Basis and framework

British writer Agatha Christie (1890-1976) was (and is) a unique phenomenon in literary history worldwide. Due to this, from the perspective adopted in this study, a topic such as “Agatha Christie in Brazil”, just like “Hemingway in Europe” or “Tolstoy in Germany”, might be envisaged from the point of view of what Goethe termed as “World literature” (for more information on Goethe’s concept of Weltliteratur, see Eckermann (1930). But, considered this way, such topics are never simple, since the entire issue of world literature is never simple. As a matter of fact, not even the supposedly “less broad” concept of “national literature” is a simple one. And it certainly is paradoxical to talk about world literature without first considering the concept of national literature.

At first, the idea of world literature sounded almost blasphemous, mainly because the Nation-State strongly cultivated the idea of a national literature. At the end of the 18th century, Goethe linked the idea of national literatures with particular cultural traditions, but only after the second half of the 20th century literary theoreticians and comparatists such as René Étiemble (1975) and Jonathan Culler (2007) discover that the very concept of literature was hardly any older than the idea of national literature.

Of course without the understanding that different countries and cultures had their own literary traditions, research on national literatures and their development and, furthermore, research on the various kinds of relationships between these different national literatures (such as what is proposed in this article) would probably never have started. However, as is well illustrated by Baldensperger and Friederich’s 1950 bibliography on Comparative Literature, the study of what was called world literature was for quite some time a delicate matter, for example dividing continents in those author’s post-World War context and, thus, going against Goethe’s original view of uniting continents.

On the other hand, the world literature phenomenon was exemplified in a practical way long ago. Several writers from the 19th century, for example, were fascinating examples of international success stories: Hoffmann and his Contes Fantastiques as well as Edgar Allan Poe, just to mention a few among many names, became famous abroad before being really successful in their own countries. Their national reputation was largely influenced by their international success. And, perhaps coincidentally (but most likely not), just like Agatha Christie the aforementioned authors also wrote popular books that could fall into the category of “thrillers”.

Only decades after literary comparatists and theorists had established their discipline, sociologists rather than literary scholars realized that, in the age of globalization, literary systems are much more than just an accumulation of national traditions (for more insights on this topic, see Casanova (1999)). It took a long time before Comparative Literature scholars realized that the interaction between literatures deserves to be examined as a key component of the social, cultural, economic and religious relationships between countries, nations, communities, etc.

However, little by little, the study of peculiar aspects in the historical development of literary dynamics within a broader spectrum has been making progress, which in its turn opened the path for studies considering diversities in the development of literary traditions in different places. Thus, space was opened to what may be called a wave of colonization and decolonization investigation within literary studies, which made it clear that “literature” has not simply been brought to the colonized world as part of “civilization”, together with economic, social or religious goods, but underwent different processes for obvious reasons.

Nonetheless, the history of this modern effort in research is still very complex. And much more complexity is added to this panorama when one considers research on translation phenomena. Therefore, it would not be wise to start any research about an author such as Agatha Christie and her position in a Latin American country without first explaining this background.

Brazilian cultural dynamics and Agatha Christie’s translations: what’s the connection?

A lot has been written both about Christie and her novels, as well as about Christie and “the translation issue.” Analyses with titles such as “Agatha Christie in Brazil” and “Agatha Christie’s works in Portuguese” are probably available in Comparative Literature publications, even though, surprisingly, a search on Google Scholar did not provide any results for such titles. However, neither is that what the reader should expect from this article.

One may ask, while considering Christie’s work: did the development of Brazil’s own literary traditions favor or disfavor “peripheral” genres coming from Europe? Has the question of the canonic literary discourse, if not the mere availability of a standardized language, been a handicap for the production of translations? These general questions indicate the type of discussion proposed by this article.

This text tells a story of languages, nations, communities... Agatha Christie is a full world, not only from the perspective of mapping Brazil, but also from the perspective of Portugal, France, etc. After all, one of the surprises when analyzing Agatha Christie in Portuguese is to see that such countries pop up in her books in so many different aspects.

The particular point about the distribution of Christie’s books in Brazil is that it enters a cultural complexity where several developments unfold regarding discourse, fiction, literature in society and many others on the basis of so-called post-colonial studies. As Simon (1997, 463) would put it, “translation studies [...] investigates the linguistic and textual realization through which cultural exchange takes place”. The goal here is to put this into practice, checking this cultural exchange between Latin America and Europe represented in the translations of Agatha Christie, and verifying what can be learned from them.

Orality, literature and translation

One fascinating option for studies in the area of translated literature in Brazil is the question of written representations of oral discourse. Orality or, more specifically, the question of oral vs. written language, is a complex and sensitive subject in Brazil, which has to do with the fact that language is not as institutionalized as in other countries with older written and literary traditions. According to Aguiar (2013, 12), impelled by the recent abolition of slavery and the country’s independence, a nationalistic trend emerged in Brazil at the end of the 19th and beginning of the 20th centuries, and along with it came a drive to consolidate Brazilian orthography. Aguiar also cites Veríssimo, who states that Portuguese still had no established orthography during this period in either Brazil or Portugal. Each writer and each publishing house developed a different, peculiar spelling style, somewhat like the situation of Elizabethan English. This confirms Hobsbawm’s statement that “historically, the coexistence of peoples of different languages and cultures is normal; or, rather, nothing is less common than countries inhabited exclusively by people of a single uniform language and culture” (Hobsbawm 1996, 1068). Brazil, a Portuguese-speaking nation, proved to be a country of many Portuguese languages on both the orthographic and spoken levels.

Thus, the focus of this study is orality, i.e., how oral discourse is used in popular translated fictional literary works in Brazil, observing and describing how particular kinds of written representations of oral discourse are filtered and rendered into Brazilian Portuguese.

