Self-translation: the case of João Ubaldo Ribeiro and his Sargento Getúlio / Sergeant Getúlio

Self-translation: the case of João Ubaldo Ribeiro and his Sargento Getúlio / Sergeant Getúlio

Maria Alice G. Antunes[1]

[1] Universidade do Estado do Rio de Janeiro (UERJ), ORCID 0000-0003-2858-4637


In this article, we discuss the work of self-translation of the Brazilian writer João Ubaldo Ribeiro who translated his novel Sargento Getúlio (1971) into English. Our analysis deals with Ubaldo’s choices for the translation of the general noun ‘coisa’, used in the original text as a form of treatment and in reference to the prisoner Getúlio takes to Aracaju. Based on the notions of autonomization and approximation (Britto, 1996), the results of our analysis suggest an apparent greater number of movements of autonomization but a also choice to to stay within the limits imposed by the original.

Keywords: self-translation; novel; original; autonomization; approximation


Neste artigo discutimos o trabalho de autotradução de João Ubaldo Ribeiro que verteu para o inglês o romance Sargento Getúlio (1971). Nossa análise se concentra nas escolhas de Ubaldo para tradução do substantivo “coisa”, usado no original como forma de tratamento e em referência ao prisioneiro que Getúlio conduz para Aracaju. Com base nas noções de autonomização e aproximação (Britto, 1996), os resultados de nossa análise sugerem certa prevalência dos movimentos de autonomização, mas também uma tendência do autotradutor à escolha pela permanência dentro dos limites do original.

Palavras-chave: autotradução; romance; original; autonomização; aproximação

Self-translation is an activity undertaken by a number of authors from different language backgrounds. Some of the most famous writers who have translated their own works are the Nobel Prize Winners Frédéric Mistral (1904), Rabindranath Tagore (1913), Karl Adolph Gjellerup (1917), Luigi Pirandello (1934), Samuel Beckett (1969), Isaac Bashevis Singer (1978), Czeslaw Milosz (1980) and Joseph Brodsky (1987). But they are not the only self-translators. Rainier Grutman states that “in today’s world, there are probably writers translating themselves on every inhabited continent” (2013, p. 2). Jan Walsh Hokenson and Marcella Munson, the authors of The Bilingual Text: History and Theory of Literary Self-translation (2007) also affirm that “a roll-call of self-translators would summon up the stellar figures of many literatures and languages” (p. 1). However, most of these self-translators’ work, that is, the self-translated text itself, has not deserved attention from many researchers throughout the world. Rather, it is the self-translators themselves – the “stellar figures,” so to speak – that have attracted the curiosity of most of the scholars interested in self-translation. We thus believe time has come for researchers to concentratre on the work of self-translators and on the choices they make when they translate their texts so that we understand the characteristics of the result of their work. Can self-translated texts be called new originals? Only research that focuses on the text translated by its author can shed some light on this question. This is therefore the focus of this paper.

The Brazilian writer João Ubaldo Ribeiro (1941–2014) translated into English two of his most popular novels, namely Sargento Getúlio (1971) / Sergeant Getúlio (1979) and Viva o Povo Brasileiro (1984) / An Invincible Memory (1989). His work as a self-translator, however, has not been frequently explored. In fact, little has been said about the actual characteristics of Ubaldo’s work, namely his self-translated novels. The present article thus aims at making another contribution to filling in such a gap in the history of self-translation in general and in the history of this popular Brazilian writer’s career. We will do that by presenting the results of a comparison between his acclaimed prize winning novel Sargento Getúlio and Ubaldo’s own translation of this novel into English: Sergeant Getúlio. We do not intend nevertheless to present an analysis of all of Ubaldo’s translation choices since that would be too great a task for the purposes of a single article. We will concentrate our analysis on the use of the noun coisa (thing being its most usual equivalent in English) and in the choices made by the Brazilian writer in his translation of this noun into English. The noun coisa is most frequently used to refer to an object, or an item, when you cannot or do not want to refer to the object or item in a more specific way. It is used in the Portuguese text when Getúlio, the main character, talks to and/or about the prisoner he is supposed to take to Aracaju, the capital city of the state of Sergipe. In the Portuguese text, coisa is therefore both a form of address and a noun used in reference to the prisoner, transforming him from a person into an object, or item, or a lifeless being. Does João Ubaldo Ribeiro choose its most usual equivalent, thing, to refer to the prisoner in the English self-translated text? Does he make other translation choices? Which choices does he make? What is the impact of his choices? Do Ubaldo’s choices turn the prisoner into a lifeless being in the English text? Our analysis is based on Paulo Henriques Britto’s concepts of autonomization and approximation (1996) and it is on these concepts that we will concentrate our discussion at this point.

