Abstract

This article presents the preliminary results of a questionnaire-based reception study of some of the strategies and tactics translators opt for in subtitling when faced with the challenge of translating non-standard discourse. The study focuses on two main strategies – presented as strategy of centralization or normalization of discourse and strategy of decentralization in the present study – and several tactics identified as the most common in a descriptive study based on twelve Portuguese translations of Bernard Shaw’s Pygmalion and Alan Jay Lerner’s My Fair Lady. The strategies identified in this study are briefly presented in this article as well as the contextual factors and working assumptions behind them. This is followed by the discussion of the data collected in the reception study, which brought some of those assumptions into question.

Keywords: Translation of linguistic variation, literary translation, theatre translation, audiovisual translation, reception

1. Introduction

The use of linguistic varieties in a fictional product – a literary text, a film, a theatre play – raises important questions for the study of translation as such use is always embedded in the source text with pragmatic and semiotic significance. The translator faces a challenge in relation to the presence of a linguistic variety in the source text which reflects the close relationship between the speaker, the medium and the context in which it appears. Such linguistic variety evokes and explores extralinguistic knowledge (hierarchically organising both the varieties and the speakers) and becomes a moment of tension (Lane Mercier 1995; 1997), a ‘culture bump’ (Leppihalme 1997), as the linguistic elements are both culturally conditioned and socially regulated.

The main purpose of this paper is to present the preliminary results of an experimental reception study of subtitling which focused on investigating the impact of some of the strategies and techniques employed by translators when faced with the challenge of translating non-standard discourse. After a brief discussion of the function fulfilled by linguistic variation in fictional works, this article will revisit some of the strategies and techniques identified in a small corpus of twelve Portuguese translations (for print, stage and screen) of George Bernard Shaw’s Pygmalion and Alan Jay Lerner’s My Fair Lady. This will be followed by a discussion of the results of a reception study which focused on investigating the effect of some of those strategies and techniques on the viewers’ definition of the characters’ profile, power relations and overall interpretation.

2. Linguistic varieties in literary text

The creative use of linguistic varieties in literary dialogue contributes to informing the reader who is speaking and the circumstances in which he/she is speaking. This type of use proves to be a textual resource which defines the character’s sociocultural outline in addition to his/her position within the sociocultural fictional context. This element also leads to the stratification of the participants in the dialogue since based on extra-linguistic factors, the speakers tend to associate the standard variety (officially established as the correct language use) with greater prestige and to devalue all other varieties, which are culturally associated to peripheral geographic spaces and to lower sociocultural status. The degree of linguistic mimicry is thus mediated by the author’s aesthetic, narrative, thematic and stylistic objectives, but also by consideration for their readership/audience and factors such as legibility, intelligibility and medium (Blake 1981, 1995; Page [1973] 1988; Chapman 1994). The literary recreation of a linguistic variety is then based on a previous selection and is truly a ‘pseudo-variety’ (Rosa 2004), a fiction Olga Brodovich has labelled as ‘scenic dialect’ (1997: 26). When recreating linguistic varieties, the author, as well as the translator, resorts to sociolinguistic stereotypes which are known to form part of public knowledge, i.e., those which are associated with a subcode which the public understands easily (Blake 1981, 1995; Page [1973] 1988). For this reason, it is important to discuss the translators’ decision whether or not to recreate and the way in which he/she chooses to do so, as this decision can modify, or even subvert, the work’s system.

3. The Portuguese translations of Pygmalion and My Fair Lady

In a previous article (Ramos Pinto 2009), I discussed in detail the different strategies and techniques identified in the twelve Portuguese translations of George Bernard Shaw’s Pygmalion and Alan Jay Lerner’s My Fair Lady. This corpus included translations from the 20th Century (1945-2001) intended for print publication, theatre performance and subtitling, thus allowing the identification of the main strategies and techniques translators opted for when translating non-standard discourse. Additionally, it also enabled the unveiling of certain contextual factors mediating such a choice, namely censorship, the prestige associated with the written and oral discourse, literary tradition, readership/audience design, and legibility.