Orality is an alluring topic, probably because it is connected with the essence of humanity and human interaction/communication. Walter Ong, one of the main proponents of orality as a scholarly theme, wrote that “human society first formed itself with the aid of oral speech, becoming literate very late in its history, and at first only in certain groups” (Ong 2002, 2). However, at a certain point most cultures did become literate and, speaking simplistically, writing began to proliferate not only for strictly utilitarian purposes, but also to entertain and to share philosophical, religious and political points of view in the form of fictional and non-fictional literary works.

Even though literary scholars have only recently been obliged to recognize both world literature and translation as an important component of the new globalized world, literary works have spread throughout history on smaller or larger scales for different reasons (including as a means of colonization). This same process ongoing in most, if not all, of the developing nations of the globalized world is also occurring in Brazil, where “literary colonization” is still in full bloom. According to Wyler (2003, 13), based on data from 1987, 80% of the prose, poetry and reference books sold in Brazil (besides manuals and catalogues) are translated. Thus, it is not at all an exaggeration to say that Brazilian readers are mainly – and almost exclusively – reading translated foreign literary works, and also translated foreign texts in general, beyond literature. But the marriage between oral and written discourse is more particular in fictional texts, and it seems to be a privileged issue in detective stories like those by Agatha Christie.

In literature as well as in general discourse, the particular discursive patterns and models in circulation may impact the dynamics of languages via translations. In what kind of Brazilian Portuguese do Brazilian readers integrate the new genres and the new international writers? More than in the dominant languages around the world, Brazilian translations help shape the new canonized language.

Thus, this study first presents an analysis of two Christie novels translated into Brazilian Portuguese and the approaches used to translate them. These two works happen to represent very general and dominant features of the discursive Brazilian Portuguese Agatha Christie world. They illustrate what can be observed in other Agatha Christie stories, as well as in many other detective stories from the same environment. This is so simply because oral discourse appears to be a problematic issue in fiction from non-canonical literary genres that happen to become successful.

Rather than particular novels from individual writers, we try to observe how a particular kind of “style” (register) is developing in a country where the language of literature and the language of dialogue have still been taking form in recent years. From this perspective, Agatha Christie is a challenge to the extent that her literary style tends to play on the borderlines between “literature” and everyday life while anticipating patterns that Roland Barthes would label “le degré zéro de l’écriture”.

From macro to micro: considerations on the corpus

Toury’s concept of translation norms (1995) and Lambert and van Gorp’s 1985 guidelines for describing and analyzing translations (on preliminary, macro- and micro-levels) serve as the basic theoretical and methodological platform for the study. The preliminary and macro-structural characteristics of the translated texts will be analyzed first, including attention to paratextual details of physical copies. Written oral discourse will then be examined on the micro-level using elements of direct speech, particularly the register.

Hatim and Mason (1993, 47), working within the descriptive framework established by Halliday, McIntosh and Strevens (1969, 87), recognize two dimensions in language variation: user-related varieties (called dialects), and use-related varieties, known as registers. They point out that registers differ primarily on formal questions of language (e.g. grammar and vocabulary class), and make the following comments:

The category of register is postulated to account for what people do with their language. When we observe language activity in the various contexts in which it takes place, we find differences in the type of language selected as appropriate to different types of situation.

“Situation-appropriateness”, therefore, will be taken as a key concept for analyzing the use of register. The aim is not, however, to assess the quality of the translations, but rather to look for regularities in certain situations in Brazilian translations of Agatha Christie. Regularities are understood here as synonyms of what Toury (1995, 53) calls norms. He explains that translation is – most often – target oriented, focusing on the target culture, and that every culture presents certain norms that regulate how (certain) texts are approached.

Moreover, as far as oral and written discourse are concerned, experts in translation studies have demonstrated one particular function of the register issue, i.e. its link with narrative functions within narrative texts and genres. Thus, texts on narrative discourse by Hermans (1996) and Schiavi (1996) are also considered as part of the theoretical framework for all levels of analysis.

Preliminary and macro-analysis of the corpus: publishing houses, translators and general paratexts

The translations are analyzed according to the chronological order of their publication in English, since it was not possible to determine the exact year of publication of the first edition of one of the Brazilian Portuguese translations. Two different editions of the same translation of each book are considered, with a gap of some decades between them, in order to verify any diachronic changes in approach.

Murder on the Orient Express was first published in the United Kingdom in 1934, and is the first novel we will analyze. The older edition of Assassinato no Expresso do Oriente used for this study was published in Brazil in 1979. However, this was already the seventh edition of this title. No information regarding the publication date of the initial Brazilian edition was found.

This book, produced by publishing house Nova Fronteira, is part of a collection that is still in print. The back flap of the 1979 edition enumerates 31 titles in the Nova Fronteira Agatha Christie collection. However, 68 titles are listed in the 21stedition of the same book published in 2002. This indicates the rapid progression of Agatha Christie’s prestige in Brazil, sufficient to warrant the publication of 37 other titles in little more than 20 years.

The 1979 cover boasts Christie’s name in very large print, describing the mystery, immediately underneath, as a Poirot case (which shows that the detective was already famous in Brazil). Only after that comes the title (in much smaller print than the author’s name).

On the title page we find once more the author’s name, the title (this time slightly larger than Agatha Christie’s name), and then, in much smaller print, that it is a translation by Archibaldo Figueira.

Online research indicates that Archibaldo Figueira is mainly known for this specific translation, i.e. one of Agatha Christie’s most famous best-sellers. Although some results from the area of sociology are connected with an Archibaldo Figueira, it would be difficult to determine if this is the same person. However, any academic work notwithstanding, he cannot be described as a successful author, and, moreover, did not write fiction. Thus, for the objectives of this study Figueira can be classified as a “conventional” translator.

