Abstract

This article focuses on the analysis of technical constraints and critical points in the voice-over translation of a single episode of The West Wing, an American TV series set in the White House. The aim was to investigate how the mode of translation (voice-over for fiction genres) with its numerous constraints influences the translator’s decision-making. We first performed a quantitative analysis of technical aspects, including time constraints, text reduction and the quality of the recording. We found significant target text reduction (by 31 percent) and higher lexical variety of the target text, which suggests removal of repetitions and oral discourse markers. In the qualitative analysis we focussed on such critical points in the translator’s decision-making process as culture-specific items, metaphors, or irony, and tried to evaluate the decisions in the context of the observations made in the quantitative analysis. We found normalization to be the most frequently applied strategy in translating culture-specific items. This means that administrative and military terms from the American culture were frequently replaced with culture-free words, which led to some loss of the cultural context. Our analysis shows that in voice-over translation for a fiction genre the translator has to satisfy technical constraints first and only then can he strive to find the best solutions to the critical points created by the original text.

Keywords: voice-over narrator, culture-specific items, decision-making in translation, fiction genre, technical constraints, critical points

1. Introduction

On the audiovisual translation map presented by Gottlieb (1998), Poland was identified as a voice-over country, that is a country with voice-over as a predominant mode of audiovisual translation. This mode of audiovisual translation gained popularity in Poland in the previous decades mainly due to economic reasons. Nowadays, the majority of foreign audiovisual content broadcast by TV stations is still translated by means of voice-over – again, due to economic reasons, but also because viewers are used to and expect this type of translation. Subtitling and dubbing do exist – the former is the dominant AVT mode in cinemas, the latter is applied to animated films to facilitate their reception by young audiences. Alongside Russia, Bulgaria and some other Eastern European countries, Poland seems to use voice-over not only for documentaries (as is the case in some predominantly dubbing countries, such as Spain), but also for fictional genres, including feature films and TV series. The episode of The West Wing analysed in this special issue was thus translated by means of voice-over in Poland and as such was presented to the Polish audience.

The aim of this article is to analyse the translation of the said audiovisual material in the voice-over mode with a special focus on technical constraints and critical points in translation decision-making. We will base our analysis on a single episode of The West Wing entitled “In excelsis Deo”. We will identify the limitations the translator had to face in preparing a voice-over script for that episode resulting from both the technical characteristics of voice-over and the incongruence of American and Polish culture exemplified by certain culture-specific items (as well as from other aspects that may be challenging during the translation process). To that end, we will first present an overview of voice-over as a lesser known and much lesser researched audiovisual translation technique. We will then apply a quantitative and qualitative analysis to the translated episode to shed more light on the translator’s choices as determined by voice-over-specific limitations. The subsequent section will comprise a qualitative analysis of the translator’s decisions taken at critical points identified in the original text.

2. Voice-over as an audiovisual translation mode

Voice-over is considered to be the ‘ugly duckling’ of audiovisual translation (Orero 2006 after Woźniak, 2012) when it comes to both the number of research-based publications (for a detailed report see Franco et al. 2010) and the researchers’ opinions (see below). Voice-over as a term originally borrowed from Film Studies is defined in an abundance of ways (see Franco et al. 2010 for a detailed analysis), so it is necessary to specify what we mean by voice-over for the purpose of this paper. Díaz-Cintas and Orero define voice-over as:

a technique in which a voice offering a translation in a given target language is heard simultaneously on top of the SL voice. As far as the soundtrack of the original program is concerned, the volume is reduced to a low level that can still be heard in the background when the translation is being read. It is common practice to allow the viewer to hear the original speech in the foreign language at the onset of the speech and to reduce subsequently the volume of the original so that the translated speech can be inserted. The translation usually finishes several seconds before the foreign language speech does, the sound of the original is raised again to a normal volume and the viewer can hear once more the original speech  (Díaz-Cintas and Orero, 2006: 477).

Franco et al. 2010 add other important features to the above definition. Voice-over ‘is the revoicing of a text in another language, or a translating voice superimposed on a translated voice’ (2010: 23), is spoken in synchrony with original speech, recognisable words and actions (kinetic/action synchrony discussed in more detail below), derives from unedited material (production voice-over) or from edited material (postproduction voice-over); ‘can render content more closely to the original (voice-over translation) or less closely to the original (what the authors have decided to call free voice-over translation); reproduces mimetic features to a certain extent (accent, age, emotion, gender, intonation, orality markers, stress)’ (2010: 23). As we will see later, due to the specific use of voice-over in Poland some of these features do not directly apply to the Polish voice-over analysed in this paper.

