Lost in Translation

(1)

Department of English

Master Degree Project English Linguistics

Spring 2020

Lost in Translation

Voice-over Translated TV as a Source for Incidental Language Acquisition

Anna Justyna Sochacka

(2)

Lost in Translation

Voice-over Translated TV as a Source for Incidental Language Acquisition

Anna Justyna Sochacka

Abstract

Voice-over translation (or “audio subtitles”) is the only audio-visual translation (AVT) used for feature-length films on Polish television. Since the reader (lektor) only partially overrides original dialogues and some parts of the actors’ speech are clearly audible, it can be considered as marginal L2 exposure. The aim of this study is to investigate the attitudes of Polish speakers of English towards voice-over translated television and to determine if TV shows and films with a lektor can be a source of incidental language acquisition. The study consists of two parts, a survey and an experiment. The survey investigated the attitudes of Polish speakers of English to different types of AVT, especially their opinions about the influence of voice-over translated TV shows on their English proficiency. The data reveals that Polish speakers of English have rather negative opinions about this AVT and numerous instances of unfaithful translation are easily spotted by them, which creates an image of unprofessional and inaccurate AVT. They seem not to consider films with a lektor as L2 exposure and tend to think that introducing subtitles to television would improve the L2 competence in Poland. However, the informants admitted being able to spot unfaithful translation, meaning that they pay attention to the foreign language in the background. The experiment was conducted to examine the possibility of incidental language acquisition from voice-over translated TV shows. 26 high school students, Polish learners of English watched 4 clips with voice- over translation with instances of unfaithful translation. The study has empirically proven that Polish learners of English are able to spot the differences between the original dialogue and its translation. Thus, watching voice-over television programming can be considered L2 exposure.

Keywords

Second language acquisition, audio-visual translation, voice-over translation, lektor, subtitles, incidental language acquisition, L2, Polish learners of English.

(3)

1. Introduction

There is no doubt that exposure to different types of foreign language (L2) media can facilitate the overall language learning process. As Webb (2015) stated, due to a large amount of time spent in front of a TV, “television might be the greatest source of first language input” (p. 159). Therefore, we can assume that by increasing their L2 exposure by watching TV shows, L2 learners may significantly improve their language competence, such as vocabulary knowledge or listening skills (Webb, 2015, p. 162).Even though it has been proven that incidental language acquisition from listening might be less significant than from reading, there is no denying that it has a substantial influence on the learning process (Rodgers, 2013, p. 74; van Zeeland and Schmitt, 2013, p. 620).

What is more, watching TV-shows in L2 may quite often be the main source of informal language (Bahrani et al., 2012, p. 347) and fill in the gap between formal language taught in the classroom and language in use.

While watching films and TV shows with subtitles is exposure to L2 and a dubbed motion picture does not provide any L2 input, it is quite difficult to categorize films with voice- over translation. The voice-over translation technique is the only type of audio-visual translation (AVT) used by Polish TV broadcasters for feature films and TV series. It involves original dialogues partially overridden by a single male lektor who reads translated lines. The translation is designed to be significantly shorter than original dialogues and the lektor reads the translated script so that at least the first and last seconds remain audible in order to preserve the artistic value of the actors’ play. Thus, we can view voice-over translation as “partial dubbing” or “audio subtitles” (Franco, Matamala and Orero, 2013, p. 27, Braun and Orero, 2010, p. 178, Wozniak, 2014, p. 214).

Voice-over translation is considered a relic of communism (Wozniak, 2012, p. 210) and to date, the linguistic research on this type of AVT focuses mainly on the translation process. The possibility of television programming with a lektor being a stimulus in language learning is virtually non-existent. Thus, this study will shed some light on voice- over translation in the field of L2 learning by examining whether Polish learners of English who are used to TV shows and films with a single lektor are able to focus on the original English dialogues and whether they can notice the differences between original dialogues and the translation. In order to answer the question if watching films with a lektor could be considered as exposure to L2 input and thus lead to incidental language acquisition, a survey and an experiment were carried out. Moreover, this paper investigated the attitudes of Polish speakers of English to different types of AVT.

1.1 Research questions

In this study, I investigate voice-over translation in the field of L2 acquisition and answer the following research questions:

1. To what extent are Polish learners of English able to focus on the original language of films with voice-over translation? To what extent can we consider films with voice-over translation as exposure to L2?

2. What is the attitude of Polish speakers of English as L2 to different types of AVT?

Is there any connection between their proficiency and their choice of AVT?

(7)

3. Since there is an age gap between the informants of the experiment and survey, what is the difference between their preferences of choice of AVT?

1.2 Thesis structure

In the Literature review below, the previous research on language exposure and incidental language acquisition and AVT will be described. It is followed by the Methods and Materials section where the study will be presented. The study was divided into two parts, a survey conducted among Polish adult speakers of English and an experiment involving high school students (learners of English). The findings will be presented and discussed at the end of this paper followed by conclusions.

2. Literature review

This section will highlight some aspects of incidental language acquisition and L2 exposure. In this context, different types of audio-visual translation will be listed and discussed. Since this paper will focus on how voice-over translation can contribute to incidental language acquisition, the newest research on this particular topic will be reviewed.

2.1 Incidental language acquisition

Incidental language acquisition is an indispensable part of the language learning process.

Often L2 learners intuitively choose to read books, watch TV-series and films in the foreign language to facilitate and accelerate the learning process. With less cognitive effort, they can acquire new words and phrases. L2 learners are aware of the beneficial influence of L2 exposure on their foreign language competence like improving their pronunciation, intonation and knowledge of colloquial expressions. Furthermore, encouraging learners to read novels, watch television or even to play video games in L2 should be incorporated in the L2 curriculum in order to teach those different learning strategies (Kuppens, 2010, p. 80). However, it may not be easy to draw a clear line between incidental and intentional learning. Even though the goal is the same, namely to develop L2 skills, incidental L2 learning can occur whenever a learner is exposed to L2 input and gains vocabulary or grammar knowledge without explicit instructions or deliberate intention to learn (Leow and Zamora, 2017, p. 33). Gass (1999) defines that incidental L2 acquisition is “a by-product of other cognitive exercises involving comprehension” (p. 319) while intentional learning is defined as “a deliberate attempt to commit factual information to memory, often including the use of rehearsal techniques, like preparing for a test in school or learning a song by heart” (Hulstijn, 2013, p. 1).