It is not possible (nor is it my intention) to present in detail all the aspects discussed in that previous study. However, considering that the results of that analysis formed the basis of the reception study which produced the preliminary results presented in this article, in this section we will briefly present some of the strategies and techniques identified. This will allow a better understanding of the results regarding the effect of these strategies and techniques on the viewers’ overall interpretation of the scene and on the definition of the characters’ profile and power relations.

The corpus analysed includes the following translations:

Table 1: Portuguese translations of Pygmalion and My Fair Lady for print, stage and screen

The use of a particular substandard variety of British English – cockney – is central to the plot of both Pygmalion and My Fair Lady, which can be summarised as the teaching of a standard variety by a phonetician, Prof. Higgins, to a common flower girl and cockney speaker, Eliza. The use of non-standard features, perceived to be representative of a low sociocultural group, unveils the characters’ social peripheral status and low educational level. These features serve the communicative purpose of indirectly distinguishing Eliza’s character, depicting her as belonging to a low social status group and possessing a low level of education.

The first challenge faced by the translator is the asymmetry between the Portuguese and English diasystems. In one article, the translator of the 1972 translation refers to the difficulties of this translation, stating that there is no Portuguese low-urban-class variety corresponding to cockney[1]. This situation motivated the translator’s attempts to create a pseudo-variety using a combination of features from different dialects and which would generally be evaluated as ‘popular’, ‘uneducated’, sometimes ‘incorrect’, by the Portuguese readership. But what did other translators opt for?

The published translations opted for preserving the linguistic variation in the TT through the use of features familiar to the TC reader and which would generally be perceived as indicative of a low level of education and low social status. However, the presence of non-standard features is much lower in the translations completed prior to 1974[2]. In this context, forms of address become an important way of distinguishing the participant’s social status and oral discourse features, namely ellipses and contractions, which are used to mark the discourse as non-standard. There is a visible preference for lexical and graphic features (the so-called ‘eye-dialect’[3]) witnessed through the loss of morpho-syntactic features; however, translations with a higher standardisation of discourse display a greater presence of lexical features, while translations which emphasise the non-standard nature of the discourse display a greater presence of graphic features.

The translations intended for theatre performance also convey the decision to preserve the linguistic variation, accomplished through the use of features familiar to the TC public and generally perceived as denoting low educational level and low social status. Furthermore, as identified in the published translations, theatre translations also utilise oral discourse features as non-standard features. When compared to the published translations, there is a verifiably greater presence of both non-standard and oral features in the theatre translations. Accordingly, the percentage of standard discourse is much lower, not only in the published translations but also in the source text. In relation to the textual-linguistic features, one should note the higher percentage of morpho-syntactic marks, which are almost non-existent in the published translations, and the use of several types of graphic marks to indicate changes in the vowel quality, monothongisation, and metathesis or nasalisation of the vowel at the beginning of a word.

In the subtitling of the films under consideration, it is possible to distinguish between two groups and their respective illustration of two different strategies. The first group, comprised of 1987 and 1994 translations for RTP, the state-sponsored television channel, and a second group made up by the 1995 and 1996 translations for SIC, a private channel. In the first group, the discourse has been highly standardised: graphic non-standard features are absent and most ST features of oral discourse are omitted, such as ellipses, interjections, contractions, etc. The few remaining non-standard features are lexical features and it could even be said that social differences are only understood through the use of certain forms of address. In the second group, there is evidence of the use of features generally perceived as indicative of low educational level and low social status, as well as oral discourse features used as non-standard features. As with the published translations, even though morpho-syntactic features are almost non-existent, there are a higher percentage of lexical and graphic features in relation to the rates of the first group.

Even though all the translations, regardless of the medium, opted for the linguistic variation present in the source text, it is possible to distinguish between two main strategies. On the one hand, a strategy of standardisation[4] of discourse in which the variation is restricted to lexical features and forms of address and on the other hand, a strategy that preserves the variation without standardising the discourse.