In the context of Gottlieb’s classification of audiovisual translation modes, voice-over would fall predominantly into the isosemiotic category, that is it involves transfer between the same channels (the verbal auditory channel in the original including characters’ utterances transferred into the verbal auditory channel in the translation including the reading of the translated lines by the voice talent). However, it may also be diasemiotic, that is involving different channels since certain inscriptions, captions or letters shown to the viewer but not read out by the characters will be translated and presented orally by the voice talent, that is the verbal visual channel will be transferred into the auditory one. 

In the majority of countries voice-over is usually applied to the translation of programmes that come under the umbrella term of the factual genres (including news, documentaries, talk shows, debates, corporate videos, interviews, instruction videos, infomercials) (Franco et al. 2010). In Poland, voice-over (called wersja lektorska) is used predominantly in television for fiction genres as well, such as feature films and TV series. It thus differs from the voice-over used for non-fiction genres in the West.

3. Voice-over as an AVT mode for fiction films in Poland

In principle, Polish voice-over is delivered by one voice talent, irrespective of the number of characters and their gender. It is usually the male voice for fiction genres and either male or female for documentaries. In the rest of the analysis, we will focus on the characteristics of Polish voice-over for fiction films only. According to Polish scholars: ‘słowo dociera do widza podwójną drogą: w wersji tłumaczonej i częściowo interpretowanej przez lektora, oraz w wersji oryginalnej, która zostaje jednak znacznie wyciszona, stanowiąc w nowym przekazie zaledwie tło dźwiękowe’ [the word reaches the viewer in two ways: translated and partially interpreted by the voice talent, and in the original version with a significantly lowered volume so that it becomes just a sound background in the new message] (Hendrykowski 1982 after Tomaszkiewicz 2006).

Although the majority of Polish viewers express their preference for voice-over in the translation of audiovisual content in surveys and opinion polls (Bogucki 2004), Polish scholars shun this mode and consider it as inferior to subtitling and dubbing. Belczyk (2007) claims that low costs may be the only advantage. According to Garcarz: ‘Polscy widzowie (telewizyjni) nie wyrażają niestety gotowości do wprowadzenia wersji napisowej (dominującej w dystrybucji kinowej) na codzienny użytek w przekładach telewizyjnych’ [Polish TV viewers are unfortunately not willing to accept subtitles (that predominate in cinemas) as a translation mode on TV] (Garcarz 2006: 115). Tomaszkiewicz (2006) explicitly states that voice-over should not be used for fiction.

Similarly, voice-over for fiction genres is negatively viewed in Western countries. The New York Times asked through one of its headlines: ‘Why Is Marilyn Monroe a Polish Baritone?’ and described this audiovisual translation mode as follows:  ‘As actors and actresses open their mouths to speak, their words are drowned out by the voice of a seemingly omniscient Polish male off screen. Joan Collins’s put-downs on Dynasty are thus heard by Poles as a local baritone. Marilyn Monroe’s breathy come-ons in Some Like It Hot are heard as a deep monotone, and Jane Fonda’s seductive voice in Barbarella emerges as flat drone’ (Glaser 1991). The article even quotes Izabela Cywińska, former Polish Minister of Culture, thus explaining the predominance of voice-over: ‘It is a matter of laziness. The Polish people deserve better than these idiotic voices who invade films and characters’ (Glaser 1991). Regardless of personal views, voice-over seems to be an established mode of audiovisual translation in Poland and is starting to be regarded as an interesting object of scientific analysis. When approached without prejudice, it proves to be as complex as other audiovisual translation modes, subject to its own characteristic technical constraints that continue to pose challenges to translators.

Since the definitions of voice-over presented in section 2 pertain to voice-over as applied mainly to non-fiction genres while voice-over is used also for fiction genres in Poland, it is important to pinpoint some differences crucial for the understanding of the nature of Polish voice-over. There are also differences in the implementation of voice-over in countries that use it for films and TV series. For instance, there are typically two voice talents in Russian voice-over for films – a female voice for all female characters and a male voice for all male ones. In Poland, all lines are read by one male voice talent who in general does not reproduce mimetic features of the characters (such as accent, age, emotion, gender, intonation, orality markers, stress, as indicated above by Franco et al. [2010]). The voice is rather neutral and the viewers know that the characters are shouting or stuttering because they can hear the original soundtrack slightly decreased in volume in the background. In fact, if compared with Polish voice-over for films from decades ago, contemporary voice-over seems to leave the original soundtrack quite audible to make it an inherent part of translation (we elaborate on this rather thought-provoking comment in section 5.3). Other differences regarding various synchrony types will be discussed in more detail in section 5.

4. Methodology

The present study includes both quantitative and qualitative analyses of various aspects of the voice-over translation of one episode of The West Wing, the material that has been subject to manifold analyses in the present special issue of the journal. We are especially interested in seeing how the mode of translation (voice-over for fiction genres) with its numerous constraints influences the translator’s choices.