Apart from this single mention, the presence of a translator is not noticeable in the paratexts. There are no translation notes and no paratextual elements written by the translator or the editor. Moreover, the reader is frequently left with the impression that he is reading a Portuguese skin stretched over an English text. For example, on p.30: “Poirot, naquele momento, já sabia tudo sobre a filha de Mrs. Hubbard. Todos os passageiros que conheciam o inglês o sabiam". This sentence indicates that English is the lingua franca on the train, which underscores the fact that the text is a translation from English. Furthermore, Mrs. Hubbard’s declarations are uniformly presented in Portuguese, despite the fact that she is American. Hermans (1996, 29) calls this kind of occurrence in translation a “linguistic self-referentiality”, pointing out that cases such as this tend to “foreground” the voice of the translator, usually carefully hidden in the vast majority of literary work.

In the 2002 edition (vaguely termed as the 25th impressão, which could mean either edition or printing) of Assassinato no Expresso do Oriente, there are a few changes, although purely aesthetic ones. The style of the cover has changed considerably, but the main element is still Christie’s name, this time in the form of her signature (positioned below the much smaller name of Poirot). Figueira’s name appears again in the same place on the title page. The new edition is longer, due to the larger font size. There is an editorial note explaining that Constantinople is now called Istambul on page 11, as well as certain changes regarding the telling of time (e.g. on page 213, 0h 37 replaces 12h 37m). Other than that not a single difference can be found, despite claims on the final page of two editors and three revisers having worked on this edition.

In the preliminary and macro-analysis of Cai o Pano, we will examine to what extent this second translation has been treated along similar lines. Curtain was first published in the United Kingdom in 1975 to a warm critical reception, which probably explains its quick translation in Brazil: the first edition of Cai o Pano was published in 1977. The older of the two editions examined in this study featured neither a year of publication nor an edition number (it could actually have been the first edition). Since it was part of the same collection as the above-mentioned 1979 edition of Assassinato no Expresso do Oriente, there is reason to believe that it was published either at the end of the 1970s or the beginning of the 1980s. The first Brazilian translation mentioned by the Index Translationum is from 1984, allegedly an 11th edition.

The book is presented in an almost identical fashion to that of Assassinato no Expresso do Oriente. The name of the translator appears in the same place, on the title page without any special mention, even though Lispector was already a famous author at that point. However, there is no doubt that the translator takes a more visible role in this novel. For example, Christie’s word, sycamore, is translated as plátano on page 10.(1) Since this species is generally uncommon in Brazil, Lispector adds a parenthetical italic explanatory note (tipo de árvore). Schiavi (1996) envisages a translation as a split message from two addressers, the author and the translator. In this case, such ambiguity is in high relief: Christie takes the reader’s prior knowledge of this specific tree for granted, whereas the translator felt obliged to clarify the word for her Brazilian audience, thus adding a layer of readership to the writer’s address.

Another case highlighting the translator’s presence is on p.53, where the word espírita appears. The religion of Spiritism was not nearly as well-known in Agatha Christie’s United Kingdom as it was in Lispector’s Brazil; therefore, this is a clear case of domestication. The English text actually referred to a ‘psychic’.

Such choices by Lispector might have been conditioned by her position in the Brazilian literary polysystem. Her style, both as a translator and as an author, was considered quite innovative, eventually becoming prestigious. Espírito Santo (2011, 27) discusses certain characteristics of Lispector’s writings, including her particular use of Portuguese grammar, which plays a decisive role in the translation of Curtain, as will become clear in the micro-analysis that follows.

In the 2009 edition of Lispector’s Cai o Pano, the layout coincides almost entirely with that of the 2009 version of Assassinato no Expresso do Oriente. However, compared to the initial version of Cai o Pano, at least two differences should be mentioned: (1) an appendix providing titles and summaries of other Agatha Christie works in this collection, including the translators’ names and, surprisingly, (2) the explanatory translation notes (e.g. for plátano) have simply disappeared. Thus, translator autonomy, despite their more visible place in the paratext of this new series, is actually undermined by the editors/revisers in the details of the text.

Micro-analysis of the corpus: the speech of Christie’s characters

In this micro-analysis, different text levels are treated in different ways with respect to register. The following scheme of abbreviations will be used in this section: Assassinato no Expresso do Oriente will be Assassinato; Murder in the Orient Express will be Murder; Cai o Pano will be Cai; and Curtain will remain unchanged. The page numbers always refer to the earlier versions of the translations. Certain cases are presented in which the newer versions were also consulted for the sake of comparison.

At the outset, it must be mentioned that both translations maintain the irrefutably international and exotic characteristic of Christie’s books. The foreign expressions or even entire sentences in foreign languages from the English originals were also found in the Brazilian translations. Four examples follow (the names of the characters speaking are in parenthesis). The second example was completely left in English, whereas in the original its function was just to emphasize the accent of the character:

–Mais oui… Nunca estive antes em Istambul… seria uma pena passar por lá comme ça... (Hercule Poirot, Assassinato, 2002, 10) –I speak a LEETLE, yes. (Condessa Andrenyi, Assassinato, 2002, 111) –Já não lhes disse tudo que sabia? Per Dio… (Antonio Foscarelli, Assassinato, 2002, 207) –Sabe meu amigo, você está bem conservado. Les femmes, elas ainda se interessam por você? (Hercule Poirot, Cai, 13)

The first translation we will consider individually is Assassinato. The speech of Hercule Poirot is analyzed along with that of Pierre Michel, an employee of the train company, and Mister MacQueen, the American secretary of the murdered victim. The choice of these three characters was based on the probability that their discourse would differ: Poirot speaks English as a second language with serious French interference; Pierre Michel is also a non-native speaker of English, but belongs to a lower social class; and MacQueen is a native speaker of American English, as opposed to the British English used by most of Christie’s characters. Even though the source text is not of absolute priority in this analysis, seeing how such a broad spectrum of characters are handled should reveal the (non)treatment of different registers in Portuguese.