Autonomization and approximation

In an article entitled “Translation and creation” (1996), Paulo Henriques Britto argues that “translating and writing are indeed qualitatively different activities” (p. 241, author’s italics) and proposes a processual definition of translation on the basis of a comparative analysis of his translation of the poem “Sunday Morning” by Wallace Stevens, and of his poem “Pessoana” (p. 243-250). Britto defines the main characteristics of the two processes central to his analysis: autonomization and approximation. 

From the examples presented by Britto in his article, autonomization is understood as a movement by which the translator (or the author) distances himself from the original – or the originals – when translating (or when writing, in the case of the author) a text. This distancing or departure may manifest itself in very different forms. Among them there is a selection of distinctive textual strategies, of varied stylistic resources or even of lexical items with distinct literal senses that allow for new interpretations. Approximation, on the other hand, encourages the translator to choose techniques that promote the recognition of a previously existing text since they represent a search for equivalence. It is, therefore, only natural that autonomization tends to predominate when an author writes a text, inasmuch as he or she seeks to get rid, so to speak, of the originals he had previously read. It is also natural that approximation predominates in the work of the translator since, as a rule, translators struggle to be faithful to the original. Approximation may prevail in writing, but this may result in plagiarism. Yet when autonomization prevails in translation, the result may be an adaptation, a new original or a betrayal.  In relation to his own creative works and translations, Britto points to the predominance of autonomization in the case of creation, while in translation “the structure is more or less balanced” (p. 250-251). He goes on to assert that the source text exerts a “controlling effect” (p. 251) and describes that, once he perceived his gradual departure from the original “Sunday Morning”, he discarded the movement so as to prevent the text from becoming excessively autonomous (p. 251). In writing “Pessoana”, the opposite occurred: once he realized the text was veering too close to an original (“Autopsicografia” by Fernando Pessoa), Britto discarded the movement and searched for other solutions to make “Pessoana”, a poem of his own, a more autonomous text. Therefore, he concludes that the original does not exert the “controlling effect” over the new text – the creation – as it did in relation to translation.

The relevance of Britto’s conclusions lies in the possibility to demonstrate that the “controlling effect” is not only exerted by the original. We consider that ideology, patronage, and the poetics discussed by André Lefevere (2007), play key roles in translation, regarded as a process that initiates at the moment a piece of work is accepted for publication in another cultural system. We also contend that the “controlling effect” exerted by the original over the creation may be seen as a kind of inside-out control. The writer must distance him or herself from an original and keep at a distance from it to avoid being accused of plagiarism if movements of approximation predominate in his or her writing. It is interesting to observe, however, that the attempt to keep a distance from a specific original does not ensure that the writer will actually move away from previously conveyed ideas or from other originals. It is our belief that it will not be possible to ensure the production of a text that is pulled apart from all the originals the author had contact with, since they constitute an integral part of his or her encyclopedic competence, which, either conscious or unconsciously, he or she resorts to during the original writing process. However, it may be asserted that a writer consciously tries to keep a distance from a previous original so as to produce a text that may seem autonomous at a first glance. Nevertheless, if in the case of writing autonomization surely is the expected movement, in the case of self-translation, the answer is not that simple and, furthermore, there is no single answer for that.

Research and articles on the work of self-translators, such as Samuel Beckett and Vladimir Nabokov (FITCH, 1988; CONNOR, 1989; COATES, 1999), indicate that there are movements in self-translations that demonstrate the author’s freedom – once freed from the control of the original text, the author may alter it significantly without being accused of betrayal, since he or she is seen as the one who has authority over his or her texts. On the other hand, it is undeniable that freedom is not absolute and that the original text exerts control over the self-translated one, otherwise professional readers would not recognize in self-translated texts the one that was previously written and (very often) published. On the other hand, João Ubaldo Ribeiro argued that the source text has a determining and controlling role in his self-translating process.

We shall now examine the extent to which this control occurs. Our analysis will be restricted to the novel Sargento Getúlio / Sergeant Getúlio. We will address the translation of the general noun coisa in an attempt to detect whether there is a predominance of either autonomization or approximation or whether “the structure is more or less balanced” (BRITTO, 1996, p. 250-251), in João Ubaldo’s self-translation.

Original and self-translated texts

Translations of Sargento Getúlio (1971) have been published in many countries around the world. Among them we may include: the U.S.A., Germany, England, France, Italy, Portugal, Spain, Holland, Sweden, Cuba, Hungary, Norway, Finland, Denmark, Russia, Israel and Canada. The publication of the novel, however, did not make of Ubaldo an acclaimed author in international literary systems. As the author himself stated in an interview, “o país onde [seus] livros têm maior aceitação é a Alemanha” (CADERNOS, 1999, p. 41). Despite the lack of success abroad, the publications in different countries and languages have guaranteed the writer’s acclaim in the Brazilian literary system. The writer is, for example, a member of the Academia Brasileira de Letras.