The Polish translation is authored by Alicja Mołoniewicz, produced by ITI Studio and commissioned by TVN, one of Poland’s nationwide private broadcasting corporations. The series was screened in Poland under the title: Prezydencki poker (back translation into English: Presidential poker). The first four seasons were aired by TVN and TVN7 as late night shows. The series was not very popular and subsequent seasons were not shown in Poland. The text of the translation was very difficult to access so we simply transcribed the episode available through one of Poland’s official online pay-as-you-go services offering television content. We thus arrived at a very small parallel bilingual corpus. Parallel corpora typically include original texts and their translations, thus lending themselves easily to direct comparisons between the source text and the target text (Bosseaux 2007; Kenny 2001). Our corpus included a text of the original scripted dialogues of the episode “In excelcis Deo” in English and its translation into Polish.

We first performed a quantitative and qualitative analysis of technical aspects of the VO, including time constraints, text reduction and the quality of the recording. Then we focussed on some critical points in the translator’s decision-making process, such as culture-specific items, metaphors, irony or grammar, and tried to evaluate the decisions in the context of the observations made in the first part.

5. Technical aspects of the VO

Since voice-over is under-researched it is usually considered less demanding for translators because they do not have to bother about lip-synchronisation (as in dubbing) or space constraints (as in subtitling). But, as we hope this study will show, voice-over sets various limitations that eventually determine the translator’s choices. This section will focus on time constraints in voice-over and various synchrony requirements it must meet (voice-over isochrony, kinetic synchrony and action synchrony). We will use the definitions provided by Franco et al. (2010) and identify differences in those requirements between non-fiction and fiction genres. Examples will be provided from the TV series in question.

5.1. Time constraints – isochrony

Time constraints are one of the most obvious limitations one can observe when watching voiced-over content. The spoken translation cannot be longer than the original lines. This parallelism between voice-over versions and their originals is referred to as isochrony (Franco et al. 2010: 121), or ‘the need to create a fluent translation that is going to be read aloud and which fits in the space available’ (Franco et al. 2010: 46). It is common practice to have seconds at the beginning and at the end of an utterance with no translation so that the viewer can hear the original soundtrack, the speaker’s voice, and so on. According to Orero (2004), the translation begins two seconds after the original and finishes a couple of seconds before the end of the original dialogue. According to Ávila (1997), the delay is approximately 4 seconds while Carroll (2004) presents a meaning-based approach and states that the delay might be equivalent to the unit of meaning (similarly to the delay in simultaneous interpreting during conferences). It also happens that the voice-over version might end simultaneously with the original or even be longer. Such post-synchrony is frequent in Polish voice-over. The translation typically starts a few seconds after the original and frequently ends later than the original.

5.2. Text reduction

The isochrony or post-synchrony requirement entails much reduction in the translation. As Orero and Matamala (2009: 25) rightly point out: ‘fluffs, hesitations, repetitions [are] eliminated from the target language version because otherwise it could not be understood.’ In fact, this makes the voice talent’s task easier on the one hand since no acting out is required and a redundancy-free text can be neutrally delivered. On the other hand, this means that the voice-over text aims to sound natural and spontaneous even if it is deprived of orality features.

The most frequently omitted elements from the original speech include (after Tomaszkiewicz 2006 and Belczyk 2007): nominative forms of address (Monica, you guys), phatic devices (oh, uhm, well, you know, you’re right, really?), repetitions, fixed situational phrases (welcome, goodbye, identification in a phone call), question tags (aren’t you?), false starters (He’s... I don’t know), terms of endearment (honey, darling), linking devices (because I don’t want it).

In the translated dialogue list of “In excelsis Deo” we have identified all of the above categories of text reduction. In addition, we have noted that most of the unfinished sentences have been omitted as well. For example I know you're not, but that doesn't... has been translated as Wiem o tym (‘I know that’). Also, some repetitions have been dropped, such as in the following exchange between Sam and C.J. when she finds out that her new secret service code is ‘Flamingo’ and when she thinks that it is a ridiculous looking bird: – You're not ridiculous looking. – I know I'm not ridiculous looking translated as  – Ty tak nie wyglądasz. – Wiem (‘You don’t look like that. – I know’). What is more, we have noticed quite extreme cases of condensation, for example I wasn't planning on doing that, but now that you suggest it has been translated as Niezły pomysł (‘Not a bad idea’). And finally, there were numerous cases where whole sentences were dropped, though without loss of meaning for a given exchange. For example, when the president is getting ready to do some last minute Christmas shopping one of his co-workers notes: I saw the black suburban in back. President's slipping away, huh? In the translation the first sentence has been omitted completely and the line rendered as Prezydent gdzieś się wymyka? (‘The president is slipping away?’)

