The intricacies of voicing over documentaries from English into Arabic: implications for translator training

Abstract

The present paper is designed to shed some light on one of the main subfields of audiovisual translation, namely voice-over, and mainly attempts to capture the intricacies involved in English–Arabic translation of a BBC television documentary entitled NW Great. The sample of the study consists of 10 MA translation students at Al-Quds University, enrolled in Audiovisual Translation I for the academic year 2016/2017. The paper confirms what Orero (2004) is inclined to conclude, that besides voice-over's main defining features, namely faithfulness and synchrony, etc., student translators are commonly faced with several technical (e.g., synchrony in voice-over, observable in the same way as Orero (2004) argues, close-ups, etc.) and linguistic problems and potential semantic and stylistic loss. The study also has a pedagogical angle thought to be useful to trainers and trainees of the audiovisual translation mode of voice-over.

Audiovisual translation, Arabic voice-over, subtitling, linguistic problems, cultural problems, technical problems.

1. Introduction

Holmes (1988b, 2004: 18) categorised audiovisual translation (AVT) as one of the “medium restricted theories”. A year later Delabastita (1989: 196) described it as “a virgin area of research.” It has since enjoyed a good deal of popularity among translation theorists and practitioners and has gained momentum and weight all over the world. Nevertheless, AVT's most common modes (subtitling, dubbing and voice-over) are full of many linguistic, technical, semiotic and cultural problems. In light of these problems, the job of the translator may be viewed in concrete terms as challenging; as Karamitroglou (2000: 104) convincingly argues, “the number of possible [AVT] problems is endless and a list that would account for each one of them can never be finite.” As a consequence, “no one has ever come away from a foreign film admiring the translation” (Nornes 1999: 17).

It may be assumed that the requirements for AVT exceed a translator of great sagacity and immense experience to a ‘techie’ because AVT is inevitably different from literary translation in that it requires a ‘translator plus’, that is, someone who has not only a linguistic competency but also a technical competency, i.e., someone who is au fait with technology-related issues. Thawabteh (2011: 25) points out that “alongside linguistic and cultural competence, a fully fledged [translator] should be a ‘whiz-kid’ computer expert so that technical constraints, of which AVT is full, can be handled.” In addition to cultural competence, i.e., “hands-on experience of living and breathing the way of life of another culture” (Skuggevik 2009: 198), Neubert (2000) speaks of four levels of translation competence: (1) language competence; (2) textual competence; (3) subject competence, which is mainly to do with the knowledge of outside world at the disposal of the translator; and (4) transfer competence. As for linguistic competency, Neves (2004: 135) argues,

Fidelity lies in two extreme points, the source-text or the target-text, in [AVT] fidelity is particularly due to an audience that […] is in need of communicative effectiveness, rather than in search of artistic effect— as is the case in literary translation— or of exact equivalence— as happens with technical translation.

For technical competency, however, translation is viewed from the vantage point of a technology-laden process. As Neves (2004: 135) puts it, “the problems which arise are somewhat similar to those of literary translation with the extra stress that the fidelity factor is dictated by constraints that lie beyond words or languages.”

Gambier (2003: 172) points out that the dominant types of AVT are “interlingual subtitling, dubbing, consecutive interpreting, simultaneous interpreting, voice-over, free commentary, simultaneous (or sight) translation, and multilingual production.” We shall see in the following discussion the difficulties involved in voice-over translation, as the current focus seems to be on topics different from the one under discussion; see for example Matamala (2019) addressing cultural and linguistic aspects, translation and synchronization techniques, ideology and manipulation, and the role of technology. The main objectives of the current paper are then to examine these difficulties (be they technical, linguistic, semantic and/or stylistic) encountered by MA translation students and to explore the strategies to overcome such difficulties. Typically, subtitling and dubbing constitute the most common translation practices in the Arab World, whereas voice-over has hitherto received minimal attention in the existing TV documentaries. We should admit that voice-over is not adopted in the Arab World for fiction films, but for non-fiction ones (e.g., documentaries and news interviews). Conversely, subtitling and dubbing are the most common modes. In this regard, Thawabteh (2011: 25) aptly remarks,

In the Arab World audiovisual programmes (e.g., sitcoms, documentaries, soap operas, TV series, cartoons, etc.) diversify mainly via two different forms of AVT- subtitling and dubbing. But “which is the preferred form?” is a question still not answered properly.

