This article investigates the transfer of meaning in literary translation and presents the findings of a recently concluded study (Botha 2015). The article fulfils a two-fold purpose. The first is to prove that a literary translation may be well written and may reflect the content of the source text correctly, but may still be inadequate due to the fact that not all types of meaning have been translated adequately. This is proven by comparing Elsa Joubert’s renowned novel Die swerfjare van Poppie Nongena (1978) with its English translation, The long journey of Poppie Nongena (1980), to demonstrate a significant shift in meaning due to an insufficient transfer of social dialect and informal register. The second purpose is to propose and illustrate a solution to the identified problem, which involves a discussion and an application by retranslation of Juliane House’s (1977) strategy of overt translation, which entails a foreignisation of the translation in order to make the source culture and the situational context apparent. The retranslation illustrates an alternative translation strategy, but does not suggest that this is “the ultimate” alternative. It is simply one which seeks to retain a crucial aspect of the source text and is also open to critiquing and alteration.
The reason for using an actual translation for the research was to demonstrate a general problem with reference to a real translation and the aim was not to comment merely on an individual translation. Toury (2012:5) explains that “individual studies into translation are bound to yield isolated descriptions, an obvious result being a gradual accumulation of discrete pieces of knowledge”. The evaluation in this case is therefore not an end in itself, but a procedure for highlighting a general problem in translation practice, for which a practical solution is provided and demonstrated. In this way, theory and practice are combined to produce meaningful results and the strategies illustrated here can be applied in translations of a similar nature.
The texts used in this study were chosen very specifically, however, and require an introduction. Die swerfjare van Poppie Nongena is a biographical novel based on the life story of an Afrikaans-speaking black woman born in the early 20th century and her struggle under the apartheid regime. The book was an immediate success, receiving three major South African literary prizes the year after its release. It is characterised linguistically by its incorporation of many different language varieties, such as Orange River Afrikaans, township slang, informal language and antiquated language, which add a strong cultural flavour to the novel that no doubt contributed to its success. The English translation of the novel, undertaken by Joubert herself, was first published in 1980 as The long journey of Poppie Nongena (the 1980 American edition was renamed Poppie Nongena, and subsequent paperback editions in 1981 and 1985 were simply called Poppie). The English translation, intended for both an international and a local market, made the novel into an international success. However, in spite of the positive reception of the English version internationally, the translation shows some major deficiencies and one may go as far as saying that it is an inexact or weak representation of the original as far as cultural and social reflection are concerned.
An assessment of any translation as inadequate requires theoretical backing, of course, and the theoretical structure of this research consists, firstly, in a distinction between two types of translation: overt and covert translation, as defined by Juliane House (1977). Overt translation is obvious translation which is usually applied in the translation of literary texts that are culturally or historically bound. As was mentioned, it entails foreignisation, in which the original cultural context is made apparent in the translation. Foreignisation primarily involves strategies such as the retention of certain source text elements (such as words, grammatical structures, titles, names and idioms) and non-standard handling of grammar. This approach is in opposition to covert translation, which is usually applied to pragmatic texts and which aims to produce fluid, natural texts. Achieving fluidity and naturalness often involves large-scale adaptation of the source text and domestication of the source culture. The distinction between overt and covert translation forms the main theoretical structure for this research, since it advocates the type of translation to be used for a specific type of text. In this sense, it functions as a measuring device. It is also a solution to the identified problem, as is explained later.
The second major theory which supports this study involves a probe into semantics. In referring to the first purpose of the article, mention was made of the existence of different types of meaning. A definition of meaning and its relation to dialect and register formed a major part of the theoretical framework of the research. Linguistic meaning has been defined in many ways, but most models can be distilled into two categories at their most basic level, namely denotation and connotation (see Ogden and Richards 1946; Nida 1964; Newmark 1991; Chandler 2002). Denotation, simply put, refers to the so-called dictionary meaning of a word, and connotation to word associations. Whereas pragmatic texts usually require mainly the transfer of denotation, literary texts demand more than the transfer of content, since their function is not merely to impart facts. In literary texts, the way in which an utterance is expressed is often just as important to the expressed content, since the manner of expression renders valuable information regarding characters (through the use of dialect) and regarding the communicative context (through the use of register). In literary texts, therefore, dialect and register do not function merely as embellishing devices, but contribute significantly to the meaning of the original text. Mona Baker’s (1992:13–7) model describing linguistic meaning was used for this research, since it makes provision for the description of meaning that arises from the use of dialect and register under the category of evoked meaning. The degree of retention of evoked meaning thus constitutes the main assessment criterion according to which The long journey of Poppie Nongena was judged as inadequate.
The transfer of evoked meaning was further linked to Kitty Van Leuven-Zwart’s (2004) two general criteria for measuring translation quality. Van Leuven-Zwart proposes that translations be judged according to both creativity and faithfulness and criticises exclusively aesthetic (or creative) quality assessments of translations. In the case of The long journey of Poppie Nongena, the language shows literary elegance and would thus score well in terms of creativity. In terms of faithfulness, however, there is a large degree of departure. The comparison of evoked meaning in the two texts is thus conducted within the general category of faithfulness, which relates to the theory of overt translation. In terms of the practical analysis, House’s (1977) first model of translation quality assessment was used to determine the quality of Joubert’s translation. House’s model, which makes use of various language user and language use dimensions, is slightly adjusted and simplified, however, to focus only on the main problematic dimensions, namely social dialect and social attitude (expressed through informal register). According to House’s model, social dialect and informal register are firstly described in the source text and then compared with their occurrence in the translation. Based on this comparison, the quality of the translation is determined.
Using the theory of overt translation, Baker’s definition of meaning and Van Leuven-Zwart’s and House’s criteria for measuring translation quality, it could be concluded that the presence of a higher degree of lexical and structural formality in the translation led to semantic shifts in the representation of (i) the socio-economic and cultural identity of characters, and (ii) the informal relationships between characters. These shifts are not merely stylistic, but influence the meaning of the translation. These shifts were then corrected by a fuller application of House’s strategy of overt translation, illustrated in the presentation of sections of retranslated text. Reference is made to a fuller implementation of overt translation, since the overtness or covertness of a translation is not absolute and can be measured only relatively, in terms of degree. Joubert does employ foreignising strategies to a degree (especially in the retention of foreign words); however, overt translation is not carried out consistently and its effects are adversely affected, or even negated, by the formality and literary quality of the English translation. The presentation of retranslated sections of text functions as a verification or proof of the use of House’s theory as a solution to the problem of retaining evoked meaning. It illustrates strategies that can be used in the translation of other texts that are similarly culturally loaded and linguistically complex.
Thus, the problem of retaining evoked meaning in literary translation is explained by means of the analysis of a real translation and it is solved by the full application of overt translation strategies.
Keywords: covert translation; dialect; evoked meaning; faithfulness; literary translation; overt translation; register