The standardisation strategy can be identified in the book translations published before 1974 and in the subtitles broadcasted in state-sponsored TV. This can lead to the conclusion that the effort of keeping the high level of standard written Portuguese might be motivated by the pressure of censorship, in the case of books published, and by the condition of public channel defining itself as public service, in the case of subtitles. The preference for lexical features instead of grammatical or graphical features, along with the use of italics to highlight the ‘deviant’ forms, support the conclusion that translators are conscious of the importance of the non-standard discourse in this play/film, but want to preserve a high degree in writing.

However, the avoidance of graphical features in particular and the strategy of standardisation in general were probably motivated by another factor, namely, concerns for legibility. Regarding the published translations, it is important to note that, with the exception of the 1966 translation, these translations were published in very popular collections[5] aimed at a middle-class readership with poor reading habits and unacquainted with the ‘eye-dialect’ tradition. In the case of subtitling, it is important to remember that this strategy and techniques are followed in accordance with the traditional subtitling practice of avoiding legibility issues presumed to occur when orthographic norms are not followed. Despite the lack of empirical data related to this matter, translators follow the working assumption that subtitles (which address a very diverse public with different cultural understandings and reading skills) are not easy to read if they frequently present graphic features. This is especially true for the young (10-15 years old) and the elderly (55-80 years old), who would be the target public of a film, such as Pygmalion and My Fair Lady, broadcast at 2pm. Furthermore in relation to subtitling, one final consideration might be behind this strategy. Given the multimodal nature of the audiovisual product, the choice to not portray non-standard features could have been motivated by the assumption that the visual mode would provide viewers with similar information.

The strategy of preserving the variation without standardising the discourse might be interpreted as an effort towards adequacy to the oral register of the source text as well as adequacy to the oral discourse in the target culture. This is particularly noticeable in theatre translations, as these present a written translation which will essentially be brought to life via a different channel. In written translations, non-standard discourse can be conveyed through the use of oral discourse which, despite representing a deviation from the written norm, would not be considered as non-standard by linguistics. In theatre, given the nature of the spoken mode on stage, oral discourse features are interpreted by the audience as natural and expected and thus cannot be perceived to be deviating features as in written translations. In light of this, translators tend to make use of non-standard features more frequently. Even in relation to the influence of state censorship before 1974, we can see that censorship appeared to accept non-standard discourse on stage in oral discourse, even though it did not allow it in published written discourse.

As previously stated, the translations published later and in the subtitling broadcast on the private television channel demonstrate the use of what is called ‘eye-dialect’, possibly with the intention of improving the public’s reaction. In the case of subtitling, where the source and target texts appear simultaneously, translators have to face the fact that there is always someone, nowadays a large majority of the viewers, who understands the source language, a risk described by Gottlieb as ‘feedback effect’ (1994: 105). If the inclusion of oral or written non-standard features can, in fact, bear the risk of the translation being considered a bad one (Lefevere 1992:70), nowadays, the contrary is equally valid as the public who understands the source text is normally very critical if the subtitles omit discourse characteristics recognised in the source text. The same could be said in relation to the book collection translations, not only because this is a very well-known play, but also because one of the translations was published in a bilingual edition. We can thus reach the conclusion that the decision to employ a strategy which preserves the variation can be seen as an attempt to produce an accurate and adequate translation of the source text. In the case of subtitling, since this tendency has been noted in the subtitled version aired by a private TV channel, our data appears to confirm the hypothesis put forward by Rosa (1999; 2001) namely, that private companies may feel less responsible for upholding the standard.

4. Between standardisation and non-standardisation. Are viewers getting the message?

The data collected from the descriptive study, presented in the previous section, together with the information collected through a series of interviews conducted with professional translators, enabled the identification of certain operating assumptions motivating the choice between a strategy of standardisation or non-standardisation, including those hypothesised in the previous section. It was also evident that translators form their working assumptions without empirical testing. Owing to the lack of empirical data relating to the effect of translation strategies on viewers’ interpretations of the text (as well as data on viewers’ assessments of those same strategies), translators will continue to be left to their own devices and will replicate subtitling practices which might be inadequate, out-dated and unfulfilling of the full communicative purpose.