According to Volmar (as cited in Karamitroglou 2000: 132), “several governments of countries with large Arabic-speaking populations see in subtitling a means for encouraging the masses to learn to read and write,” and more importantly, subtitling is even viewed from a political vantage point: “the nascent Egyptian film industry, keenly aware of the competition coming from Hollywood, opted not to dub American films for fear of killing off the local industry” (Gamal 2009: 2). By the same token, Karamitroglou sees dubbing as a threat to the local film industry: “In Egypt in 1947, local film directors, actors and producers protested against the dubbing of American and other films into Arabic and called upon the Ministry of Social Affairs to pass a law that would forbid the release of dubbed foreign films” (Motion Picture Herald, as cited in Karamitroglou 2000: 132). However, according to Volmar (as cited in Karamitroglou 2000: 132), “[f]or languages like Arabic, for example, which many can understand but relatively few can read, revoicing offers the opportunity to the masses to enjoy TV without effort.” But it should be pointed out that subtitling is the cheapest option, economically speaking.

2. The praxis of voice-over

Termed as the ‘ugly duckling’ of AVT (Orero cited in Matamala 2019), voice-over is a mode of AVT in the circulation and consumption of documentaries and a site of dissemination of scientific ideas and knowledge. Although voice-over has its “relative simplicity […] due to the abandonment of the constraints imposed by lip synchronisation” (Luyken, 1991: 80), “[t]ranslation for voice-over has not been successful in attracting the attention of the academic community” (Orero 2004: 76; see also Díaz Cintas and Remael 2014: 41). Nor has it done so at the professional level; Grigaraviciute and Gottlieb (1999: 45) complain about the “monumental lack of professional and academic interest.” In the same vein, Orero (2004: 76) states that

In Gambier's 1997 bibliography, out of a total of some 1,300 titles only nine deal with voice-over. The translation studies on-line database BITRA […] had in February 2004 some 700 entries on AVT with nine of those paying attention exclusively to voice-over (see Endnotes 1).

Another search for ‘voice-over’ in the Translation Studies Bibliography, which took place between April 11 and April 24, 2015, returned 33 and 45 hits in abstracts and keywords, respectively (Rosa 2016: 196). More to the point, Díaz Cintas and Remael (2014: 41) touch on the state of AVT:

Despite the growing importance of AVT in our daily lives, many universities have been rather passive in the preparation of students in this area, and dubbing, subtitling and voice-over have been largely ignored in the curricula. Lack of interest, prohibitive software prices, absence of teacher expertise, vested interests, or mere blindness may be some of the reasons behind this state of affairs.

In the Arab World, studies pay scant attention to AVT. A similar search in BITRA1 returns one article on AVT with the word ‘Arabic’ in the title, two on subtitling and no articles on dubbing or voice-over.

As a point of departure towards a more operational definition for voice-over, Franco et al. (as cited in Darwish and Orero 2014: 5–6; see also Luyken 1991: 80; Gambier 2016: 898) define the features of voice-over as follows:

• 1.is mainly applied to non-fictional audiovisual programmes, although in certain East European countries it is also used in fictional programmes;
• 2.renders the words of interviewers/interviewees (dialogues)/talking heads (monologues);
• 3.is the revoicing of a text in another language, or a translating voice superimposed on a translated voice;
• 4.is oral or spoken rendering that is delivered simultaneously and in synchrony with original speech length, recognizable words and images (kinetic/action synchronicity);
• 5.does not account for lip synchronization;
• 6.usually starts a few seconds after the original;
• 7.is prepared and recorded before programme broadcasting, so never produced live;
• 8.is derived from unedited material (production voice-over) or from edited material (postproduction voice-over);
• 9.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 voiceover translation);
• 10.can personify the original speaker (first person voice-over) or report his/her words (third person voice-over);
• 11.reproduces mimetic features to a certain extent (accent, age, emotion, gender, intonation, orality markers, stress);
• 12.keeps its performer (voice talent) invisible.

It might be worthwhile to differentiate between narration and voice-over. The former is an “extended voice-over” in a “more formal grammatical structure” (Luyken 1991: 80). Díaz-Cintas and Anderman (2009: 4) state that “the original dialogue is replaced by a new soundtrack in the target language in a process generally known as revoicing. The replacement may be total, whereby we do not hear the original, as in lip sync dubbing and narration,” whereas in the latter, it can be “partial, when the original soundtrack can still be heard in the background, as in voice-over”. More precisely, Luyken (1991: 80) states that:

The faithful translation of original speech, approximately synchronous delivery, used only in the context of monologues such as an interview response or a series of responses from a single interviewee. The original sound is either reduced entirely or to a low level of audibility. A common practice is to allow the subsequently reduced so that the translated speech takes over […] alternatively if the translation is recorded as part of the original production, it may follow the original speech exactly.

Unlike Luyken (1991: 80) and Díaz-Cintas and Anderman (2009: 4), Karamatriglou (2000: 6) adopts a rather different view on the distinction of narration and voice-over:

I do not find […] the argument sufficiently convincing to justify “voice-over” constituting an independent category from “narration” since a distinction based on the “extendedness” of the target text or the “formality” of its grammatical structures would be completely arbitrary, not to mention the fact that the number of voices involved do not interfere at all with the translation strategies to be followed in either method.

With this quote in mind, we come up with a rigorous and comprehensive view of voice-over as expounded by Luyken (1991) and Baker (2018), but in terms of fidelity, it is argued that “fidelity and synchrony are two concepts rarely observed in this field” (Orero 2004: 79), though “the translation tends to be reasonably faithful to the original and quasi synchronous” (De Linde and Kay 1999: 46). With regard to synchrony of the original utterance with non-verbal components, Rosa (2016: 201) believes that “the consideration of non-verbal components is all too often reduced to the question of how to create (as much as possible) a synchronous target text (be it subtitled, dubbed, voiced-over or interpreted).”

It might also be worth mentioning that “simplification and a lack of understanding of the media and its process have made voice-over to be seen in the same light as dubbing, which is certainly a different mode, subject to different translation and production processes” (Orero 2009: 131).

It ensues, therefore, that it is very important to contemplate a well-thought-out training that corresponds well to different professional orientations, since voice-over has profoundly and rapidly gained momentum and visibility in non-fictional output (e.g., news interviews, documentaries and commentaries), as can be seen on many Arab satellite channels, e.g., Al-Jazeera Documentary and Abu Dhabi National Geography, among many others. These two channels bulge with round-the-clock voice-over translations of documentaries.