In order to benefit from L2 exposure, a learner needs to reach a minimum level of language proficiency, since

“acquisition involve(s) a certain degree of syntactic and lexical knowledge that learners heavily rely on for comprehension, and it is this relationship that needs to be taken into account for vocabulary learning purposes” (Restrepo Ramos, 2015, p. 160).

(8)

It has been stated that after learning the first 1,000 words, incidental L2 vocabulary learning can occur (Restrepo Ramos 2015, p. 159). However, De Wilde and Eyckmans’

research (2017) investigated whether L2 gain may occur without any instructions before the learners start their English course. Their results concluded that Flemish primary school students who just started their formal L2 learning were able to complete tasks on the A2 level. While there was no clear difference between pupils who spent their time watching subtitled TV-shows or listening to music in English, students who played online games and used the Internet received even higher scores. Thus, it can be stated that there is not a clear definition of the minimum proficiency required for incidental L2 acquisition and it may vary for different age groups.

As Gass (1999) points out that it is impossible to state which words or structures were learned incidentally which were learned consciously (p. 320). However, Hulstijn (2013, p. 4) and Kuppens (2010, p. 80) clearly state that incidental acquisition cannot replace intentional learning due to several factors, such as low effectiveness of incidental learning, or the fact that learners may simply ignore unfamiliar words and structures. On the other hand, there is a significant difference between the formal language taught at schools and the actual spoken language. This loophole can result in learners’ inability to communicate in more casual situations, for instance, Lin and Siyanova-Chanturia (2014) brought up an example of a student graduating with a degree in English who still felt lost in communicating with native English speakers while living in England. The student managed to fill in the gap by watching American TV series online. Webb (2015) directly suggests creating a regular practice of watching television in L2 in order to help the overall learning process. Simultaneously, Zuo and Yan (2019) underline that there is a necessity to combine both intentional and incidental learning, in order to gain better L2 knowledge and its retention (p. 40, 43).

Incidental L2 gains are vastly studied in the field of second language acquisition.

However, the majority of the studies focus on incidental vocabulary acquisition. The reasoning behind this is that L1 words are naturally gained without any explicit instructions, so the assumption that it also occurs in L2 learning arose (Leow and Zamora, 2017, p. 36). For instance, Kweon and Kim’s study (2008) of Korean learners of English has proven that extensive L2 reading has a significant impact on incidental vocabulary acquisition and the retention of learned words is high. There are of course many factors that will influence incidental word gain such as frequency, similarity to L1, or even a word class (Kweon and Kim, 2008, p. 206, 208).

The effectiveness of incidental vocabulary acquisition may vary depending on the type of exposure. A study by Vidal (2011) lead to the general conclusion that incidental L2 vocabulary gains as well as retention of newly learned words from listening are less significant than from reading and this is applicable for all the levels of L2 proficiency.

Furthermore, this effectiveness differs on different stages of learner’s L2 competence.

Vidal (2011) found that there is a correlation between language proficiency and the ability to acquire vocabulary from listening, namely, written input has a more significant

(9)

influence on the development of vocabulary among less advanced learners, while advanced learners are able to acquire more words from listening (p. 243).

In many cases, the researchers investigated the influence of watching L2 TV shows with or without AVT on L2 acquisition. The experiment of Yuksel and Tanriverdi (2009) was conducted on two groups: group A watched videos with captions (subtitles in the original language or intralanguage subtitles, see Section 2.2.1), and Group B without. The results show that there is no significant difference between watching a film with or without subtitles, however, this small difference worked in the favour of Group A. These finding were upheld by Harji, Woods and Alavi’s (2010) study which involved a similar experiment among TEFL Iranian students, however, it was found out that students watching films with captions gained a significantly better result than the students watching videos without any AVT (p. 41). What is more, Westergren (2010) proved that even though the gain of new vocabulary after watching a film without subtitles once is relatively small, there is no doubt that such gain occurs in a short term. Therefore, watching television content in L2 with or without subtitles can contribute to expand L2 vocabulary knowledge.

Furthermore, different types of subtitles may have a different effect on the number of L2 gains. For instance, Winke, Gass and Sydorenko (2010) found that intralanguage subtitles can facilitate beginners whose comprehension of L2 is limited (p. 81). Also, Lommel, Laenen and d’Ydewalle (2006) stated that the most convenient situation is when a learner with sufficient knowledge is exposed to both the original soundtrack and captions, then the acquisition of new lexical items is most effective (p. 244). The effectiveness of inter- and intralanguage subtitles in incidental vocabulary acquisition was recently investigated by Frumuselu, De Maeyer, Donche and Colon Plana (2015). Their study on college students with a different linguistic background majoring in English (BA level) showed that the intralanguage captions have a greater effect on their L2 skills than interlanguage subtitles.

The above experiments focused mostly on the short term impact of L2 exposure. Study by Kuppens (2010) has shown the long term influence of exposure to English language films, programs and games on Flemish pupils’ L2 competences. She found a clear correlation between the exposure to these English language media and improvement on the students’ translation skills (Kuppens 2010, p. 78). That leads to a conclusion that the more expensive L2 exposure is, the better effect can be observed.

Grammatical rules can also be acquired by L2 exposure. However, an experiment by Lommel et al. (2006) involving a single film viewing proved that it is not sufficient to have a significant impact on the L2 grammar gain. Lommel et al. (2006) have also shown that grammar may be too complicated to understand and better effect can be achieved by presenting the rule in advance, i. e. combining it with explicit learning. A more recent study by Aka (2020) performed an experiment on Japanese learners of English. After reading 40 sentences, the students gained knowledge on the specific grammatical feature.

In other words, L2 reading may have a greater impact on the grammatical knowledge than L2 listening.