In this final section, we present the preliminary data of a reception study aimed at analysing the effect of complete standardisation and non-standardisation strategies on the reception of subtitled films. This study was devised primarily to answer the following questions:

- Are the viewers able to access the communicative meaning of the linguistic varieties if there is no recreation of such variety in the subtitles, i.e., if the discourse is standardised?

- Does the level of competence in the source language have an impact on the viewers’ ability to interpret the communicative meaning of the linguistic varieties? Will competent speakers of the source language be able to recognise the linguistic varieties and their communicative meaning solely through the spoken mode?

- Given the established tradition in the written mode, how will viewers react to a deviation from the norm? Will the viewer interpret a deviation from the norm as a bad translation?

The study involved 48 participants, aged between 19 and 45, divided into two groups according to their level of competence in the source language. Two film excerpts – of 3 minutes each – from the film Educating Rita (1983) directed by Lewis Gilbert were used in this experiment. The film tells the story of a working-class English girl, Rita, who decides to join the Open University in an attempt to improve her level of education and her life, and the friendship that develops between Rita and her tutor, Frank, a highly-educated professor and poet disillusioned with his students and life in general. In excerpt A the subtitles followed a standardisation strategy and presented subtitles devised in accordance with the written norms of standard Portuguese. In Excerpt B the subtitles followed a non-standardisation strategy and recreated the linguistic variation identified in the source product both at the semantic and orthographic levels, as exemplified in Figure 1. The subtitling was produced by a professional translator with more than 15 years’ experience.

The semantic level recreation involved the use of certain vocabulary associated with a colloquial or ‘lower’ use of the Portuguese language. The orthographic level recreation involved the use of certain graphic marks. This is not possible in every language, but it is possible in Portuguese, as colloquial and dialectal discourse is associated with the ‘disappearance’ of certain vowels and consonants as illustrated in the following examples (the grey boxes show how it would be written in standard Portuguese):

Figure 1: Screenshots illustrating the rec reation
(at both semantic and orthographic levels) of a linguistic variety in subtitling

The study collected simultaneous data through the use of eye-tracking equipment and consecutive data through a questionnaire completed after viewing each of the two clips. In this article, only data collected through the questionnaire will be discussed.

4.1 Excerpt A: Standardised subtitles without recreation of the linguistic variety

When asked to describe character Rita, 9 per cent of the viewers with high level of English (Group A) mentioned that the character spoke in dialect and only 3 per cent mentioned that she was from a lower social class. The remaining 88 per cent of viewers mentioned a myriad of aspects ranging from the characters’ age to the type of clothes she was wearing. In Group B (viewers with a low level of English) none of the viewers mentioned her social status, her educational level or any particular aspect regarding her speech. The fact that the linguistic variety or the social class were not mentioned could not be taken by itself as a sign of the viewers’ inability to interpret the communicative meanings of those elements, but it is certainly a sign that it was neither something of primary importance to the viewers or as central to the storyline.

The communicative meanings expressed in Rita’s speech were in fact something that most of the viewers appeared to have grasped, as 67 per cent (Group A) and per cent (Group B) answered ‘working class’ when asked directly about the character’s level of education and social class. There were, however, a considerable number of viewers (33 per cent in Group A and 50 per cent in Group B) who were not able to assess the character’s level of education and social class correctly. This seems to indicate that a high level of competence in the source language does not guarantee the identification of linguistic varieties and, consequently, the interpretation of their communicative meaning. This questions the operating assumption that a good knowledge of the source language supports the choice of a less conspicuous standardisation strategy in this context. When one third of the viewers are unable to identify or interpret the linguistic variety and its communicative meaning as a result of a standardisation strategy, questions can be raised regarding the adequacy of such a strategy.