We calculated some descriptive statistics to examine the reduction of text in the voice-over translation of The West Wing episode. We decided to compare the total number of characters rather than the total number of words since word lengths differ in the language pair under consideration. We deemed the number of characters a better measure of text length since we believe that, roughly speaking, the number of characters will be a better reflection of the time required to read a text than words that differ in length (English words are usually shorter than Polish ones and thus are faster to read; in fact, the actual mean word length in the English original was 4.09 characters and in the Polish translation – 5.08 characters). The total number of characters in English was 28,003 against 19,271 characters in the Polish voice-over version. Thus, the translation is approximately 31 per cent shorter than the original. If we consider that Polish is usually longer than English (since the words are longer and the average translation of an English text into Polish would be about 30 per cent shorter), this outcome certainly reflects various reductions that took place in the course of translation.

We also looked at a measure of linguistic variety, the type-token ratio (TTR), which is the ratio of different words that are used in a text (types) and all words in that text (tokens). TTR as used in a parallel corpus, like ours, can be used to ascertain the existence of one of translation universals – simplification (Laviosa 2003). We might expect a lower TTR value for the translated text, which would mean that the Polish translator used a smaller range of vocabulary and thus simplified the language (this measure obviously pertains to lexical simplification only, not to syntactic). Surprisingly, TTR for the English text was 21.18 and its STTR – standardised type-token ratio (measured to avoid corpus length as a confounding variable (Bosseaux 2007)) was 39.48. This proved to be lower than the measures for the Polish text: TTR – 40.24, STTR – 46.43. To explain the outcome, we have to take into consideration the language pair specificity. Polish is a highly inflective language, which means that all inflected forms of the same word will be treated as separate tokens. In English, this is not the case since inflection suffixes are few and far between. This difference alone can account for the higher TTR and thus higher lexical variety in Polish. Additionally, this may be also explained by the fact that the Polish translation is greatly reduced as compared to the English original and that reduction naturally applied to repetitions.

To investigate this potential explanation further, we compared the frequency of some content words in the corpus. This time, we compared the frequency of occurrence of lemmas (all inflections of a word are treated as the same word) (Kenny 2001: 34) rather than unlemmatised types (that is all infections treated as separate words)  to account for the structural difference between Polish and English. We selected two common nouns (prezydent – president, bezdomny – homeless) and four proper names (Josh, Toby, Leo, Sam). Table 1 shows that, with the exception of president, repetitions of these words were more numerous in the original rather than in the voice-over translation, thus further adding to the explanation of the higher type-token ratio in Polish and further indicating types of reductions performed by the translator.

Table 1. Occurrences of selected words in the corpus

It seems that there is much need for reducing terms of address, utterances with phatic function, greetings and so on in Polish voice-over. According to Hendrykowski (1984 after Garcarz 2006), one of the most serious sins committed by voice-over translators is the repetitive transfer of meanings that are included in the original version as fully comprehensible without translation. These meanings include, but are not limited to, names pronounced by the characters and greetings. Majewski (personal communication after Garcarz 2006) confirms that unprofessional voice-over scripts usually include repetitions of the original soundtrack elements that are audible and comprehensible to the voice-over version users. He states that translators working with him for the Polish public television have always been instructed to leave such elements untranslated and audible in the original (personal communication). 

5.3. Voice-over as voice-in-between

Given such substantial text reduction, as described in 5.2, the translated dialogue list of “In excelsis Deo” is much shorter than the original one. As a result, we can expect that when read out by the voice talent it will not fully overlap with the original speech. In fact, this opens up new possibilities for the voice-over mode. Woźniak proposes that ‘in order to achieve maximum invisibility and unobtrusiveness of voice-over, the  principle of superimposition should be replaced with that of juxtaposition: in other words, voice-over should be transformed into voice-behind or voice-in-between’ (Woźniak, 2012: 216). She suggests that the original soundtrack should be clearly audible and the voice talent ‘should (…) be able to deliver the text in pauses and gaps in the original dialogue, or – if this is not possible – to reduce the impact by leaving whole sentences or coherent parts of them audible’ (Woźniak 2012: 216) (though it should not compromise the audibility of the voice talent). In this way the viewers can compensate for condensation and reductions in the translation.

Inspired by Woźniak’s analysis of the voice-over of Star Trek (Woźniak 2012), we decided to have a closer look at the voice-over of The West Wing episode and determine which portions of the original soundtrack are audible. Because of the space limitations of the article, we limit ourselves to one fairly short exchange from the episode when the president is meeting with children before Christmas. In Table 2 the audible portions are highlighted in bold.

Table 2. Audible portions of the original soundtrack

As the analysis shows, substantial parts of the original are audible through the voice-over. The exchanges between the kids and the president are quite fast with few gaps to be filled in by the translation. The voice talent’s voice covers some of the original speech, however the original soundtrack is loud enough and the audible portions long and coherent enough for the viewers who know enough English to be able to make sense of them and to experience the acting not only through the visual but also through the aural.