Sargento Getúlio has been described as a novel in which language plays a central role. In the introduction to its North-american edition, Jorge Amado states that “the language in Sergeant Getúlio, artistically molded on the speech of people, is often terse, hard and cruel” (1978, p. xi). Also, Amado argues that “this language gives very proper expression to the strong substance of the book” (p. xi). Rodrigo Lacerda refers to the impressive use of language in Sergeant Getúlio (2005, p. 70) and he lists several examples of this use that make Getúlio’s monologue striking. Some of them are the intentional mistakes, the onomatopoeias, the archaic expressions and syntactic structures, the complex neologisms, the regional accent, the intentional repetitions and, many times, the combination of these effects (p. 72). We will add to Lacerda’s list the creative use of words. This technique is described in this paper as the use of a term that activates meanings which are different from the ones listed in a dictionary. That is the case of the general noun coisa, a frequent authorial trace of Sargento Getúlio.

Both the general noun coisa and its most frequent equivalent in English, thing, can be classified as general nouns, that is, they belong to the category described as “a small set of nouns having generalized reference” (HALLIDAY AND HASAN, 1976, p. 274). The noun is generally used informally and the Houaiss Dictionary of the Portuguese Language lists twenty-one different possible meanings for it. None of those meanings, however, is the meaning that Ubaldo chose to give it, consciously or unconsciously, in Sargento Getúlio. In the Brazilian novel, coisa refers to a human being, that is, the prisoner that is being taken to Aracaju, whose name is never mentioned in the novel. We should emphasize Ubaldo’s unusual choice since the list of meanings in the Houaiss Dictionary does not include the reference to a human being or the use of the noun as a form of address as possible uses of coisa. We shall now examine some examples of the use of coisa both in reference to a human being – the prisoner – and as a form of address to the same prisoner. We include here examples of the Portuguese original and of the corresponding English translation.

(1) Achei que ia estrompar as gengivas do coisa. Acho que vai estrompar suas gengivas, coisa. (RIBEIRO, 1971)
(1a) I thought they would be likely to wreck his gums. “I think they are going to wreck your gums, thing.” (RIBEIRO, 1979)

(2) Baixei o capuz e botei a cara no lume e dei um arrasto no coisa, vem traste, só sabe gemer por baixo dessa mordaça, [...], tome a bença do padre, estava todo meio abestalhado com a situação [...]. (RIBEIRO, 1971)
(2a) I threw back the hood and put my face to the light and gave the creature a pull, come here junk, all you know how to do is to groan under that gag, […], ask for the priest’s blessing. The creature was a little stupefied with what was going on […]. (RIBEIRO, 1979)

(3) Aquela força que vem, coisa, aquela força que vem pelo rio atravessando, [...] ... E só vem fardado, veja bem, coisa, não vem um paisano [...]. Antes que eles queiram me acabar, coisa, eu ainda sou capaz [...]. Não vejo nem a cara, coisa, [...]. Aquela for­ça, aquela força, coisa, é uma fraqueza, [...], o que vosmecê nunca fez na vida, TREMPE, aquela força é uma fraqueza, [...], espie aí, coisa, [...], essa é uma terra de macho, viu, traste, [...]. (RIBEIRO, 1971)
(3a) That party that is coming, creature, that party that is com­ing across the river, […]. And there are only men in uniform, look well, creature, there is not one single civilian[…]. Before they finish me, creature, I am still likely […]. I don't even want to look at faces, creature, […]. That force, that force, creature, is a weakness, […], which is something you never did in your life, trash, that force is a weakness. […], take a look, creature, […], this is a man's land, hear junk, […]. (RIBEIRO, 1979)

(4) [...], oi coisa, olhe a vida, lá vem a força, [...], porque a gente não dá umas risadas, coisa? que é que está ven­do aí, coisa, o chão? isso tudo é um verdume só, coisa, [...], já cho­rou uma certa feita, coisa? [...], porque estou com um pouco de vontade de chorar agora, seu coisa, seu traste, seu trempe, possa ser que eu chore agora [...]. (RIBEIRO, 1971)
(4a) […] hey creature, look at life, there comes the force, […], why don't we laugh a little, creature? what are you seeing there, creature, the ground? this is all one big green­ness […], have you once upon a time cried, creature? […], because I feel like crying a little now, you thing, you junk, you trash, maybe I will cry now, […]. (RIBEIRO, 1979)

(5) Bem, criaturo: se fizer efeito no Município ... (RIBEIRO, 1971)
(5a) Listen good, creature: If you let your insides pour out in this county… (RIBEIRO, 1979)

As we read the examples, we are able to see that the general noun coisa is used in the ways we explained in the previous paragraphs. In the excerpts 1, 2, 3, 4 and 5, it is used to talk about the prisoner. In the excerpts 2, 6 and 7, on the other hand, it is used as a form of address, when Getúlio is talking to the prisoner. The result of both uses, however, is quite similar: the prisoner, a human being, becomes a worthless, lifeless, voiceless object dragged along the backlands of Sergipe towards its capital city.

