bard. An example is lines 10 to 12, “vosso louvor lugar/Aos olhos achará das gerações vindouras/Que passarão, até o Juízo Final” (thy praised place/To the eyes of the coming generations will find/That will pass, until Judgement Day) (Mendes 1969: 837). This syntactic organization makes it difficult to grasp the meaning. We see that Mendes, like Ramos, seems to want to create a scholarly translation, full of lexical and syntactic choices referring the reader back to the text produced centuries ago.

3.3. Wanderley’s translation

Jorge Wanderley is the only translator in the present analysis who employs the ten-syllable verse and the same rhyming scheme as the English text, giving priority to musicality in poetry translation. Wanderley’s translation is indeed clearly musical and fluid, a characteristic enhanced by the use of modern Brazilian Portuguese, accessible to a wider public. He uses phonic effects which are non-existent in the original, such as the alliteration of [v] in line 6, “E virar pelo avesso alvenarias” (And turn masonry inside out) (Wanderley 1994: 141), and of [m] in line 9, “Contra a morte e o maligno esquecimento” (Against death and evil forgetfulness) (Wanderley 1994: 141).

Despite the choice of current lexical items, Wanderley manages, like the other two translators, to keep a syntactic organization similar to that of Shakespeare: in line 5, “a guerra as estátuas devastar” (the war the statues devastates) (Wanderley 1994: 141). He postposes the verb and, in line 6, translated above, postposes the subject and separates it from the complement (“virar pelo avesso alvenarias”) (Wanderley 1994: 141). As Wanderley believes that “the musicality of a translated verse […] will always speak louder” (Wanderley 1994: 20), he makes clear his intention to maintain the sonnet’s archaic syntactic structure in order to preserve the music which is a part of this same structure.

4. A dilemma in translation: fidelity to self or to the other?

From the motivations inferred from the three Brazilian translators’ commentaries on form/content choices, which could be seen in this brief analysis of their translations of Sonnet 55, I have deduced a very particular and traditional concept of translation, which sees the task as a negotiation process. The idea that a translated text must try to strike a satisfactory balance between what is lost and what is gained in the linguistic process is quite clear in the explanations of additions and suppressions of elements, slight modifications and blatant alterations.

By choosing one verse form instead of another, by finding one lexical item more formal, and therefore more in tune with a late 16th /early 17th century text, the translators transgress their traditional contract of fidelity and impose, once and for all, their authorship upon the texts. Their (re)reading of the earlier text is based on conceptions of language, translation, and poetry, as well as certain ideological affiliations peculiar to the sociohistorical, cultural, and linguistic community they belong to. According to Arrojo,

the traditional notion of fidelity […] is nothing more than an effective resource which enables the translator to take refuge from the feeling of guilt, a result of the ‘transgression’ produced by his inevitable interference in the text he translates.

Arrojo 1993: 81

This guilt suggests fidelity to the translator’s own voice, which cannot be erased from the living text he is to produce. Complete fidelity to the Other of the author, as tradition states it, is an impossible task, a fruitless attempt to reproduce word for word, rhyme for rhyme, intention for intention, what the author produced and apparently meant to say in his moment of creation. Thus the translator’s immediate context has an important part to play in his production.

In the present case, Péricles Ramos provides at the same time a critical analysis and a poet’s labour, placing the literal translation against the poetic one and delimiting the boundary. Ramos makes choices which relate directly to the aesthetic-literary affiliation of the Generation of 1945, such as the use of the 12-syllable verse with rhyme, the formal, refined vocabulary, and an eloquent tone.

Mendes’ translation is characterized by an accessible vocabulary and a near-prose style without rhyme, as well as syntax evoking a late 16th /early 17th century text. The use of this archaic syntax may be Mendes’ idea of compensating for blank verse, calling out to any impending critic: “Throw the first stone, those who have never sinned in matters of translation” (Mendes 1969: 737). He reinforces the logocentric vision which sees in translation a sinful act, a transgressor of norms and foresees criticism of his rupture with tradition’s fidelity contract.

Wanderley’s more contemporary text, in its turn, is characterized by the musicality of rhymed 10-syllable verse. Similarly to Mendes, Wanderley presents language accessible to the modern public, though with archaic syntax. By stating that “translation is a being in permanent motion” (Wanderley 1994: 23), he regards the modifications in his text as positive. This attitude differs from Ramos’ characterization of poetry translation as an unfaithful task, and from Mendes’ apologetic tone when accounting for differences in his text.

In summary, this view of translation rejects the indeterminate play of meanings evoked by Derrida, for the idea of losses and gains presupposes the old notion of transfer and transport, ignoring the context surrounding the translator which makes the translation a text in its own right. In this era of post-modernism and post-structuralism, a redefinition of the translator’s role should be assumed by all those involved in the field and especially by the translators themselves, who must be aware of and accept their undeniable presence in the text they sign. In a word, the semiological tradition that sees translation as a series of losses and compensations has constituted a long denial of style, context, and difference.