5.4. Kinetic synchrony, action synchrony

It might seem that since so much text has been reduced in the VO of “In excelsis Deo” (see section 5.2) and since the translation is not read out in full synchrony with the original (see section 5.3), both action synchrony (a requirement that translation must correspond to the action shown on screen) and kinetic synchrony  (a requirement that translation must match body movements) would be severely compromised (see Orero 2006 quoted in Franco et al. 2010). However, no such cases have been identified in the case at hand. This is perhaps because there were no quick shot changes that would affect action synchrony and the delays in reading out the VO script were not significant enough to have an impact on kinetic synchrony.

In the sections above we have focused on the technical constraints of the voice-over mode. In the next section we move on to discuss some of the critical points in the translator’s decision-making process and how it may be affected by the technical constraints.

6. Critical points of decision-making in the translation

When translating a text the translator sometimes stumbles upon a word, phrase or even a whole sentence that poses a particular challenge and requires an extra effort and a conscious decision. Such stumbling blocks, which we call critical points of translation decision-making (cf. Munday 2010), may include – but are not limited to – culture-specific items (CSIs), proper names, idioms, allusions, word play, slang, appraisal, metaphors or even forms of address and certain grammatical structures. They coincide to a great extent with ‘translation crisis points’ which ‘constitute turning points at which the translators have to make active decisions, and these points are thus indicative of overall strategy and to what norms the translator professes’ (Pedersen 2005: 1). Critical points in translation also overlap to some extent with Leppihalme’s ‘culture bumps’ (Leppihalme 1997), though it should be stressed that her culture bumps are limited to cases of intertextuality in translation, a limitation which does not apply in our analysis. The very term ‘critical points’ with reference to the translator’s decision-making was borrowed by us from Munday, who applies this concept to evaluation and appraisal in translation (Munday 2010; see also Martin and White 2005).

In what follows we present our analysis of critical translation points in The West Wing episode: we identify such points, present the translator’s solutions (with back translation into English) to the selected problematic source-language items, and then discuss and evaluate some of them in the light of the technical constraints discussed above. We have divided the identified items into several categories including: CSIs, metaphors, word play, irony, forms of address and grammar.

6.1. Culture-specific items

As expected, the first and by far the largest group of critical points identified by us are CSIs. Culture-specific items (for example Aixela 1996), also known as ‘cultural words’ (Newmark 1988), ‘extralinguistic cultural-bound references’ (Pedersen 2005) or ‘culture specific elements’ (Leuven-Zwart 1989) are understood here as ‘extralinguistic references to items that are tied up with a country’s culture, history, or geography and tend, therefore, to pose serious translation challenges’ (Díaz-Cintas and Remael 2007: 200). These may include geographical items (for example place names), ethnographic references (for example to objects from daily life, work, art and culture or descent) and sociopolitical references (for example to institutions and functions, socio-cultural life, military institutions and objects)  (Díaz-Cintas  and Remael 2007).

As the issue of translating culture-bound terms is very well covered in the literature, we will not give an extensive overview of the topic. Instead, in Table 3 we list the main translation procedures for CSIs, as proposed by various scholars (though we are aware that given the number of existing classifications, the list is not exhaustive), grouped under seven umbrella terms of importation, calque, extra information, normalization, deletion, addition, substitution, along with an exemplary definition of each of such terms (cf. Kwieciński 2001). 

Table 3. Types of procedures for translating CSIs

When placed on a Venutian scale the first three procedures would be closer to the foreignisation end of the continuum, whereas the last four would be considered more domesticating (with importation being the most foreignising and substitution the most domesticating) (cf. Venuti 1995).

For ease of analysis we have divided the identified CSIs into four categories: the U.S. administration and White House, Washington D.C. realia, the U.S. military and general CSIs, all of which are discussed below. Next to each translated CSI we specify the translation procedure applied.

6.1.1. The U.S. administration and the White House

In a TV show about a fictional president of the United States and his senior staff one could expect a lot of terms related to the U.S administration and the White House. And in fact, in the episode concerned we have spotted seven such items, which are presented in Table 4.

Table 4. Translation of CSIs related to U.S. administration and White House

The first three items are related to staff positions or functions (secret service, Secretary of Labour, Press Secretary). Interestingly, the translator is not consistent here, as all of the items are rendered using a different procedure (normalization, substitution and calque). Similarly, the next three items related to departments or offices are also treated differently: the Labor Department is normalised, the I.R.S. (Internal Revenue Service) replaced by its Polish counterpart, while the V.A. (the U.S. Department of Veterans Affairs) is again normalised (‘veterans office’). As for the last item in the table, the Mural Room, the translator decided to drop this reference altogether. Admittedly, in reality there is no such room in the White House (there is the so-called ‘Diplomatic Reception Room’ or ‘Dip Room’, which serves the same functions as the Mural Room in the show and looks very much like it), however a certain ‘presidential’ feel seems to be lost here. This observation also applies to the other items in the group – by applying mainly domesticating procedures, the U.S. administrative reality is not preserved for the Polish viewer, though the use of such procedures is understandable given the severe time constraints.