The following, for example, are some quotes from Hercule Poirot:

–Não há pressa. (10) –Poderia arranjar-me um leito? (18) –Não se aborreça meu amigo – interrompeu Poirot –, viajarei num carro comum. (21)

In the two first lines, Poirot is speaking to a train company employee and, in the third, with a friend who also happens to be in a higher position in the same company. Nevertheless, irrespective of social class or friendship, Poirot is made to adopt the same formal tone. Such register is simply inappropriate for these kinds of conversations in Brazilian Portuguese. In this type of an environment, what would be expected of a native speaker would be a less formal tone marked by the use of a different selection of verbs (certainly the verb há in the first line would become tem), a different verb-pronoun collocation in the second (arranjar-me sounds too formal), and a compound verb form to express the future: viajarei would become vou viajar.

In this conversation between Poirot (P) and MacQueen (M), the detective is investigating the crime, and questioning M as part of his investigation routine.

P: Minha tarefa – disse Poirot – é saber de todos os movimentos dos passageiros no trem. O senhor não deve se ofender compreende? É questão de rotina. M: Claro. Prossiga e deixe-me esclarecer-lhe sobre meu caráter, se eu puder. [...] P: O senhor notou que o trem havia parado? (76-77)

Even though in this dialogue the circumstances would justify a certain formality, the poetic register utilized here does not correspond to either common speech or the register of the English, which follows below:

"The task before me," said Poirot, "is to make sure of the movements of every one on the train. No offence need be taken, you understand. It is only a matter of routine." "Sure. Get right on with it and let me clear my character if I can." [...] "You noticed that the train had stopped?" ”Oh, yes. We wondered a bit." (64-65)

MacQueen’s speech involves colloquial vocabulary (e.g. “get right on with it”), whereas in Portuguese he takes a more formal tone. And Poirot’s register was also elevated in Portuguese, changing even the pronoun “you” from the English text to o senhor, which is quite formal. Such an approach can be found in other Agatha Christie translations as well as translations of other English-language authors, resulting in awkward situations and, more importantly, drawing the cultural barrier between nations and languages into sharp focus.

HP: O que aconteceu exatamente? PM: Bati à porta, mas ele respondeu que se enganara. (70)

The same kind of register observed in the previous situations is present here, particularly in Pierre Michel’s speech. Christie’s phrasing was “I knocked at the door, but he called out and said he had made a mistake”, i.e. only commonplace English expressions are used. Thus, a question arises: if these three situations present register higher than the English original and, apparently, Inappropriate for their contexts, what is it, exactly, that makes them high? What rings (or could ring) in these lines to a Brazilian ear as inappropriate?

The fact is that the register’s elevation is chiefly due to its approach to conjugation, lexical choices of verbs and verb-plus-pronoun collocation. The use of verb constructions such as viajarei,(2) deixe-me,(3) havia,(4) or enganara (5) to represent oral discourse which, although still considered perfectly correct grammatically, are rapidly approaching extinction in common everyday speech in Brazil. Even though it would be problematic and difficult to clearly demonstrate (as if it were a homogenous unit) how Brazilian Portuguese has changed since the publication of this translation 35 years ago, it cannot be ignored that the same translation is still being sold today virtually unchanged. In addition, as Martins (2012, 174) points out, significant grammatical changes were already underway in Brazilian Portuguese by the 19th and possibly as early as the 18th century, opening the door for different approaches to the Portuguese language on every level in Brazil. Furthermore, in a 2003 study, linguist Ane Schei (155) focused on the use of pronouns in Brazil both in spoken and literary Portuguese, analyzing the discourse in six books published between 1975 and 1997. She considered their use of the language as contemporary Brazilian Portuguese, which supports our classification of the present study’s corpus and the relevance of the constructions it uses to represent oral discourse.

Last but not least, pronoun collocation is another indicator of high register in the present sample. Arranjar-me (used by Poirot) or deixe-me (used by MacQueen) are not part of common oral discourse in Brazilian Portuguese. Bagno (134) points out that:

In all varieties of Brazilian Portuguese oblique pronouns have been replaced by nominal pronouns in constructions such as “deixa eu ver”, “mande ele entrar’, ouvi ela chorando’, etc., where normative grammar prescribes “deixa-me ver”, “mande-o entrar”, “ouvi-a chorando”, etc., a morphosyntactic innovation that completely took over the spoken language and is greatly present also in the written language, even in closely monitored texts. (6)

Therefore, since the translation uses Portuguese grammar which, at least according to Bagno, is “outdated”, the final result is a text far removed from “real” speech. The translation makes perfect use of Brazilian Portuguese “book” grammar, but current everyday life conversations have already incorporated many innovations that are not yet recognized by the Academy of Letters.

So far we have concentrated on one particular translation. Yet within this particular text, the trend towards literary, socially distinguished speech is quite systematic. However, before moving onto issues involving dissemination or diachrony, it may be worthwhile to observe how established Brazilian writers may integrate the particular patterns from a given detective story into their everyday world, and vice-versa. In this context, it seems relevant to introduce Clarice Lispector, the translator of Curtain, in greater detail.