(10)

To sum up, previous research has shown that exposure to original L2 input has a positive influence on the learner's knowledge. If vocabulary is gained involuntarily, as a by- product of some other activity than intentional language learning, it is called incidental vocabulary acquisition. Even though incidental vocabulary acquisition occurs more frequently from reading than from listening, watching a film with an original soundtrack can promote better pronunciation, understanding and even affect the learner's grammar and vocabulary knowledge. Thus, the choice of the AVT type for television may have far-reaching consequences not only for an individual learner but also for the whole country. Taking into consideration the time spent watching TV, by broadcasting films and TV shows with subtitles the country may extend L2 exposure and facilitate language learning.

2.2. Audio-visual translation

This part of the essay will discuss what the difference among subtitling, voice-over translation and dubbing is and highlight some aspects of voice-over translation regarding its possible influence on incidental second language acquisition. This subsection will also explain the translation conventions for films with a lektor.

2.2.1. Dubbing and subtitling

Dubbing and subtitles are in general commonly used in the Western countries. Dubbing, popular in France and Germany, involves creating a new audio track with translated dialogues, leaving the music soundtrack and other sound effects unchanged. The dialogues imitate the actor’s performance and the synchronization between the translated and original speech must be preserved. Subtitles, which are widely used in, for example, the Scandinavian countries, the Netherlands, and Belgium can be divided into two subgroups: interlingual and intralingual (also called captions, unilingual or bimodal subtitles) (Pelletier, 2012, p. 2). Intralingual subtitles are used in case the spoken language cannot be understood by the audience due to the low quality of the material, distorted voice, a strong accent of the speaker, etc. They facilitate the audience who may find it challenging to understand the speech but also help people with hearing impairment.

Interlingual subtitles provide translation of a foreign language with the unchanged soundtrack of the source material. In most cases, subtitles are placed on the lower part of the screen and include only one language. However, both Flemish and French subtitles are used simultaneously in Belgium (Pelletier 2012, p. 2).

As Pelletier (2012) states it is usually the cost of production that had a decision-making role which AVT would be used as the main form of translation and so countries with a fairly small number of speakers, such as Sweden or Greece tend to use subtitles while Germany and Spain could afford dubbing (p. 3). The choice of AVT is not only a question of preferences or habits in a particular country; it is also a political matter. The surveys on the European language preferences created by the European Union always include a question about AVT. The importance of watching films with subtitles in L2 learning was noticed by the European Commission, and the choice of a certain type of AVT was considered as one of the indicators demonstrating attitudes toward multilingualism. In

(11)

their report based on a survey, Europeans and Their Languages (ETL, 2012), it is said that:

“In spite of their generally positive views towards multilingualism, a small majority of Europeans prefer to watch foreign films and programs dubbed, rather than with subtitles. More than two in five respondents (44%) agree that they prefer subtitles to dubbing (…) However, the balance of opinion appears to have shifted slightly in favour of subtitles since 2005, when 37% of respondents agreed that they preferred subtitles to dubbing“ (ETL, 2012, p. 117-118).

Furthermore, in the countries with a high number of citizens who speak at least one foreign language, the audience prefers to choose subtitles; for instance, in Sweden, 91%

of respondents speak at least one foreign language and 93% prefers subtitles to dubbing while in Hungary only 35% claim to master another language than their mother tongue and 27% would choose subtitles (ETL, 2012, p. 16, 118). However, it must be emphasized that the choice of AVT does not have to influence foreign language proficiency. In Germany, the majority of motion pictures broadcast on television are dubbed and still, the level of proficiency in English is quite high. Foreign language input from television, cinema, etc. even though important is not a conclusive factor. In other words, a good knowledge of a foreign language in a certain population is an effect of many elements, for instance, the differences between learners’ L1 and L2, education system, or even the attitude being a result of the political and economic situation. On the individual level, it is hard to state whether those who master a foreign language more willing to improve their skills in their free time by watching a film with subtitles or whether watching a foreign film with subtitles is not that tiresome for them.

Researchers usually compare dubbing with subtitling considering them the main AVT techniques. Thus, many studies were conducted in order to present or contrast those two types. For instance, Wissmath, Weibel and Groner (2009) and Perego, Del Missier and Bottiroli (2015) researched on how subtitles and dubbing influence the comprehension and enjoyment of watching a film. Their results prove that there is no particular difference in comprehending the content of the feature films, what is more, AVT does not affect the audience's enjoyment. Both studies completely ignore the third AVT type, voice-over translation.

2.2.2. Voice-over as an ”ugly duckling” of AVT

Films with a lektor were present in movie theatres at the beginning of the cinematography in the nineteenth century when a single reader delivered dialogues or commentaries in the silent films (Franco et al., 2010, p. 17). Within the time the readers’ role evolved and nowadays, voice-over translation is an AVT technique which involves the original film’s dialogue overridden by a lektor or lektors who read translated script or dialogues. Voice- over translation is known in most countries, but it is rather an unpopular technique of audio-visual translation in the Western countries, where a lektor can be heard almost exclusively in the news or interviews; foreign language feature films and TV series are subtitled or dubbed (Franco et al., 2013, p. 25). In Eastern Europe, voice-over translation is used for feature-length films and other television genres (Franco et al., 2013, p. 24).

This so-called “partial dubbing” or “read subtitles” can be perceived as unprofessional in

(12)

many aspects. However, for the audience which is used to it, voice-over translation provides them a unique possibility to listen to the original soundtrack and to understand the foreign language, and this gives them a stronger feeling of faithfulness of the translation (Franco, 2013, p. 27-28).

In Poland, Estonia and Latvia all genres are voiced-over and there is always one male reader in feature films, who provides the translation without any emotional input.

However, in Ukraine and Belarus, there are at least two lektors, male and female and often more readers can be involved to imitate the emotional content of the dialogues (Wozniak, 2010, p. 215). Voice-over translation for feature films is also popular in China and Japan (Braun and Orero 2010). However, this thesis will discuss only a single male lektor and the other types of AVT are not a topic of this essay.