6.1.2. Washington D.C. realia

As the show is set in Washington D.C. one would assume that it would include many references to the city’s topography and landmarks. Table 5 lists those identified by us in the analysed episode. 

Table 5. Translation of CSIs related to Washington D.C. realia

The first item, the Mall refers to the National Mall, an open-air national park area in downtown Washington, within which major landmarks are located, such as the Capitol, the Lincoln Memorial and the Washington Monument. It also includes the Korean War Veterans Memorial, which in the episode is the place of death of a homeless Korean War veteran. This reference is lost in the translation, though the viewers can make the link themselves, as the stainless steel figures representing a squad on a patrol are clearly visible in the background when we see the dead body of the homeless man lying on a bench. This is a very good example of how the visual and the verbal interact in the film and where something that is understated in translation can be inferred from the image.

The second item in the table is a reference to Washington’s topography: by the way this location is specified we can infer that what is meant here is an intersection of two streets: Capitol Street and P Street. This information is misrepresented in translation, as the location is specified as ‘around the Capitol’ (which – to make matters worse – is mistranslated as ‘Kapital’  (‘the Capital’), though it may be the case that the word is simply mispronounced by the voice talent). However, it should be noted that this mistake is a low-risk one (cf Pym 2004), as it does not affect comprehension of the scene. What is more, providing a more accurate translation would unnecessarily add to the length of the translation.

Finally, the third example, Georgetown, being one of the top U.S. universities and given that the show is set in Washington, is absolutely transparent for the American viewer, but is not necessarily so for the Polish viewer. That is probably why the translator decided to make this item more explicit for the Polish viewer by adding the word ‘studies’, thus making the translation more domesticating and transparent.  

6.1.3. The U.S. military

As has already been hinted at in the previous section, the episode at hand features a homeless Korean War veteran – Walter Hufnagle – who died of exposure on a bench at the Korean War Veterans Memorial (see above).  His body is shown to Toby Zeigler (whose coat – which he had given to the Goodwill – with his business card left in the pocket the man was wearing). Toby, who is a communications director at the White House, arranges for an honour guard funeral for the deceased man. In the episode we also hear a story of Mrs Landingham’s twin sons, who died in Vietnam. We therefore have a number of CSIs related to the U.S. military, which are listed in Table 6.

Table 6. Translation of CSIs related to U.S. military

In the first two examples the division in which the soldier served is translated by normalization and calque (though it must be noted that not all of the elements of the original term have been translated). The Purple Heart, which is a U.S. military distinction awarded to those who have been wounded or killed in battle, is also normalised, with the name of the medal omitted. The final item is lottery number, which refers to a draft lottery system introduced in 1969 to determine the order of call to military service during the Vietnam War. This reference is subject to normalization and as such is completely lost in the voice-over. Had the reference appeared in a book or an article, a translator could have provided a detailed description of the system in a footnote, and by the same token better familiarise the readers with this portion of U.S. history. Because of the constraints of the voice-over mode, however, the viewers are deprived of the opportunity.

6.1.4. Culture-specific items (general)

In the episode we have also identified a few general CSIs, which are enumerated in Table 7.

Table 7. Translation of general CSIs

We would like to draw the readers’ attention to three of them: eggnog, Hail Mary’s and the Keystone Kops.  Eggnog was translated using substitution (‘ajerkoniak’ – ‘eyercognac’). Although both drinks are made from similar ingredients (mainly eggs, milk, sugar and liquor), they are not identical. More importantly, eggnog has an important connotation, which ‘ajerkoniak’ lacks: in the U.S. eggnog is mainly associated with Christmas and the festive winter season. In the episode, which takes place around Christmas, Mrs Landingham’s comment about the president being allergic to eggnog makes perfect sense, which is not conveyed in the Polish rendition.

In the second example, Hail Mary’s were left in the original in the Polish version, which is strange, especially since Poland is a Catholic country where the Polish counterpart of the prayer (‘Zdrowaś Maryjo’) is very well known. Chances therefore are that the translator simply did not recognise that the name refers to a well-known prayer, the effect being that some meaning of the scene has been lost, as the fact that the adolescent assailants made their victim say the prayer while they tortured him only adds to their cruelty.

Finally, the Keystone Kops is an allusion to silent film comedies of the 1920s, which featured incompetent policemen. Nowadays this phrase is used to criticise a group of people for their lack of coordination and the ensuing mistakes. Again, due to the limitations of the AVT mode, this reference is lost in the Polish voice-over, however without any loss of meaning of the scene for the Polish viewer.  