3. Academisation of voice-over

Translation studies has had a relentlessly upward trajectory in the Arab World since time immemorial. By the same token, attention-grabbing AVT practices (particularly subtitling and dubbing) in the Arab World have been Ubiquitous for the past few decades (see Karamitroglou 2000; Al-Adwan 2009; Gamal 2009), but voiceover has been much less popular due to the dearth of educational programmes, and even worldwide, “hardly any preparing their students for voice-over” (Díaz-Cintas and Orero 2003: 371). Moreover, “[t]ranslation for voice-over has not been successful in attracting the attention of the academic community” (Franco & Orero as cited in Orero 2004: 76). Voice-over, according to Khoshsaligheh and Ameri (2016), has failed to receive academic recognition. To the best of our knowledge, the existing voice-over training programmes are only at the fringes of two Arab universities. Hamad bin Khalifa University offers a unique MA programme in AVT, which is the only one in the MENA region. Among the courses taught at this university is a specialised obligatory course called Voicing, in which students are introduced to and learn the basics of voice-over and dubbing. The skills of voicing are further enhanced and polished in two more courses, namely Advanced Dubbing and Internship. The second institute is Al-Quds University, which offers two courses on AVT, covering subtitling, dubbing and voice-over. Here students are conceivably trained to voice over an audiovisual text (e.g., a documentary or an interview) in a purposeful manner. Gamal (2020) addresses other new academic programmes in the Arab World. A master's degree programme was established at Effat University, Saudi Arabia, in 2016 (see Endnotes 2). This programme includes one technical elective course in the second year (semester 5) on Dubbing and Subtitling.

4. Voice-over software for Arabic

The advent of digital technology has contributed to the future health of AVT, including especially the voice-over mode, which is used for many languages, including Arabic. One of the free pieces of user-friendly software which are readily available is Windows Movie Maker (various versions). Meanwhile, at Hamad Bin Khalifa University, students use the professional software Adobe Audition CC 2019 in their in-class voice-over assignments and projects. They usually record their parts in well-equipped labs and then submit their recordings to their department's audiovisual specialist for mixing and mastering. Having borne in mind the all-encompassing definition of voice-over offered by Franco et al. (as cited in Darwish and Orero 2014: 5–6), student translators may start training thereof. The first embryonic stage takes place within traditional translation practice classes, where students may watch the video, do all kinds of analyses, look up difficult words in various dictionaries, consult as many online sources as possible, write down the translation, etc. We may term this process as ‘unedited material’. The second stage is likely to be technology-related classes; fastidious attention here is given to acquiring suitable technological competence, e.g., the student translator can ‘ferret out’ details in Help, upload a video, start a voice-over session and save work regularly. We may term this process as ‘edited material’.

5. Methodology

The major objective of this study is to provide an overview of the intricacies of voice-over translation from English into Arabic. The data comprise a 5-min BBC documentary entitled Great NW, with an apparently global reach. To pinpoint the problem under discussion, a sample of 10 MA translation students was chosen from Al-Quds University for the academic year 2016/2017. All students had a BA in English Language and Literature, and Arabic was their native language. The students first watched the video and then were asked to voice it over into Arabic using Windows Movie Maker software. Then the transcription of their work (i.e., into Arabic) was generated for the sake of appropriate discussion as can be illustrated in Discussion and Analysis section below. The English script is provided in the Appendix.

Our discussion will mainly encompass two points: first, the ways student translators go about voicing over the documentary of this study will hopefully be beneficial for designing training programmes on voice-over translation. The second point has to do with the implications drawn from the analysis of the experimental part that will hopefully be conducive to a more optimal voice-over translation.

6. Discussion and analysis

Although “the lack of difficulty when translating for voice-over is seen by many as a result of belonging to a particular genre, away from fiction and firmly positioned within nonfiction” (Orero 2004: 77), we identify two major voice-over problems insofar as English-to-Arabic translation is concerned: technical problems and unnatural translation problems. The former refer to aspects of the production and postproduction of the audiovisual material (e.g., synchrony), whereas the latter include problems related to translation, e.g., elocution features, literal translation, insensitivity to context, ungrammaticality, unnecessary formality, stylistic loss and potential semantic loss (see also Luyken, 1991: 1055). Based on an empirical investigation of voice-over translations of television documentaries, translation problems also include what Franco (2001) terms “terminological and conceptual difficulties”. The problems identified are based solely on sound empirical evidence from our observations.