Given the multimodal nature of the audiovisual product and the fact that there might be a certain overlap in the meaning expressed through the different modes, the viewers were asked about the basis for their assessment of Rita’s character as possessing a low educational level and low social status. The fact that none of the viewers were able to identify Rita’s discourse as dialect highlighted that this group of viewers had been able to infer the linguistic variety’s communicative meaning through other modes asides from the verbal one. This was confirmed by the viewers themselves, who admitted to having inferred the meaning from elements such as Rita’s clothes and the second character’s actions of tearing down a wall with a large hammer. This was certainly a surprising result, as such action are not normally perceived as indicative of social status, but the viewers’ interpretation followed the logical reasoning that if the character was tearing down the wall himself, it was due to the fact that he could not afford someone else to do it for him; consequently, the viewers inferred that he was neither rich nor from a high social class and he possessed a low level of education. As the viewers had established that the characters in the excerpt were a married couple, the interpretation of the male character was transferred to the female character, Rita.

Therefore, it seems possible to conclude that other modes besides the verbal one played an important role in enabling a correct interpretation of this scene due to the overlap of meanings being expressed via the different modes and the proximity of the source and target cultures, which allowed the viewers to interpret the characters’ actions as common to one social class in particular.

It is, however, important to bear in mind that such an interpretation was only possible due to two elements: the overlap of meaning and the proximity of the Portuguese and English cultures. Had there been a lesser degree of proximity between the source and target cultures, the viewer would a) not have been able to proceed with the same type of reasoning (as the source product would present social practices not shared by the viewer) and would not have been able to assess the communicative meaning of the linguistic variety or b) would have made a false interpretation. As  a confirmation of this, 50 per cent of viewers from Group B were unable to infer the communicative meaning of the linguistic varieties from the moving images and sounds. It is also possible to assume that by not making this information available in the subtitles, the viewer assigned greater importance to other modes and extracted information from elements that would have only overlapped indirectly with the meaning expressed by the presence of a non-standard variety.

Another relevant and surprising result was that, when asked directly to characterise the character’s speech, 45 per cent of the viewers in Group A answered ‘standard’, 45 per cent answered ‘dialect’ and 10 per cent answered ‘I don’t know’. In Group B, 100 per cent answered ‘standard’ bearing in mind that 33 per cent of the viewers in Group A demonstrated that they had not understood the linguistic variety’s communicative meaning, this means that 12 per cent of the viewers were able to identify the speech as dialect, but were not able to extrapolate its communicative meaning. It is possible to assume that this percentage would be even higher if the source and target cultures had less in common.

Thus, it is important to bear in mind that even though the standardisation strategy might not result in a loss of meaning, its use relies on the viewers’ ability to extract those meanings from other elements, which is only possible if certain conditions are met. A good level of competence in the source language as a facilitating factor does not seem to guarantee the identification of the linguistic variety and its communicative meaning. Moreover, were there no overlap of meanings between different modes, the meaning most probably would have been lost or, perhaps worse, it might have been interpreted inaccurately.