Voice-over translation with a single male reader raises mixed feelings among AVT researchers (Wozniak, 2010, p. 210). It has been called an ”ugly duckling of AVT”

(Orero, 2006) and “a rare bird, with each country establishing individual technical and aesthetic standards of its execution” by Holbut (2014, p. 138). Wozniak (2010) states that the differences in opinions are mainly caused by the fact that voice-over translation is not a very well-known technique (p. 210). She continues that

“(t)he peculiarity of (...) the voice-over translation strategy (...) seems particularly puzzling to many foreign scholars, who even resort to negative or sarcastic comments about it” (Wozniak 2012: 210)

For instance, mentioned above Pelletier (2012) while describing the usage of dubbing and subtitles in the EU defines ”voice-over” as a technique from ”emerging new states from the former USSR. They are relatively new to the need of translation (…) and so distributors are still reluctant to invest much in a market” (p. 6). Pelletier (2012) is also incredibly sure that voice-over translation is doomed to disappear from AVT in order to give place to more advanced types and suggests ”to hire people to do actual dubbing and not just voice-over” (p. 6). This is not surprising since in the Western countries voice- over translation is used only in non-fiction, news or documentaries. Thus, there is not much literature dedicated to this AVT type compared to subtitling or dubbing (see Franco et al., 2013, pp. 21-22).

2.2.3. Translation in voice-over television films and programs

Literal or very faithful translation is not always possible in any AVT type. Orero (2004) stated that there is not a particular difference between fidelity of translation in voice-over translated films and those with subtitles or dubbing. Although there is no need to adapt the translation to lip movements like in dubbing and some subtitling strategies can be applied, the loss in voice-over translation is caused mainly because there is only one reader (Sepielak and Matamala, 2014, pp. 147-148). Wozniak (2010) also summarizes that “this type of translation is in fact far from being an exact reproduction of the original speech and that it tends to simplify and reduce the amount of information” (p. 214).

The most distinctive feature of voice-over translation is having two audio tracks, one with the original soundtrack and the other with the read translation, played simultaneously.

(13)

Franco et al. (2010) have distinguished and named four types of synchrony in this type of AVT translation:

 voice-over isochrony – utterances of a reader are much shorter than original dialogues, so that the beginning and the end of the original utterances are audible, not overridden by the translation (Sepielak and Matamala, 2014, p. 149). This type of voice-over translation reduces the content of original dialogues to a great extent by omitting certain elements, for instance, address forms, exclamations, repetitions or question tags (Sepielak and Matamala, 2014, p. 150);

 literal synchrony is the opposite to voice-over isochrony; this type refers to a very accurate translation which should be read exactly in the same time when the original utterance is said, in this way the lektor's voice entirely overrides the original dialogues (Sepielak and Matamala 2014: 149); this technique is designed to prevent the audience from hearing original dialogues and to perceive the translation unprofessional or non-credible (Sepielak and Matamala, 2014, p. 155);

 kinetic synchrony's (orfacial expression synchrony) main feature is that it allows a reader to express verbally some gestures and body language of the actors (Sepielak and Matamala, 2014, p. 156), “for example, if an actor nods affirmatively, a contradictory translation saying “no” is not heard” (Sepielak and Matamala, 2014, p. 149);

 action synchrony is similar to kinetic synchrony but apart from giving a description of gestures, a lektor refers to the action (Sepielak and Matamala, 2014, p. 157). Sepielak and Matamala (2014) exemplify action synchrony with exchanging a noun with a pronoun, when the object is visible on the screen and pointed by the actor (p. 158).

All four types are used on the Polish television, often used interchangeably, appropriately for the situation.

Lektors usually have a low timbre; their speech is monotonous and neutral. Their static intonation cannot express any emotional content leaving their voices as an imperceptible part of the background. As advantages of voice-over translation, Wozniak (2010) lists such aspects as no need to focus on subtitles and also the fact that there is no loss in the original intonation and actors' play (p. 2015). Almost always in feature films, the lektor starts reading the translation with a few seconds of delay and finishes before actors, which is an example of voice-over isochrony. Even though no rules specify how long the gap should be (Franco et al., 2013, p. 80), Wozniak's (2010) analysis of those delays has shown that that in certain films up to 90% of the original dialogues (actors’ speech) is clearly audible for the audience, however, in others it is only 42% (p. 221). Often, the translation is an excerpt from what is actually said in the film, so naturally, the lektor's parts are shorter. If longer pauses occur in the original soundtrack, the lektor reads his text in those breaks between actors' speech. Thus, literal synchrony can hardly be found in feature films. Wozniak even calls voice-over translation “a voice in-between” or

“voice-behind” (Wozniak, 2010, p. 216). “The first and last words will not only be heard by the audience but very often be understood by some of them” (Luyken et al., 1991, p.

141) and Franco et al. (2013) adds “if not understood by the audience in the foreign

(14)

language, (infidelity in translation) can nevertheless be spotted” (p. 81). Further Franco et al. (2013) points out that due to the fact that the audience is not familiar with the challenges of translation they also can spot also non-literal translations (p. 80).

Still in the context of this research, films with a lektor could be considered as a foreign language input that could help in the second language acquisition. Of course the viewers' perception of a voice-over translated film is rather a subjective experience for each individual but it can be assumed that the process of listening to a lektor is similar to reading subtitles, so done automatically and unconsciously (Wozniak, 2010, p. 221). In the end of her study Wozniak (2010, p. 225) also states that “such a strategy can help viewers understand the original dialogue only if they already have some knowledge of the language spoken”, so the viewer needs to achieve a minimum L2 proficiency to benefit from watching films with a lektor.

2.2.4. Audio-visual translation in Poland

In Poland subtitling, dubbing and voice-over translation are used for non-fiction AVT.

However, all of them have certain domains in which they occur. Voice-over translation is used exclusively on television, but it is used for both fiction and non-fiction films and TV programs. This technique previously was used for some cartoons and all feature- length films for children and adults. Nowadays, more and more cartoons and films for children and young adults are dubbed.