Table 8 summarises the types of procedures used in the episode (in absolute numbers and percentages).

Table 8. Translation procedures used in the episode

It turns out that by far the most popular procedure in the translation of the CSIs in the episode was normalization (39 per cent), followed by importation (22 per cent). Interestingly, only one item was made more explicit by the addition of extra information (but it should also be noted that only three items were subject to deletion). Such findings could be explained by the constraints of VO (see section 5) – it seems that as time is of the essence, the translator cannot afford more elaborate procedures, such as adding extra information. The effect of the translation of CSIs in the analysed episode was thus more domesticating, with a range of cultural elements getting lost in translation and the viewer being deprived of the possibility to learn about a foreign cultural reality. 

6.2. Metaphors

In the analysed episode we came across a number of metaphors (presented in Table 9) which could be potential critical points in the translation process.

Table 9. Translation of metaphors

An interesting metaphor – which is used several times in the episode – is related to stepping on the gas pedal in a car (put my foot on the gas, barely grazed the gas). It is used with reference to a situation where the Press Secretary C.J. openly expresses her personal opinion about hate crimes at a briefing, for which she is reprimanded by her colleagues. The metaphor is not preserved in the Polish rendition, nor is it replaced with a different one. Nonetheless the sense of the exchange is preserved. 

6.3. Idioms

The “In excelsis Deo” episode does not abound in idioms, however we noted one example, where the translator managed to turn one into a kind of a pun (Table 10).

Table 10. Translation of idioms

In Polish there is a colloquial phrase ‘nie robić z czegoś szopki’ meaning ‘not to make a big deal out of something’. However, the word ‘szopka’ signifies a ‘manger’ in Polish, so given the scene in which it is used (the President sneaking out to do some last minute Christmas shopping by himself) this phrase gets a double meaning and is a great play on words.

6.4. Irony

Another possible critical translation point is irony. It may pose a particular challenge to an AV translator, as irony can be expressed not just by words, but also by non-verbal information such as body language, gestures, facial expression and tone of voice. In the episode concerned we detected one case of irony which – as it turns out – did in fact cause problems to the translator.

In the analysed scene Toby is on the phone trying to establish some facts about the death of the homeless man (see section 6.1.3. above), but he is constantly put on hold. While he is holding, Mandy – the White House media consultant who is in charge of  getting the White House ready for Christmas – comes into Toby’s office and we witness the following exchange presented in Table 11:

Table 11. Translation of irony

Toby is being ironic when he says: ‘It might seem trivial?’. This is expressed by both the grammatical structure he is using (the modal verb ‘might’) as well as his intonation (with emphasis placed on the word ‘might’). None of these two features are preserved in the voice-over: the translator has replaced the modal verb with a ‘to be’ verb, while the voice talent read out this line using regular, rather flat intonation. In effect, the Polish viewer is deprived of the ironic effect. This example points to the fact that, firstly, the AV translator must not only concentrate on the dialogue list and the image, but also on how the words are uttered, and secondly, that there must be better cooperation between the translator and the voice talent as regards the way certain lines should be read out. However, since studios responsible for preparing VO versions try to cut their production costs, voice-over translators are less likely to be present at the recording to make last-minute improvements or to intervene when the oral rendition by the voice talent is unsatisfactory (Majewski, personal communication).

6.5. Forms of address

In Polish there is a distinction between second-person pronouns that are used depending on the degree of politeness, courtesy, familiarity or social distance between participants of a social interaction (the so-called T-V distinction). In Polish, the T-forms would include addressing the addressee using the pronouns ‘ty’ (you singular), ‘wy’ (you plural) as well as their first names (nominal phrases), whereas V-forms include such forms of address as ‘Pan’ (Mr), ‘Pani’ (Mrs), Panowie (Gentlemen), Panie (Ladies) or ‘Państwo’ (Ladies and Gentlemen, Mr and Mrs), all of which are expressed in English by one pronoun ‘you’. 

When translating a dialogue list from English into Polish, the translator is often faced with a dilemma of what forms of address to use to render the ubiquitous ‘you’ into Polish. They then have to use their common sense and intuition to assess the degree of familiarity between the speakers on screen and determine whether they would be on a first or rather third name basis in Poland. However, it should be noted that such assessment is not always successful in Polish film translations, and cases where characters address themselves using a T-form during their first encounter or when there is a substantial social distance between them are not a rarity (on translating English forms of address into Polish see also Szarkowska 2006). Table 12 presents interactions which could be potential critical points for the VO translator.