6.1. Technical problems

Desktop computer users need software that enables them to do their translational tasks as efficiently and effectively as possible. However, computer programs entail technical barriers to translation for which the translator must seek solutions. For example, the desktop software Windows Movie Maker necessitates “unedited material production voice-over” and “edited material postproduction voice-over” (Franco et al. as cited in Darwish and Orero 2014: 5–6). This may create problems for the voice-over translator, e.g., low-quality voice, undesirable periods of silence in the voice-over translation, or having a clipped cadence, among others.

6.1.1. Synchrony

“An approximate synchrony between original and translation also helps to give the programme the feeling of ‘authentic’” (Orero (2004: 79). Generally speaking, synchrony is pertinent to the complementary aspects of AVT (Sepielak and Matamala 2014).

This problem is ascribed to failure to successfully master the programme. It is paramount that edited text be synchronised with the images, which is not the case in most of the examples below. Thus, the feeling of authenticity is entirely lost. One example has the voice-over of “the American pioneers travelling west to California met this unforgiving desert” taking place while pictures of exotically beautiful scenery are shown, thus failing to preserve the coherence of communication between ST dialogue and voiced-over translation. It is then true, as Darwish and Orero (2014: 21) state: “The lack of synchronicity of voice-overs stems from the rhetorical features of the voice-over styles of delivery and the idiosyncrasies of the voices.”

6.2. Unnatural translation problems

It goes without saying that language competence is a “feature which determines the degree of simplicity when translating” (Orero 2004: 82) in all modes of translation. Orero (2004: 82) further argues that ‘elocution features’ “can be added [as] the language habits commonly found in the delivery of individual speakers.” Insofar as our sample is concerned, only some (totalling five) have actually made good public speakers. The tool used to measure it is based on our analysis of the examples. It is important for the trainee to have a good share of training to be competent in a language and as a public speaker. Lertola (2019: x–xi) suggests that “revoicing tasks prove effective in improving learners' speaking skills, in particular pronunciation, intonation, awareness of prosodic features, and fluency.” Prior to the AVT course, all the students had already studied rudimentary stylistics of communicative acts in at least one interpreting course, whose objectives were to train for public speaking and improve native language competency, among others. By the same token, Hussain and Khuddro (2016: 18–19) suggest that it is a must to hear the first and last couple of words of the original when voicing over a sound-bite, as opposed to voicing over a narration or commentary, in order to give credibility to the TT.

Elocution-wise, voice quality and accent of Arabic tend to be problematic. Although all the students were Palestinians, they came from different cultural backgrounds and social classes, which may influence delivery, intonation and accents. For example, Hebron students had a strongly marked regional accent. Their pronunciation of words is always elongated: a long-drawn out process. That is fine for the sake of training. Tilting towards a particular training that would develop a slight Modern Standard Arabic (MSA) accent can be made. It has been observed by the authors that many of the voice-over translations on well-known satellite channels like Al-Jazeera are in MSA. Orero (2004: 81–82) believes that “[a]ccents can be another hindrance to understanding the original because of regional variations”, and this may be consequence of a voice-over translation that is not intelligible insofar as the target audience is concerned. It is then preferable to have a neutrally accented voice-over for the convenience of the target audience; an effective way could be via “spontaneous speaking pre- and post-tasks” (Lertola 2019: 41).

Another type of oft-repeated problem has a bearing on a clipped speech effect, whereby students may have a clipped way of speaking that their clipped tones crackled over. This is undesirable, as voice-over translation's ultimate goal is to maintain the flow of intercultural exchange in communicative transactions. To diversify our argument, let us now consider Example 1 below:

Example 1.