Subtitling is almost never used on Polish television. The only exceptions are Polish films broadcast for the international audience on Channel TV Polonia, videos with bad audio quality or any material which can cause problems in understanding the speech (distorted voices, regional dialects, etc.). On the other hand, subtitling has dominated films in movie theatres, where only films for children are displayed dubbed, and motion pictures for adults have subtitles. Voice-over translation is never used in the cinema. However, documentaries with a lektor are displayed in some special movie theatres such as IMAX.

It can be said that the choice of the AVT for a motion picture on the television or the cinema is rather a matter of habit, not rules or aesthetics.

What should be said is that voice-over translation is not related to the level of illiteracy or the wealth of the society as suggested by Glaser (1992) or Perego et al. (2005), but it is rather a cultural phenomenon. In Poland voice-over translation is commonly called

“films with a lektor” (“film z lektorem”), that is why in this essay this expression will also be used.

2.2.5. Voice-over translation in Poland

Surprisingly, there are no strict rules that would state how text should be read by a lektor.

The readers are left with the whole range of possibilities, which they have developed themselves (Wozniak, 2010, p. 216). Thus, an unwritten rule in Polish AVT as Wozniak (2010) claims is that only male readers are employed to read feature films' translations (p. 213). The majority of nonfictional motion pictures are read by male readers as well.

(15)

Although, some of them have a female reader, for example, a popular lektor, Krystyna Czubowna who narrates nature documentaries. The number of lektors is rather limited, the lektors’ voices are easily recognized by the general public. The audience can easily indicate which lektor works in a particular channel and they also have their preferences.

It is the lektors themselves who created the high standards of their job, often calling it “an art”. Krzyzaniak (2008) states that “(t)he best lektor is the one we do not hear after a couple of minutes so we do not lose anything from the original actors' timbre and intonation of voice” (p. 6 cited in Wozniak, 2010, p. 215). As said above, the lektor's voice can be easily ignored by the audience so that they experience an illusion of understanding the actors' speech. This neutrality in the reader's voice must be sustained throughout the whole film to avoid a conflict between his intonation and the emotional content of the film, in other words, a lektor cannot be another actor in the film, he needs to be unnoticeable or “invisible” (Wozniak, 2010, p. 214). That implies that the viewers must pay attention to the actors’ voices in order to understand the key aspect of cinematic art, namely acting, and this leads to the main thesis of this paper that the input from the voice-over translation should have a positive influence on the acquisition of English.

Furthermore, not much linguistic research has been done on voice-over translation in Poland.

The studies usually focused on issues connected to the translation process itself and was named by Bogucki (2010) ”error-hunting” showing imperfections of the translated version (p. 416). Even though films with a lektor are commonly blamed for the low level of English in Poland, this concept belongs to popular science rather than to any research- based knowledge, since there was no study on how it actually influences foreign language knowledge. However, taking into consideration what was said about incidental L2 acquisition from listening and the peculiarity of the illusion of ”understanding” the original soundtrack, it must be underlined that the aim of this study will rather be to prove that a voiced-over film can be considered as a marginal input of L2.

2.2.6. Summary

Subtitles and dubbing are considered as the main types of audio-visual translation. Since L2 input is considered as important for foreign language learners the EU institutions have carried out research on the preferences in the choice of AVT as an indicator of attitudes toward learning languages in Europe. Even though voice-over translation is widely used in Eastern Europe, it is considered inferior to dubbing and subtitles. In Poland, voice-over translation with a lektor are used in all feature films, documentaries and some cartoons broadcast on the television. Thus, lektors are blamed for the low level of English in Poland. On the other hand, due to the fact that literal translation is impossible in AVT, the synchrony is never complete and the audience can hear the original voices of actors.

Films with a lektor allow the audience to experience the actors’ play and create a feeling of understanding a foreign language.

Taking into consideration Wozniak's (2010) findings and previous research on incidental vocabulary acquisition, it can be discussed whether films with a lektor could be called exposure to L2 input and thus lead to incidental vocabulary acquisition.

(16)

3. Methods

This part of the essay contains a review of the methods and materials used in the investigation. The study is divided into two parts, a survey and an experiment. In the first subsection, the choice of the informants of the survey and the experiment will be presented as well as the background information about them. Since there is an age gap between both groups, the informants of the survey will be referred to as Polish speakers of English, and the participants of the experiment as Polish learners of English. This part will be followed by the description of the questionnaire and details on how the survey was carried out. Then the four stages of the experiment will be presented. The study is, for the most part, a quantitative study, but a qualitative approach was used to complement the findings by analysing the open question answers and the classroom discussion.

Furthermore, the chi-square test was performed on some data sets to verify whether there are any relations between the groups and their AVT preferences. The program used for statistical testing was SPSS Statistics¹.

3.1 Participants

The participants of the study are presented in the two sections below: first, the survey informants (Polish speakers of English) and second, the experiment participants (Polish learners of English).

3.1.1 Survey informants

The online questionnaire was sent via email to 120 participants, young professionals living in Poland. 94 of them completed the survey. The majority of the respondents were women (66 respondents), only 30% of completed forms were filled in by men. All of them speak Polish as their native language, English as a foreign language and 7 informants declared knowledge of other languages. The age of the informants ranged between 24 and 32 (M age = 27). Every informant confirmed learning English at school, and a great majority (90%) passed the Matura exam² with English as an optional foreign language on the basic or advanced level³. 28% took additional English courses outside the curriculum. 18 (19%) participants evaluated their level of English as basic, 37 (39%) as communicative, and 38 (40%) are fluent in English. One person admitted that their level of English is poor, however, they attended English classes in primary and high school. 15% of participants graduated from high school and did not continue their education on the academic level, 72% had a university degree, and 12% were still

1 IBM SPSSStatistic Version 25.

2 Final examination after completing all courses at secondary school. It is an equivalent of British A-levels.

3 In Poland, students can choose if they want to pass the final exams on the basic and advanced level, the basic level in a foreign language examination corresponds to B1 and advanced to B2 level (Common European Framework of Reference).

(17)

studying at a university level. The informants have shown a considerably high interest in learning foreign languages. More than two-third of the respondents estimated that learning English was a very important or important part of their educational process.