4.2 Excerpt B: Non-standardised subtitles with recreation of the linguistic variety

The identification of the characters’ speech as non-standard seems to have been achieved much more easily when the linguistic variety was recreated in the subtitles. When asked if they would characterise Rita’s speech as ‘standard’, ‘dialect’, ‘speaking badly’ or ‘I do not know’, none of the viewers in Group A answered ‘standard’, 58 per cent answered ‘dialect’, 18 per cent answered ‘badly’ and 24 per cent answered ‘I do not know’. In Group B, none of the viewers answered ‘standard’, 31 per cent answered ‘dialects’, 56 per cent answered ‘speaking badly’ and 13 per cent ‘I do not know’. In this context, it is important to mention that, of all the viewers who had identified Rita’s speech as ‘standard’ in excerpt A, 33 per cent (Group A) and 31 per cent (Group B) identified it as ‘dialect’ when watching excerpt B in which the subtitles recreated the linguistic variety. The fact that all the viewers understood that Rita was not speaking standard English and that more viewers were able to identify the speech as dialect, appears to support the assumption that the recreation of linguistic varieties in the subtitles contributes to a more accurate interpretation and recognisation of the communicative meanings at play. The answers provided in response to the questions on interpretation also supported these findings, as the viewers were able not only to identify the difference between Frank and Rita regarding their level of education and social status, but also to accurately interpret the power relationship established between them. When asked about the basis for their interpretation, all the viewers mentioned the speech as their primary source. The recreation of non-standard features in subtitling, therefore, appears to be a worthy strategy to consider using in situations where the viewer would most probably be unable to identify and interpret the communicative meaning of the linguistic varieties via other modes.

Support for the standardisation strategy is based on the assumption that the use of non-standard features would result in a higher cognitive load and impair the interpretation of the elements in the remaining modes. This assumption can be questioned in light of the fact that 100 per cent of the viewers in both groups correctly answered the memory questions designed to test iconic attention seems to allow us to start questioning that assumption. At this initial stage of analysis[6], it is difficult to draw conclusions on the cognitive effort required by the non-standardisation strategy but it does not seem to have compromised interpretation to the point of not allowing the viewers enough time to scan and interpret the elements in the moving images and sound modes (as assumed).

Even though the non-standardisation strategy has enabled a greater number of viewers to recognise Rita’s speech as dialect, it is important to discuss the considerable number of viewers who were still unable to do so: 18 per cent (Group A) and 46 per cent (Group B) answered that Rita was speaking ‘badly’, while 24 per cent (Group A) and 13 per (Group B) did not know how to characterise it. Important conclusions can be drawn from these results. Firstly, it confirms that one cannot underestimate the way in which a deviation from standard discourse and the written norm will first give rise to the speech being identified as strange and only afterwards as dialect. Secondly, it suggests that the type of recreation needs to be planned if it is to be interpreted as expected.

The impact on cognitive effort caused by different types of recreation is a topic in need further research, but the results collected in this study appear to support the conclusion that the use of easy-to-recognise and stereotypical dialect discourse markers leads to the smoother identification of the discourse as dialect.

Likewise, this also applies to the use of ‘neutral markers’ that cannot be identified with a specific dialect in the target culture. There is, however, an additional aspect mentioned by five of the participants in Group B which deserves attention – the fact that they were unable to identify a dialect and subsequently Rita’s speech simply looked incorrect to them. This suggests that the viewers answering ‘speaking badly’ were in fact accurately interpreting the communicative meaning at play, despite being unable to correctly express themselves. Finally, it is also important to mention that 10 per cent of the viewers in Group A did notice and criticise the fact that, in the final subtitle, the character was speaking in ‘alentejano’ (a well-known regional dialect from the south of Portugal), which they interpreted as strange and incoherent with the fact that the action took place in the UK.

5. Assessment

In addition to the cognitive effort evoked by the strategies of standardisation or recreation of the linguistic variation, as well as the viewers’ ability to interpret the different elements at play in the audiovisual product and their communicative meaning, this study was devised to evaluate how viewers would assess those strategies.

When questioned on the quality of the translation presented in excerpt B, 76 per cent of the viewers in group A answered that the translation was good and 21 per cent considered it to be a bad translation, while, in group B, 69 per cent considered it a good translation and 31 per cent considered it a bad translation. Those who assessed it as a good translation showed that they were able to distinguish between the translation process and the written mode, normally associated with the written norm, from which the subtitles deviated. The viewers who assessed it as a bad translation showed the opposite, as they were unable to dissociate one from the other. This is evident in their answers. When asked why they considered it a bad translation they always gave one of two answers: ‘because it is badly written’ or ‘because you don’t write/translate dialects’. On the contrary, the viewers who assessed the translation as good explicitly stated that their opinion was based on the belief that this was how the character spoke, even though it was badly written. This answer was, in fact, unexpected in relation to group B as most of the participants in this group had not been able to distinguish the character’s speech as dialect and were only able to interpret it as such through the subtitles’ recreation of it. More than just an association or dissociation between bad translation and bad speech, these answers reveal two different attitudes towards translation and the translator: one recognises the translator as a creative agent with the power to work creatively on the target code in order to better transfer the meanings in the source text, and the other perceives the translator as having a more submissive status, constrained by the established traditions for each mode and unrecognised for their ability to work the target code as deemed necessary in order to transfer the meaning expressed in the source product.