As we read the choices inscribed in João Ubaldo’s self-translated novel, we notice that the writer selected two nouns to replace coisa: thing and creature. The selection of thing is probably the choice of the self-translator who respects the limits imposed by the original since its use causes an interpretation that the original coisa had caused: the transformation of a human being into an object. When he chooses creature, on the other hand, new meanings arise. In English, a creature may be “a) a lower animal; especially a farm animal; b) a human being; c) a being of anomalous or uncertain aspect or nature” (Merriam-Webster Dictionary). As we can see, it may arise in readers meanings which are quite distinct from those inscribed in the original by means of the use of coisa.

It is also essential to highlight the fact that while the noun coisa is used more frequently in the Portuguese text, in the English text, creature is the preferred choice. We would therefore easily conclude that movements of autonomization are more frequent in Ubaldo’s activity as the translator of his novel Sargento Getúlio, when we consider the choices for the translation of the general noun coisa. However, the noun criatura(o) is also present in the Portuguese original even if much less frequently than coisa. Therefore, we can say that it makes the choice of creature a possible one since João Ubaldo states that he does not want to go beyond the limits imposed by the original text.

Does Ubaldo keep his promise? Since the general noun thing does not include among its possible uses the reference to living entities in the same way criatura does, we may conclude that the English Sergeant Getúlio shows a prisoner who is a mixture between a voiceless, lifeless being and a living being whose appearance may frighten people. However, the living being tends to prevail. Therefore, our analysis of the self-translator’s choices regarding the general noun coisa shows that the movements of autonomization are more frequent.

Final considerations

In this article, we aimed at presenting our analysis of self-translation based on the case of the Brazilian writer (and self-translator) João Ubaldo Ribeiro. We considered the writer’s choices for the translation of the noun coisa, an important trace of the Portuguese text.

Our results show that the writer’s choices for the translation of the noun coisa stresses the fact that the prisoner is a living being even though this being can be considered strange to a certain extent. This transformation is due to the more frequent selection of the noun creature to replace coisa. However, as we stated before, João Ubaldo Ribeiro can be said to have remained within the limits imposed by the original text since the noun criatura, a more literal equivalent to creature, is also present in the original in Portuguese.

Finally, it is worth pointing out that it is the detailed analysis of self-translated texts that will contribute to our understanding of the question: can self-translated texts be called new originals? In our opinion, the study of different cases of self-translation support our view that the activity must be regarded as a heterogeneous one. In other words, there is no single answer to that question. Self-translators around the world exhibit varied behaviours which appear to be due to the conditions that surround writers that undertake the task of translating their own texts (ANTUNES, 2009, p. 248).

The study of self-translation based on the case of João Ubaldo Ribeiro shows readers that the limits of the original text control the choices of the self-translator.



ANTUNES, M. A. O Respeito pelo Original: João Ubaldo Ribeiro e a Autotradução. São Paulo: Annablume, 2009.

GRUTMAN, R. Beckett and Beyond. Putting Self-Translation in Perspective. Orbis Litterarum, Malaysia, p. 188-216, 2013.

HALLIDAY, M. A. K. & HASAN, R. Cohesion in English. Londres: Longman, 1976.

HOKENSON, J. W. & MUNSON, M. The Bilingual Text. History and Theory of Literary Self-translation. Manchester: St. Jerome Publishing, 2007.

RIBEIRO, J. U. Sargento Getúlio. Rio de Janeiro: Nova Fronteira, 1971.

_____________ Sergeant Getúlio. Boston: Houghton Mifflin Company, 1979.


Submetido em 31/05/2018; Aceito em 31/08/2018

289 visualizações.


  • Não há apontamentos.


A Revista Brasileira de Literatura Comparada está indexada nas seguintes bases:

Revista Brasileira de Literatura Comparada, ISSN 0103-6963, ISSN 2596-304X (on line)

Licença Creative Commons
Esta revista utiliza uma Licença Creative Commons Atribuição 4.0 Internacional (CC BY 4.0).

Wildcard SSL Certificates