The choice of this age group was motivated by the fact that this is the first generation in Poland who received education after the collapse of communism in the late 1980s. It means that they had an opportunity to study English at every stage of their education, usually starting in the fourth grade (age 10-11).

Moreover, this group was exposed to voice-over translation as the only available type of AVT in the public and private Polish TV channels during their education in the 1990s and up to the mid-2000s. As said above, subtitled films are available in the cinema only. Some of them could have access to the foreign cable or satellite television, where other types of AVT were available. On the other hand, with the development of the Internet, different data carriers and later the streaming platforms, this group can actively choose the AVT type. However, that required the recipient to actively look for films with other AVT type than voice-over translation.

The informants were informed about the purpose of the study in a short introduction to the survey; they were asked to give their consent to their answers being used anonymously in this study. In order to ensure the confidentiality, code names were used (I1, I2, I3, etc.).

3.1.2 Experiment Informants

The participants of the experiment were senior students (grade 11) of a secondary school in Włocławek, Poland. The age of the students ranged between 16 and 17 years old. The students were preparing for the Matura examination the following year, and all of them have chosen to take an exam in English as the foreign language on the B2 level. Out of 32 students, 26 were present at school when the experiment was carried out. The informants were divided into two groups, Group 1 (14 participants) and Group 2 (12 participants).

On the one hand, the experiment participants were exposed in a greater extent to dubbing than the informants who took part in the survey, mainly due to new channels aimed at children and young adults available over cable television, which were launched in the second half of the 2000s, for example, Disney Channel Polska. On the other hand, the education system in Poland changed the approach to teaching foreign languages and put a lot of emphasis on teaching English. The students who participated in the experiment participated in obligatory English classes from the first grade (age 6-7), they also had a mandatory English test included in the final examination after gimnazjum⁴.

Since the participants of the experiment were underage, both the students and their parents or legal guardians were asked to give written consent to take part in the experiment and for the results to be used anonymously in this thesis. In addition, they were assured that

4 Gimnazjum was a type of a junior high school, grade 6-8 (age 13-16); school finished with an exam in humanities, science and English, similar to B-levels. Gimnazja were gradually closed after the education reform in 2017.

(18)

the materials used in the study (video clips) would not contain any inappropriate or controversial content.

3.2. The survey

The material involved 2 open-ended and 18 close-ended questions with prompted (fixed) responses. Implementing so many closed-items brought uniformity in the informants’

answers and helped to quantify the data (see Grass and Mackey, 2011, p. 152, 160).

However, due to the lack of opportunity to interview the participants after completing the questionnaire, extra comment boxes for additional remarks were added to get a better understanding of their views and opinions. Questions 1-8 were designed to retrieve factual information which could be relevant for the results, for instance, gender, age, mother tongues, education, and English proficiency. In Questions 9-10, the informants were asked whether they had been able to notice unfaithful translation of feature films and TV shows and how often it occurred. Questions 11-15 aimed to collect behavioral data, namely check the preferences of AVT type depending on the language or genre. In Questions 16-19, the informants were asked about their personal opinions on how films with a lektor have influenced their L2 acquisition and if implementing changes in AVT type would improve the knowledge of English in Poland.

The informants (see Section 4.1) could fill in the form online or were provided a paper form if required (Appendix A). In the introduction to the survey, the participants were informed about the purpose of the study and what kind of data would be collected. In order to submit the questionnaire, the informant had to have a Polish IP address and had to answer all questions, there was no time limitations. To prevent any misunderstandings and limiting the answers, the questionnaire was provided in Polish (see Grass and Mackey, 2011, p. 162). Quotes used in the Result section were translated into English.

Apart from Question 14, no negative questions were asked, to avoid biased answers (see Grass and Mackey, 2011, p. 163).

On the one hand, “questionnaires will often be one of the most efficient data collection methods that the second language researchers have at their disposal” (Grass and Mackey, 2011, p. 163) and this method allowed to remotely collect data from a significant number of Polish speakers of English. As any other tool to collect the quantitative data, the survey was designed mainly “to provide a broad-brush overview of general trends and relationships” (Groom and Littlemore, 2011, p. 93), namely to address the question about the informant’s preferences and attitudes to different AVT types (see Question 2 in Section 1.1). On the other hand, the survey did not allow to collect any naturalistic data.

Groom and Littlemore (2011) point out that a questionnaire “will never provide direct information about people’s behavior. It will only provide information about what they think about their behaviour” (p. 97-98). Answers to Question 9 and 10 were based on the participant’s retrospective self-observation on their own mental processes rather than on facts. Since the memories may decay or be exaggerated, the answers may have been biased and not fully address the first research question. Therefore in order to collect more accurate data, the complementary experiment was conducted to triangulate the data (Groom and Littlemore, 2011, p. 78-79; see Section 3.3).

(19)

Moreover, the survey was based on self-evaluation of the participants’ English proficiency and did not evaluate their actual L2 knowledge. Since the number of the informants was large and the survey was carried out remotely, engaging other data elicitation methods in this group was not possible. What needs to be noted is that the results obtained for Question 16 almost completely coincided with answers for Questions 11 and 12 and will therefore not be used in the study.

3.3 The experiment

This part of the study was designed to investigate the students’ cognitive process while watching an English language feature film with Polish voice-over translation and a reversed situation, a Polish language material with an English language reader. This method shared some common features with a written recall task. Grass and Mackey (2011) define

“stimulated recalls, a subset of introspective methods, are used to explore learners’ thought processes or strategies employed during a task by asking learners to reflect on their thoughts after they have carried out a task. (…) While hearing or seeing these stimuli, learners are asked to recall their thought processes during the original event.” (Grass and Mackey, 2016, p. 22).

After each clip was stopped, the participants (see: Section 3.1.2) were asked to write down the English utterance on an answer sheet immediately after hearing the target phrase (see Table 1). Thus, it can be defined as an immediate recall, which “can be distinguished from simulated recall in that a simulated recall may or may not occur immediately following the event” (Grass and Mackey, 2016, p. 60). However, due to the time limitation of the experiment, there was no chance to ask the students to elaborate any further about their thought process. After collecting the answer sheets, a free discussion was carried out about the purpose of the study and AVT in Poland. In the section below the choice of the materials will be discussed in detail and it will be followed by the description of the experiment’s implementation.