Table 12. Translation of forms of address

In the first example we witness an exchange between a young man Charlie and his much older co-worker Mrs Landingham. Despite their working together, the translator decided that the age difference was big enough for Charlie to address her using a T-form, and thus – quite rightly – used a V-form instead. The next example, on the other hand, is quite conspicuous, as it is a mixture of both a T- and V-form, which is quite rare in contemporary Polish. The translator might as well have dropped the ‘George’ (cf. Text reduction in 5.2.), as it adds nothing to the exchange and takes up precious time. 

The third example is from a scene where Josh and Sam go and visit Sam’s friend Laurie who is an elite prostitute. They want to convince her to divulge some discrediting information about influential Republicans in order to protect the career of their friend and colleague Leo, the chief of staff. In the exchange Laurie addresses Josh using a V-form whereas he speaks to her using a T-form (despite it being their first encounter). This sounds slightly condescending, but perhaps the translator decided such a form would reflect the asymmetrical power relations, though it seems that a V-form would be more appropriate in this case.

6.6. Grammar

It may seem unlikely that for a translator with good linguistic expertise grammatical issues may pose a challenge. However, as the example in Table 13 shows, grammar can also be a critical point in the translation process.

Table 13. Translation of grammar

Although from a grammatical point of view (sequence of tenses) the Polish translation is correct, given the wider context of this particular exchange (the lady says that her children are dead), the translator should take this information into account and adjust the translation accordingly, thus replacing the present form with a past one.

7. Conclusions

The above analysis of the technical constraints and critical points in the voice-over translation of The West Wing episode hopefully shows that voice-over is a rewarding subject of research in the area of audiovisual translation. The task of the VO translator is no less demanding than the task of a subtitler or a dubber. It is only that the requirements, constraints and challenges might be slightly different. In general, one of the greatest demands faced by a voice-over translator is that of reduction. The analysed translation is 30 per cent shorter than the original, which means that a considerable amount of the original text gets lost in translation. Even if this reduction does not entail content loss (when it involves reduction of proper names, greetings, and so on), it leads to a rather unnatural exchange of lines in a target language dialogue, which does not help much in creating the authenticity illusion. On the other hand, since part of the dialogue is omitted, the viewer can listen to the actors’ voices and experience some real emotions as opposed to the Polish voice talent’s deliberately flat and neutral intonation – this, in turn, adds to the authenticity illusion (created partially through the original and not only through the translation).

Bearing in mind the reduction and text condensation requirement, the translator’s task looks daunting in the context of all the cultural allusions present in the original version. Some of the original content and flavour is lost as the strategies most frequently applied to culture-specific items in this episode are normalization and importation. The former entails using culture-free words, which inevitably leads to the loss of the cultural context. The latter usually entails importing a proper name into the target culture. What gets imported is just the word, not the cultural connotation, which also leads to the loss of meaning and cultural context. This seems inevitable because any explicitation and addition of compensatory information is impossible due to time constraints. On the other hand, cultural substitution or full domestication would impair the authenticity illusion since the whole series is so deliberately set in the American political culture.

In general, the translator of “In excelsis Deo” did her best and provided solutions that – first of all – met the technical requirements and time constraints of the voice-over AVT mode and – secondly – led to as little loss as possible. It seems that the technical constraints have to be satisfied first and only then can the translator strive to find the best solutions to the critical points created by the original text. When dealing with the CSIs, the translator resorted to deletion only when importation or neutralisation were not possible. One of the greatest hurdles, apart from culture, turned out to be irony since it involves much more than just a text. Close cooperation with a voice talent would be required to find the best solution to such a critical point.

The Polish market of audiovisual translation has not changed much over recent years and it seems that voice-over is there to stay for good. Luckily, it offers great opportunities for translators to refine their skills and provide interesting translations. Moreover, as we hope to have shown, it also offers many new and exciting research avenues for the future. 

About the author(s)

Iwona Mazur works as Assistant Professor in the Department of Translation Studies, Faculty of English, Adam Mickiewicz University in Poznan, Poland. Her research focuses on audiovisual translation in general and audio description in particular. She has participated in a number of Polish and international research projects, including an AD reception study AD-Verba, Digital Television for All Project and the ADLAB Project. She serves as Executive Board member at the European Society for Translation Studies EST and the European Association for Studies in Screen Translation ESIST. More information: http://wa.amu.edu.pl/wa/Mazur_Iwona

Agnieszka Chmiel works as Assistant Professor in the Department of Translation Studies at the Faculty of English, Adam Mickiewicz University in Poznan, Poland. In her 2004 PhD thesis she focused on the neurocognitive aspects of conference interpreting. Her research interests include conference interpreting, audio description, audiovisual translation, cognitive studies, memory and visual imagery in interpreting. She works as an interpreter and has trained conference interpreters at AMU for 12 years now. She has participated in many international translation studies projects, including ADLAB and IVY. She is Head of the Postgraduate Programme in Audiovisual Translation at AMU. Profile: http://wa.amu.edu.pl/wa/Chmiel_Agnieszka

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

•  Previous article