Lispector (1920-1977) was one of the most renowned 20th-century Brazilian authors. She began her career writing short stories for newspapers and in 1943 published her first novel, Perto do Coração Selvagem, which received a prestigious literary prize and was soon followed by many others. Her writings cannot be classified in a single genre due to their unique and innovative characteristics, but her literary style has been often compared to that of Virginia Woolf and James Joyce. The New York Times has called her the equivalent of Kafka in Latin American literature. Before Lispector’s debut, regionalist commitment pervaded in Brazilian literature, and the novelty of her style at that moment garnered her much attention. In the 1970s Lispector lost her position at the newspaper and began to work as a translator, translating from French and English (she translated authors such as Edgar Allan Poe, Oscar Wilde, Anne Rice and Agatha Christie).

In Curtain we will examine the oral discourse of Hercule Poirot, Hastings (Poirot’s friend and the narrator), Mrs. Luttrell (owner of the boarding house were the crime happens) and Judith (Hastings’ young daughter). These characters were chosen because of their different backgrounds: an old foreigner; an old Englishman; an old lady who, according to Christie, tries to sound Irish; and a young ‘modern’ girl conversing with her father.

The following sentence by Hastings carefully reflects everyday oral discourse in Brazil, unlike Figueira’s approach:

Não, não, ainda não tinha reparado. Sempre imaginei ele totalmente envolvido com seu trabalho. (66)

In this example Hastings is chatting with another guest at the boarding house where Poirot is staying; the register demonstrates the informality of the situation. The device that Lispector used to accomplish this informality, however, involves the complete opposite of Figueira’s approach: the pronoun collocation agrees with current spoken Brazilian Portuguese and breaks the rules of standard grammar (imaginei ele). What is surprising is that, in the 2009 edition of this translation, this “incorrect” use of grammar was undone. The same sentence says: “Não… Não… ainda não tinha reparado. Sempre o imaginei totalmente envolvido com seu trabalho” (Cai, 2009, 77).

The following dialogue between Poirot (P) and Hastings (H) serves as another example of the informal tone of Lispector’s translation:

P: É, mas eu lhe asseguro amigo, isso é um teste, “o teste”. Quando as jovenzinhas chegam para você conversando com tanta gentileza, oh, tão gentilmente, é o fim! “Coitado do velho – dizem – vamos ser boazinhas para ele”. [...] H: Você realmente é demais Poirot. E como é que vai você? (14)

This passage uses diminutive forms (jovenzinhas) and orality markers to give a general impression of informality.

The same colloquial feel is also present in two different lines by Mrs. Luttrell:

Coitadinha. Está enfurnada naquela casinha lá no fundo do quintal. Estou alugando-a ao Dr. Franklin. Ele arrumou a casa todinha. Está cheia de porquinhos da índia, pobrezinhos, ratos e coelhos. (9) Você não tem o menor desconfiômetro. Passe essa garrafa pra cá. Passe logo, ora. (73)

Once again, the use of diminutive forms, orality markers, and even common-sense neologisms such as desconfiômetro (“common sense meter”) sets the informal register. In addition, contracting the preposition para with the article a (resulting in pra), an absolutely oral construction, helps set the informal tone. It is true that a higher register pronoun collocation (alugando-a) is also observed, but the informality generated by the other elements seems to overshadow this formal element. Nevertheless, this hesitation cannot simply be overlooked; Lispector is trying to balance the use of more and less prestigious structures for a reason, be it not to offend the grammarians to a great extent or to stress the different social and discursive layers of the original English text.

Family life conversations within this translation also provides interesting cases, as in this interaction between father and daughter, Hastings (H) and Judith (J):

H: Então o Allerton não estava com você? Judith bufou de raiva. J: Não, não estava – falou com nuances de fúria. – Eu só me encontrei com ele perto da casa e só uns dois minutos antes de encontrar você. Eu espero que você esteja satisfeito. Mas se eu tivesse passado o dia inteiro com o major Allerton, você não teria nada com isso. Eu tenho vinte e um anos, me sustento, e como uso meu tempo não é nem um pouco da sua conta. H: Nem um pouco – falei, tentando amenizar a enxurrada. J: Ainda bem que você concorda – Judith estava mais calma. Deu um sorrisinho sentido. – Ah meu querido, tenta um pouco não ser tão superpai. Você sabe o quanto é irritante. Se você evitasse essa preocupação exagerada... H: Vou evitar, realmente não vou te aborrecer mais com essas coisas. (85)

Judith uses a verb conjugation seen as incorrect based on standard grammar, but ubiquitous in oral discourse (não tenta). The 2009 edition ‘corrects’ this, using the imperative form, tente, which is prescribed by normative Portuguese grammar (101). The pronoun collocation also obeys popular oral discourse (me sustento, te aborrecer), although these were not altered in this case in the 2009 edition.

It is remarkable that the register in some of these sections is even less formal than that written by Christie, perhaps a kind of hypercorrection. Lispector attributes certain discursive characteristics that are not necessarily deducible from Christie’s text. For example, the constant use of diminutive adjectives and nouns, particularly by Mrs. Luttrell, is not nearly as systematic in the English text: Christie may use the word little occasionally, but this feature was intensified in Lispector’s translation. Compare the five diminutives in Lispector’s version (Cai, 9) with Christie’s original:

Ah, the poor girl. She’s cooped up in that studio place down at the bottom of the garden. Dr. Franklin rents it from me and he’s had it all fitted up. Hutches of guinea pigs he’s got there, the poor creatures, and mice and rabbits. I’m not sure that I like all this science, Captain Hastings. (Curtain, 10)

Authors who also translate, especially respected prolific authors, have the privilege of drawing from the treasury of resources/frameworks developed in their own original work. A cursory examination of original works can often reveal to what extent the translator recycles the patterns of his/her own repertoire. The fact that Lispector uses the same kinds of resources in this translation found in her own original writing is not a surprise. A passage of one of her most famous stories has been pointed out by Bagno as a clear example of ‘real’ Brazilian Portuguese usage (165). The predictability here was due to the fact that, in previous analysis of Lispector translations, it was proven that her own writing style was present in the translated text. Gonçalves, in an analysis of Lispector’s translations of Edgar Allan Poe in Brazil, points out that she does not follow the text formally, but retells it in a more colloquial tone (2006, 99).