3.3.1. Selection of the Films

The materials for the experiment consisted of four video clips from three rather unknown American movies with Polish voice-over translation: for Clip A Presumed Dead in Paradise (Polish title: 'Smierc w raju'), for Clips D The Family Man and Clips B and C The Ultimate Gift ('Bezcenny dar'). Clips are examples of voice-over isochrony typically found on Polish television and the students were most likely more familiar with American English films. All three films were broadcast on the Polish television with the same voice- over technique as used for the experiment in the class. The videos were a few minutes long, consisting of several scenes to achieve an effect of unaware listening. The samples contained many instances of unfaithful translation. However, after watching the informants were asked about the last audible utterances in the clips, and specific target phrases (see Table 1 below).

(20)

Clip A and B had a similar level of difficulty, the target phrases in English were fully audible because the translation preceded them. In Clip C, the original utterance was interrupted by the lektor's voice. The original phrase was 3.1 second long, while the translation took a bit more than one second and was said almost exactly in the pause between words ‘but’ and 'I (can't)'. Finally, Clip D contained one utterance which was omitted by the Polish translation and was followed by an utterance with unfaithful translation. The last English utterance ('She took my bell') was fully audible.

Table 1. Target phrases of the experiment.

Original dialogue Lektor’s translation The word sequence heard by the learners

Clip A (4min 45sec)

Wow... That's quite a story

Niesamowite (Amazing)

Niesamowite - Wow... That's quite a story

Clip B (5min 04sec)

Can I help you? O co chodzi?

(What's up?)

O co chodzi?- (pause) Can I help you?

Clip C (5min 04sec)

I would love to help you but I can't

Nie mogę ci pomóc (I can't help you)

I would love to help you but - Nie mogę ci pomóc - I can't

Clip D (3min 07sec)

I need that back. She took my bell

Oddaj dzwonek (Give me back the bell)

I need that back. She took my bell - Oddaj dzwonek

In the fourth phase of the experiment (see Section 3.3.2), the participants were presented a short video from the BBC news in which it was the Polish language overridden by an English lektor. The volume levels of the original soundtrack and the lektor voice were not changed, however, some of the students complained about the echo effect in the classroom.

3.3.2. Implementation of the experiment The experiment was divided into four phases:

1. Watching without instructions.

In the first phase the students were asked to watch one 5- or 4-minute-long clip without being given any instructions or being informed about the purpose of the experiment: Clip A for Group 1 and Clip B for Group 2 respectively (see Table 2). At the end of the fourth or fifth minute, the original dialogue varied significantly from the lektor's translation. The clip was stopped just after the target phrases were said by the actors (see Table 1 in Section 3.3.1). The students were asked to write down what was said in the original English dialogue, the right answer should have been given in English if possible. The purpose of this part of the investigation was to find out whether the learners unintentionally listen to the original dialogues in L2.

(21)

2. Control viewing.

In the second phase, the clips were switched and Group 1 was presented Clip B and Group 2 was presented Clip A (see Table 2 below). Since the first phase of the experiment revealed the purpose of the study, Phase 2 examined if the audience could gain a better result when intentionally listening to the original soundtrack. Both clips had a similar level of difficulty. The goal of this exercise was to determine if there is any difference between intentional and unintentional listening to the original dialogue. Moreover, it was designed to verify if the students could possibly have any issues with understanding the L2 input.

Table 2. The order of viewing clips during the experiment.

Phase Group 1 Group 2

1 2 3 4

Clip A Clip B Clip C and D

Clip E

Clip B Clip A Clip C and D

Clip E

3. Watching with instructions.

Phase 3 included viewing two additional Clips, C and D. The target phrases in both clips are examples of different synchrony than target phrases of Clip A and B. In Clip C the translation was read in the pause between two sentences of the original dialogue. In Clip D the translation followed the target utterances. This part was designed to verify if there is any difference in reception of different synchrony.

4. Listening to an English lektor.

In the final stage of the experiment the pupils were shown Clip E with a reverse situation. Both groups watched a short 3-minute video in which the original soundtrack was Polish and the translation was English. After watching the clip the students were asked to determine which language they focused on. The purpose of this part was to examine if the students habitually focus on a lektor, who in this case spoke English, or if they paid attention to their L1.

The students were provided an answer sheet with blank spaces for their answers. The answer sheet did not contain any instructions or questions to make sure that the students were not able to guess the purpose of the study. Since the films were broadcast on the Polish television in the past, under each blank space the pupil was asked to specify if they could recall seeing the presented film or clip. However, none of the students remembered seeing any of the films.

3.3.3 Additional questions and free discussion.

In addition, on the back of the answer sheet, there were three additional questions for students to evaluate how often they pay attention to the original dialogues and how frequently they are able to spot the “errors” in translation. The students were also asked to specify their preferences of AVT (see Appendix B). These questions were the same as

(22)

Questions 9-11 in the survey for the Polish speakers of English (see Appendix A). After completing the final task, a free discussion was held on different types of AVT and their effect on the students’ L2 proficiency. During the discussion, certain students took the lead and others may have been suggested answers, and not all students took an active part in the task. The discussion was recorded.

Compared to the questionnaire, the experiment participants provided the answers immediately after being exposed to unfaithful translation. Since the experiment was conducted in a group, there was no possibility to elicit more information apart from the written answer.

3.3.4. Experiment limitations.

The material chosen for this experiment was previously aired on the Polish television, the sound levels were exactly the same as on the television and the lektor reading translation is one of the most popular voices in Poland (see Section 2.2.5). However, the classroom environment might have had an influence on the results of the study. Because of that the participants might have tried to get a high score and so they listened extra carefully. Due to the classroom context, it was not possible to estimate if the informants could have the same level of unaware listening as they would have in their leisure time.

4. Results

In this section the results of the survey will be presented, followed by the outcome of the experiment.