Conclusion

This article started by reviewing some of the strategies and techniques translators use when faced with the challenge of translating non-standard varieties as well as by discussing some of the contextual factors and assumptions behind those strategies. This analysis was the basis for the reception study presented afterwards which focused on the way in which the strategies of standardisation and non-standardisation impact on the viewers’ overall interpretation of a given scene, the definition of both characters’ profiles and the interpersonal relationship between characters.

Even though this article presents only the preliminary data collected by the questionnaires, this analysis has already pointed us in interesting directions regarding the initial questions which the reception study aimed to answer. Participants in both groups appeared to be able to access the communicative meaning of the linguistic varieties despite a large majority of the participants being unable to distinguish the characters’ speech as a non-standard variety. This outcome highlighted two important aspects: first, that the level of competence in the source language does not guarantee the identification of linguistic varieties and their communicative meaning; second, that the audiovisual product’s remaining modes can function as compensatory elements and enable viewers to access information not conveyed in the subtitle. However, it also led to another question that this study would like to leave for future investigation: in this particular case, the information expressed in the speech mode was in accordance with the information expressed in the visual mode, but what if that was not the case? Additionally, what if the viewers are unable to interpret the visual mode due to cultural distance, for example? It is thus important to keep in mind that the choice for a strategy of standardisation (based on the assumption that viewers will access the information lacking via the visual mode) depends on the viewer’s familiarity with the visual elements as well as the relationship established between the modes. Different relationships can be established between the modes and the elements in the images mode might only work as compensatory elements of a standardisation strategy once that relationship is confirmed.

None of the participants, when asked directly, identified the character’s speech as standard, showing that the identification of the character’s speech as non-standard was clearly easier when using a non-standardisation strategy. Additionally, viewers do not seem to associate the deviation from the standard/written norm with a bad translation and are able to distinguish between the translators’ efforts for accuracy and the ‘incorrectness’ of the written discourse, it is therefore imperative to question the subtitling practice of standardisation based on the assumption that recreating non-standard features will make the subtitles too difficult to read and leave the translation open to the viewers’ assessment of it as bad.

About the author(s)

I am currently the Director of Translation Studies at the University of Leeds specializing in audiovisual translation. Before my current appointment, I held the position of Marie Curie ITN Post-doctoral Fellow in the project TIME (2011-2013), as well as teaching positions at the University of Manchester (2009-2011) and at the University of Kent (2008- 2009). I hold a PhD in Translation Studies from the University of Lisbon (2010) but during my doctoral studies, I was a visiting fellow at the University College London (2006- 2009). Currently, I also collaborate with the Centre for Comparative Studies at the University of Lisbon in the context of the Translation Studies Research Group. Parallel to my academic work, I am an experienced freelance translator in the areas of audiovisual translation, theatre translation and academic writing. My research has been focused on audiovisual translation and theater translation as I am interested in investigating the construction of meaning in multimodal products. At the moment, I am particularly interested in examining the reception of subtitled products in order to better understand the impact of specific translation strategies in the audience. In this context, I am developing a series of empirical reception studies aiming at understanding the interplay between the different modes in the construction of meaning of subtitled products, testing current subtitling practices and exploring the possibilities brought by innovative practices of subtitling. I also maintain a continued interest in linguistic variation and the challenges it brings to translation as well as news translation.

Email: [please login or register to view author's email address]

•  Previous article