In conclusion: telling crimes stories in Brazil

This study illustrates the complexity of the written vs. oral discourse question in Brazil. The different approaches to oral and written forms observed in the work of different translators highlights the hesitation and disagreement present in their choices, demonstrating that they are not at ease with Agatha Christie’s discourse. The examples shown above, therefore, are not so much a contradiction as a confirmation of this ongoing process of coming to grips with orality in Brazilian literature, especially translated literature.

This crucial issue must also be linked with the traditions of colonization and colonial discourse or culture. Brazilian Portuguese can hardly avoid the following dilemma: to support or not support the more specific linguistic characteristics of the various Brazilian states/regions, which would imply the “loyal” exploration of various different local accents and priorities. Respect for standard grammar, lexicon or metaphorical items and experimentations, however, implies respect for international ‘civilized’ norms, i.e. those of European Portuguese disseminated in Brazil. In fact, the closeness between the Portuguese language used in Christie’s Brazilian translations and the Portuguese used in Portugal was such that, according to Sampaio (2007), contrary to expectations, Brazil was actually exporting detective story translations to Portugal in the 1930s and 1940s, which underwent just some minor changes before being released in the former colonizer’s market. Since Agatha Christie is a successful non-canonized European classic, her work presents a challenge on several levels.

The approaches observed, particularly in the more recent versions of the Agatha Christie novels, indicate to a certain extent that there is a persistent pursuit of (more) prestigious literature in Brazil.(7) Different varieties of English (discourse) have little chance of being welcomed as components of (Brazilian) literature in the (post)colonial future. Hence, there is no way of disconnecting Brazilian translations of Agatha Christie from the very particular stylistic and literary values prevailing in the country: as translation theoreticians often indicate, the canonized stock of expressions and lexical/idiomatic selections reduces and stereotypes imported works, even (and particularly) in the case of less conventional works and genres. Detective stories, first of all Agatha Christie, become a test case for innovation in both genre and register, if not in the profile of characters whose discourse is the first key to their identity.

Further investigation is necessary to confirm which style of register predominates in the translations of Christie’s books in Brazil and to establish whether Lispector’s model is isolated or not. We suspect that Figueira’s approach will prove to be predominant. If confirmed, however, the presence of a consistent register clash between translations and their source texts would hardly be a novelty in worldwide terms. Edwards (2010, 102), referring to Skvorecký’s writings about the translations of Agatha Christie into Czech, states: “they made Hercule Poirot talk like the other characters whereas, in Christie’s English original, the clever Belgian detective speaks a very ‘Frenchified’ English. But a new translation made Poirot sound like a Sudeten German.”Although Poirot is made to speak a “Frenchified” Portuguese in the Brazilian translations, this quote shows, nonetheless, that oral discourse has also been manipulated in Czech translations.

Chantal Leroi (1978, 128), a Belgian who conducted a comprehensive study of French Christie translations, concluded that the translators’ language is quite sophisticated compared to Christie’s English, ascribing a higher level of discourse to the characters. The result is that, via translation into several languages, Agatha Christie has a more elevated position in international literature than she did at home. This is clearly seen in Brazil, where Christie’s works are in a certain way treated as canonized literature, for example being published in a book collection side by side with authors such as Plato, and being introduced by a publishing house as a possibility to teach literature for adolescents. Probably this was only possible because the language found in Christie’s translations makes them very similar to canonized Brazilian literature (for more information see Hanes, 2015).

Marjolijn Storm’s forthcoming research on Agatha Christie translations into Dutch and German indicates that the speech patterns employed are generally standardized. She found that the translators, with very few exceptions, do not imitate Poirot’s grammatical and idiomatic mistakes. Regarding Poirot’s franglais, in pre-war German translations all the French is translated into German, whereas in the post-war translations it is retained. However, in the Dutch translations it has always been retained. Once more, even though only for a historical period and probably due to political questions, the same standardization regarding the French language found in the Czech translations was also present in the German ones. Nevertheless, even after the war, the standardization of other elements, particularly grammatical ones, remained in written oral discourse both in German and in Dutch translations, as it also did in Brazil. Based on this information about Christie’s translations in different European countries, the translation carried out by the “conventional” Brazilian translator seems to be representative of a worldwide phenomenon, due to its language standardization and elevated register. The standardization of the discourse of all characters, regardless of social background, was present, as it was in Czech, Dutch and German. The sophistication of the register was similar to that found in the French translations.

These similarities between the Brazilian and European approaches are clear indicators of a historical stronghold of European tradition in Brazilian translation policies. Paraphrasing Even-Zohar (1990, 12), due to the central position occupied by translated literature in the Brazilian literary polysystem to this day, and at the same time to the long-term influence of European literary traditions in the country (particularly the French), such similarities were not at all unexpected. Concerning the influence of the French tradition, Wyler (2003, 56) mentions that at the beginning of the 1900s most books in Brazil were imported, and mainly from France. French prevailed as a literary lingua franca (8) in the country for quite some time, being used as an intermediary for translations. Regarding translations, Amorim (2000) states that French was used as an intermediary language to translate classics from languages that she classifies as “de difícil acesso” (60). The possible regularities in Christie’s translations may just be exactly what would be expected due to these long-standing traditions. According to Bogucki (1996: 182), “since rendering a message into another language often means trying to achieve something in the other culture (...) general rules of politeness have to be observed to avoid losing one’s audience”.