4.1. Survey results: Preferences for AVT types

The survey participants were asked what AVT they preferred in different genres. The AVT choice seems to be determined by the genre of the television content. The Polish speakers of English tend to choose subtitles or no AVT at all when watching English language films and TV series. The detailed results are illustrated in Table 3.

As can be seen in Table 3, the genre with the highest score for no AVT at all are TV series, most likely because the audience can get used to actors’ language, accents, or specific vocabulary throughout a certain period of time. It is worth noting that for documentaries, which contain more specialist language, half of the participants chose voice-over translation. Not surprisingly, more than half of the respondents chose dubbing for cartoons, it can be explained by the availability of dubbed cartoons, but also a great popularity of the Polish dubbing for animated comedy films, like Shrek, Madagascar, which engaged many famous Polish actors and comedians as well as involved many in- jokes.

The respondents who rated their level of English as fluent, tend to choose subtitles for feature films (57% of fluent respondents), or watch them without any AVT (36% of fluent

(23)

respondents). Furthermore, more than half of the fluent respondents chose no AVT for TV series. Even for cartoons, the speakers with a high L2 competence choose the original dialogues only. Communicative speakers of English prefer subtitles for all fiction genres apart from cartoons (62% would choose dubbing for animated films). Those with a basic understanding of English would mostly choose voice-over translation (44%) for fiction genres and documentaries, dubbing remains the most preferable AVT type for cartoons as well. The informant whose English competence is poor chose voice-over translation for all genres.

Table 3. AVT preferences of Polish speakers of English when watching English-language television.

Voice-over

translation Subtitles Dubbing No AVT

Feature films (all participants; n=94)

by proficiency:

fluent (n = 38)

communicative (n = 37) basic (n =18)

poor (n = 1)

16 (17%)

2 (5%) 5 (13%) 8 (44%) 1 (100%)

55 (58%)

22 (57%) 27 (73%) 6 (34%) 0 (0%)

0 (0%)

0 (0%) 0 (0%) 0 (0%) 0 (0%)

23 (25%)

14 (36%) 5 (13%) 4 (22%) 0 (0%) Documentaries (all

participants; n=94)

by proficiency:

fluent (n = 38)

poor (n = 1)

45 (48%)

14 (36%) 18 (49%) 12 (67%) 1 (100%)

30 (32%)

11 (30%) 15 (40%) 4 (22%) 0 (0%)

N/A

19 (20%)

13 (34%) 4 (34%) 2 (11%) 0 (0%) Cartoons (all

by proficiency:

fluent (n = 38)

poor (n = 1)

9 (10%)

2 (5%) 4 (8%) 2 (11%) 1 (100%)

13 (14%)

4 (10%) 5 (13%) 4 (22%) 0 (0%)

50 (52 %)

15 (39%) 23 (62%) 12 (67%) 0 (0%)

22 (24%)

17 (45%) 5 (14%) 0 (0%) 0 (0%) TV series (all

by proficiency:

fluent (n = 38)

poor (n = 1)

22 (24%)

3 (6%) 9 (24%) 8 (44%) 1 (100%)

37 (39%)

14 (36%) 20 (54%) 3 (17%) 0 (0%)

2 (2%)

1 (2%) 1 (2%) 0 (0%) 0 (0%)

33 (35%)

20 (53%) 8 (22%) 5 (27%) 0 (0%)

There is a clear connection between the L2 proficiency and choice of AVT. The more fluent L2 speaker is, the less they need to facilitate themselves with AVT. In documentaries, the language may be more specialized, thus, the L2 speakers choose AVT they are most used to, voice-over translation. The chi-square test determined that the choice of AVT is associated with the level of L2 proficiency for all genres (see Table 4 where all p values are lower than 0.005).

(24)

Table 4. Difference between the English proficiency and AVT preferences.

X² df⁵ N p

Feature films 19.2 4 93 0.001

Documentaries 9.7 4 93 0.044

Cartoons 17.5 6 93 0.008

TV series 18.5 6 93 0.005

The informants’ AVT preferences changed slightly when asked about non-English language television content (see Table 5). Only 5 informants were able to watch television content without any AVT, presumably because the rest of the informants did not gain enough language competence in another foreign language. Some of the participants chose not to indicate any preferable AVT for non-English films and programs. Similarly to the AVT preferences for English language films the most popular AVT are subtitles, dubbing for cartoons and voice-over translation for documentaries. However, nearly twice as many informants choose voice-over translation for non-English language feature films.

11 more informants would chose a lektor for non-English language TV series than for English language TV series.

Table 5. AVT preferences of Polish speakers of English when watching non-English-language television. (n= 94)

Voice-over translation Subtitles Dubbing No AVT Don’t watch such films

Feature films 30 (32%) 57 (61%) 1 (1%) 5 (5%) 1 (1%)

Documentaries 41 (44%) 34 (36%) N/A 5 (5%) 14 (15%)

Cartoons 13 (14%) 19 (20%) 41 (43%) 5 (5%) 16 (17%)

TV series 33 (35 %) 34 (36%) 1 (1%) 7 (8%) 19 (20%)

Concerning Question 9 asking specifically about the enjoyment of watching films with a lektor, 31% of the informants admitted that they enjoy watching motion pictures with voice-over translation, 53% stated to dislike this type of AVT. When asked if their desire to learn foreign languages influences their choice of AVT type (Question 17), the answers were more ambiguous. 43% of the respondents stated that motivation to learn a foreign language did not influence their choice of AVT. However, some of them commented that there were other factors that influenced that choice, for instance:

(1) While learning English at school, I didn't think about choosing a different type of AVT to improve my English. However, I still prefer to watch English-language films with no AVT, or possibly with Polish or English subtitles. (I36)

5 df value varies for genres because SPSSStatistics removed the categories that had 0, for example dubbing for feature films.

Lost in Translation

Lost in Translation

Voice-over Translated TV as a Source for Incidental Language Acquisition

Anna Justyna Sochacka

Lost in Translation

Abstract

Contents

1. Introduction

2. Literature review

3. Methods

4. Results