ST	From the everglades of Florida in the east,
TT
a.	من سهول أيفرقليدز في فلوريدا من الشرق Back-translation (BT): From the fields of everglades in Florida in the east
b.	من محمية أيفرقليدز في شرق فلوريدا BT: From the nature reserve of everglades in east Florida
c.	من مستنقعات فلوريدا الشرقية BT: From the swamps of eastern Florida
d.	من سهول فلوريدا شرقا BT: From the fields of Florida in the east
e.	من سهول فلوريدا الفسيحة شرقا BT: From the vast fields of Florida in the east
f.	من سهول فلوريدا الخضراء شرقا BT: From the green fields of Florida in the east

A close look at Example 1 above shows various voice-over translations. One is a literal translation which, according to Nida and Taber (1969), aims to reproduce the form of the original in the receptor language. In a sense, literal translation “occurs when there is an exact structural, lexical, even morphological equivalence between two languages […] and this is only possible when the two languages are very close to each other” (Vinay and Darbelnet, as cited in Molina and Hurtado Albir 2002: 499–500). It is perhaps worth noting that in the case of languages with little linguistic and cultural affinity (as is the case for Arabic and English— the former belongs to the Semitic language family, whereas the latter belongs to the Indo-European language family), such a strategy brings about awkward translations. The literal translation procedures are as follows: (1) Borrowing, whereby a word is taken directly from another language. In TTa and TTb, there is a clear tendency by student translators to hedge their bets by translating ‘Everglades’ via a loanword which narrows the phonological distance between English and Arabic as much as possible, giving rise to أيفرقليدز (/‘ayfarglīdz/). It should be noted here, however, that the two languages drastically vary in terms of consonants, vowels and diphthongs, syllable structure stress and rhythm, etc. The student translator attempted a signifier that would travel across cultures and languages into Arabic. (2) Calque, whereby a foreign word or phrase is translated and incorporated into another language (Vinay and Darbelnet, as cited in Molina and Hurtado Albir 2002: 499–500); in TTc–TTf, “[t]he concept is translated word-for-word while the translation conforms to the syntax of the borrowing language” (Armstrong 2005: 146). (3) Literal translation, whereby word-for-word, clause-for-clause or sentence-for-sentence translation is used (Vinay and Darbelnet, as cited in Molina and Hurtado Albir 2002: 499–500).

A mere glance at the translation in TTd would immediately reveal naturalness, but adding more adjectives would add more to the cohesion of the moving picture and would pave the way for flowery elocution. In TTe and TTf, the student translators opted for الفسيحة (lit. ‘vast’) andالخضراء (lit. ‘green’), respectively, with remarkable uniformity, as the moving picture shows the vastness and greenness of the Everglades of Florida.

In TTc, however, the gratuitous rendition “من مستنقعات فلوريدا الشرقية” (lit. ‘from eastern swamps of Florida’) seems to be unrelated to the picture—a head-on clash with other semiotic elements (i.e., no swamp is present). This kind of translation is the epitome of insensitivity to context on the part of the student translator. “The dissonance created by the lack of synchronicity renders the original message with a degree of infelicities” (Darwish and Orero 2014: 21). For more illustration on insensitivity to context, consider Example 2 below:

Example 2.

ST	to the great forests and mountains of the west. This continent has many rich and varied landscapes
TT
a.	إلى الغابات والجبال العظمى غربا، تزخر هذه القارة بمناظر طبيعية متنوعة BT: to the wondrous vast forests and vertiginous mountains of the west. This continent has many majestically beautiful landscapes
b.	إلى الغابات الواسعة والجبال الشاهقة غربا، تزخر هذه القارة بمناظر طبيعية متنوعة BT: to the vast forests and mountains of the west. This continent has many majestically beautiful landscapes

In TTa and TTb, the translation flows naturally with attitudinal values from the moving picture, perhaps with the exception of lexical selection, i.e., non-collocation Arabic الغابات والجبال العظمى (lit. ‘great forests and mountains’), and is concomitant with the visual elements in the documentary. To elaborate more on the point, consider Example 3 below:

Example 3.