The main regularity that stood out in both Brazilian translations was the retention of Poirot’s French words and expressions throughout the texts. This also seems to be a tendency in other countries, judging from the information found in other studies. However, more is at stake here than just the use of foreign words: the present sample of Brazilian translations left no doubt about the country’s openness to charming, quirky individual discourse by certain lead characters in detective novels. This implies a preliminary answer to the question of whether a mixture of oral and written discourse in Brazilian literary works is acceptable, which could lead, in further investigation, to the heart of the matter.

Based solely on these two translations, however, we have little claim to any diachronic conclusions. Nevertheless, to the extent that our analysis refers to what is tolerated and what is not, we have at least one definite hypothesis to be tested in other texts: register higher than that of common conversation has consistently been used to represent oral discourse in literary works in Brazil for many decades. What we are not saying, of course, is that Agatha Christie in Brazil involves no everyday language whatsoever. Instances of that can certainly be found in the translations, but it is their systematic combination with high register that make this case so symptomatic.

The limited corpus of documents investigated so far indicates at least how the limits of linguistic homogeneity are under threat in Agatha Christie’s Brazilian Portuguese texts. Whether they reflect much larger trends in the Portuguese used in literary/written/oral speech will need to be explored in a much larger corpus. At the very least, this symptomatic case of detective fiction seems to serve as a key to language(s) in flux, perhaps especially in societies where languages are in search of standardization.

In closing, two further examples may help illustrate this general panorama. On the one hand, some recurring vocabulary choices found in Vallandro’s translation of Christie’s The Murder of Roger Ackroyd, such as the word alvitre, as well as the use of certain pronoun collocations, are also constantly found in Machado de Assis, the most prestigious Brazilian author of the 19th century. This (probably subconscious) repetition shows what type of works Vallandro could have used as a model. On the other hand, Rachel de Queiroz’s translation of At Bertram’s Hotel could prove to be closer to Lispector. Queiroz was also an acclaimed author and infused her own writings with a sophisticated political background and heavy social content. What can be said about her translation at this point is that a strong use of colloquialism is apparent from even a cursory examination.

Cases such as these show that there are still many chapters of this story to be unraveled that will reach beyond Agatha Christie and possibly lead to the discovery of a general trend in the approach to translated literature in Brazil.

NOTES

(1) Plátano is not to be confused with the Spanish word for plantain/banana, i. e. it refers to the same family of tree as the sycamore, although the word sicômoro can be found in Portuguese dictionaries (http://dicionario.priberam.pt/sicômoro).

(2) Synthetic –or sometimes called “simple”– future, analogous to “will” vs. “going to”.

(3) Ênclise - hyphenated construction of a verb plus an oblique pronoun.

(4) Haver is a transitive, auxiliary and pronominal verb. As an auxiliary verb it is used as a synonym of “to have’” (i.e. perfect tenses), and it is also one of the options usually chosen to translate the impersonal “there is” or “there are”. As a transitive verb, it is a synonym of “to exist”.

(5) Pluperfect tense - indicative mode: used to refer to something that occurred earlier than the time being considered, similar to past perfect.

(6) Original text: Em todas as variedades do português brasileiro, os pronomes retos substituíram os pronomes oblíquos em construções como “deixa eu ver”, “mande ele entrar”, ouvi ela chorando”, etc, em que a norma padrão prescreve ‘deixa-me ver”, “mande-o entrar”, “ouvi-a chorando”, etc, uma inovação morfossintática que já se apoderou completamente da língua falada e já dá sinais de grande presença também na língua escrita, mesmo em textos mais monitorados (my translation).

(7) Considering the translations of The Murder of Roger Ackroyd from 1951, by Leonel Vallandro, and from 2010, by Renato Rezende, the first uses high register vocabulary for the speech of all characters, unlike Christie’s original, whereas the latter presents clear signs of attempts of foreignization such as the use of quotation marks to indicate oral discourse, which is not a common Brazilian practice.

(8) As well as the language of diplomacy: it was being used for passports as late as the 1920s.

REFERENCES

AGUIAR, Monalisa dos Reis. “As Reformas Ortográficas da Língua Portuguesa: uma Análise Histórica, Linguística e Ideológica.” Revista Filologia e Linguística Portuguesa, 9.1 (2007). Available at http://www.fflch.usp.br/dlcv/lport/flp/images/arquivos/FLP9/Aguiar.pdf. Accessed April 10, 2013.

AMORIM, Sônia Maria de. Em Busca de um Tempo Perdido: Edição de Literatura Traduzida pela Editora Globo (1930-1950). São Paulo: EDUSP, 2000.

AVELAR, Juanito, and Dinah CALLOU. “Sobre a emergência do verbo possessivo em contextos existenciais no português brasileiro.” In: Ataliba CASTILHO et al. (orgs.), Descrição, história e aquisição do português brasileiro.. São Paulo: Editora Pontes, 2007, pp. 375-402.

BAGNO, Marcos. A Norma Oculta: Língua e Poder na Sociedade Brasileira. São Paulo: Parábola Editorial, 2010.

–– A Língua de Eulália. São Paulo: Contexto, 2011.

BALDENSPERGER, Fernand, and Werner Paul FRIEDERICH. Bibliography of comparative literature. Chapel Hill: University of North Carolina, 1950.

BOGUCKI, Lukasz. "On rendering register." Translation and the (re) location of meaning. Selected papers of the CETRA Research Seminars in Translation Studies. Vol. 1996. 1994.

CASANOVA, Pascale. La République mondiale des lettres. París: Éditions du Seuil, 1999.

CHRISTIE, Agatha. Cai o Pano: o último caso de Poirot.Translated by Clarice Lispector. Rio de Janeiro: Nova Fronteira, 19--.

–– Cai o Pano: o último caso de Poirot. Translated by Clarice Lispector. Rio de Janeiro: Nova Fronteira, 2009.

–– Assassinato no Expresso do Oriente. Translated by Archibaldo Figueira. Rio de Janeiro: Nova Fronteira, 1979.

–– Assassinato no Expresso do Oriente. Translated by Archibaldo Figueira. Rio de Janeiro: Nova Fronteira, 2002.

–– Curtain: Poirot’s Last Case. Great Britain: Fontana/Collins, 1977.

–– Murder on the Orient Express. E-book: Amadio Inc., 2009.

CULLER, Jonathan D. The literary in theory. Stanford: Stanford University Press, 2007.

DELABASTITA, Dirk, and Rainier GRUTMAN (eds.). Fictionalising Translation and Multilingualism. Special Issue of Linguistica Antverpiensia, 4 (2005). Available at https://lans.ua.ac.be/index.php/LANS-TTS/issue/view/9. Accessed April 15, 2014.

ECKERMANN, Johann Peter. Conversations with Goethe. Translated by John Oxenford Dent. New York: E. P. Dutton, 1930v

EDWARDS, John. “The Treason of Translation? Bilingualism, Linguistic Borders and Identity.” In: Humphrey TONKIN and Maria Esposito FRANK (eds.), The Translator as Mediator of Cultures.. John Benjamins: Amsterdam/Philadelphia, 2010, pp. 89-106.

ESPÍRITO SANTO, Janandréa. Laços da Tradução: as Versões em Língua Espanhola de Laços de Família, de Clarice Lispector, em um Estudo Baseado em Corpus. Diss. Universidade Federal de Santa Catarina, 2011. Available at http://www.pget.ufsc.br/curso/dissertacoes/Janandrea_do_Espirito_Santo_-_Dissertacao.pdf. Accessed April 16, 2014.

ÉTIEMBLE, René. Essais de littérature (vraiment) générale. Paris: Gallimard, 1975.

EVEN-ZOHAR, Itamar. “Polysystem theory”. Poetics Today. 11.1 (1990), pp. 3-26.

GONÇALVES, Fabiano Bruno. Tradução, Interpretação e Recepção Literária: Manifestações de Edgar Alan Poe no Brasil. Universidade Federal do Rio Grande do Sul, 2006. Available at < http://www.lume.ufrgs.br/handle/10183/7123>. Accessed April 16, 2014.

GRUTMAN, Rainier. Des langues qui résonnent. L’hétérolinguisme au XIXe siècle québécois. Montréal: Fides/CÉTUQ, 1997.

HALLIDAY, Michael A. K., Angus MCINTOSH, and Peter STREVENS. The Linguistic Sciences and Language Teaching. London: Longman, 1969.

HANES, Vanessa L. L. "Desvendando um mistério: a canonização de Agatha Christie no Brasil durante os séculos XX e XXI". Tradução em Revista, 18:1 (2015): 134-147.

HATIM, Basil, and Ian MASON. Discourse and the Translator. London: Longman, 1993.

HERMANS, Theo. “The Translator’s Voice in Translated Narrative.” TARGET International Journal of Translation Studies, 8:1 (1996): 23-48.

HOBSBAWM, Eric. “Language, Culture and National Identity.” Social research, 63.4 (1996): 1065-1080.

LAMBERT, José, and Hendrik van GORP. “On Describing Translations”. In: Theo HERMANS (ed.), The Manipulations of Literature Studies in Literary Translation. London & Sydney: Croom Helm, 1985, pp. 42-53.

LEROI, Chantal. Agatha Christie en Français ou la Traduction du Roman Policier. Diss. Katholieke Universiteit te Leuven, Leuven, 1978.

MARTINS, Marco Antonio. A colocação de pronomes clíticos na escrita brasileira: para o estudo das gramáticas do português. Natal: EDUFRN, 2012.

ONG, Walter. Orality and literacy: the technologizing of the world. New York: Routledge, 2002. E-book.

SAMPAIO, Maria de Lourdes R. M. História crítica do género policial em Portugal (1870-1970): transfusões e transferências, 2007. Vol. 1, 731 f. Ph. D. thesis (Doutorado em Literatura Comparada – Programa de Pós-Graduação em Literatura), Universidade do Porto, 2007.

SCHEI, Ane. A colocação pronominal do português brasileiro: a língua literária contemporânea. São Paulo: Humanitas/FFLCH/USP, 2003.

SCHIAVI, Giuliana. “There is Always a Teller in a Tale.” TARGET International Journal of Translation Studies. 8:1 (1996): 1-22.

SIMON, Sherry. “Translation, Postcolonialism and Cultural Studies.” Meta : journal des traducteurs / Meta: Translators' Journal. 42.2 (1997): 462-477. Available at http://www.erudit.org/revue/meta/1997/v42/n2/004153ar.pdf. Accessed April 17, 2014.

STORM, Marjolijn. A Corpus-Driven Analysis of Translations of Agatha Christie’s Detective Novels into Dutch and into German. Diss. University of Birmingham (forthcoming).

TOURY, Gideon. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins Publishing Company, 1995.

VITÓRIO, Elyne Giselle de Santana Lima Aguiar. “As construções existenciais com ter e haver: o que tem na fala e o que há na escrita.” Domínios da Linguagem, 7.2 (2013), 71-89. Available at http://www.seer.ufu.br/index.php/dominiosdelinguagem/article/view/23165. Accessed March 10, 2014.

WYLER, Lia. Crônicas, Poetas e Bacharéis: uma Crônica da Tradução no Brasil. Rocco: Rio de Janeiro, 2003.