ST	But it is the south-western part of the United States that is perhaps most distinctively American.
TT
a.	ولكن القسم الجنوب الغربي من الولايات المتحدة التي هي ربما الاكثر امريكا BT: But it is the south-western part of the United States that is perhaps most differently American.
b.	ولكن القسم الجنوب الغربي من الولايات المتحدة يعتبرامريكا بامتياز. BT: But it is the south-western part of the United States that is deemed most distinctively American.
c.	ولكن القسم الجنوب الغربي من الولايات المتحدة هو لربما الاكثر تميزا في امريكا. BT: But it is the south-western part of the United States that is perhaps most distinctively American

Example 3 above merits close investigation. In TTa, the student translator seems not to be conscientious about what exactly “is perhaps most distinctively American”. The intentionality of the original dialogue is likely to show swashbuckling adventure story, so to speak. The referent هي (lit. ‘she is’) is quite astounding—breaking the communicative thrust of the utterance down, thus the ungrammatical sentence choice and irretrievable stylistic loss can most clearly be observed. It is perhaps particularly noteworthy that the subtle use of modality هو لربما (‘he is perhaps’) is perforce added in the translated dialogue in TTc, which makes the whole translation sound wordy. Take Example 4 for more elaboration:

Example 4.

ST	Here are the deserts and canyon-lands of the “Wild West”, just famous for so many of the country's most extraordinary natural wonders.
TT
a.	وهي ايضا مشهورة للعديد من الدول كاحدى عجائب الطبيعة. BT: Just famous for so many countries as most extraordinary natural wonders
b.	وهنا الصحاري والاودية في البراري الغربية، تعد احدى عجائب الطبيعة. BT: Here are the deserts and valleys of the “Wild West”, deemed to be most extraordinary natural wonders.

In Example 4 above, particularly in TTa, semantic loss is clearly observed. The translation does not tally with the overall picture composition of the documentary, for example, the long, narrow valley with steep sides, i.e., ‘canyon-land’, is likely to leave the target audience to forge its own ‘inappropriate’ interpretation. In fact, a different picture inevitably emerges.

The Arabic device serving as a conjunction/wa/with a relative pronoun, most likely with a view to establishing lexical cohesion, is also entirely missing from the target text in TTb (i.e., تعد ‘is deemed’), thus perhaps inevitably and knowingly sacrificing the coherence of the sequence in the ST.

In Example 5 below, neither TTa nor TTb does the trick in terms of a sequence flowing naturally in the TT. At first glance, in TTa, the non-finite ‘travelling’ being rendered as the finite السفر استطاع … (‘could … travel’) seems to easily relay the message quite beautifully, but the ST dialogue identifies only those American pioneers who travelled west to California, whereas the TT does not. A grammatical error happens in the use of the accusative الامريكيين ‘Americans’. The nominative case should be used here as the stretch of TTb succinctly shows, i.e., الامريكيون ‘Americans’.

Example 5.

ST	the American pioneers travelling west to California met this unforgiving desert.
TT
a.	استطاع الرواد الامريكين السفر غرب كاليفورنيا والوصول الي تلك الصحاري القاسية. BT: The American pioneers could travel to west of California until reaching this unforgiving desert.
b.	اكتشف الرواد الامريكيون تلك الصحاري القاسية. BT: The American pioneers discovered these unforgiving deserts.

TTb sounds awkward and ‘obstreperous’, as it falls short of the original as well.

7. Subtitling-cum-voice-over strategies

In this section, we refer to subtitling strategies that partly apply to voice-over. Taking our cue from Gottlieb (1998), subtitling strategies are tenfold:

• 1.expansion: expanded expression, adequate rendering, e.g., culture-specific references;
• 2.paraphrase: altered expression, adequate content, e.g., non-visualised language-specific items;
• 3.transfer: full expression, adequate rendering, e.g., slow, unmarked speech;
• 4.imitation: identical expression, equivalent rendering, e.g., proper nouns; international greetings;
• 5.transcription: non-standard expression, adequate rendering, e.g., dialects; intended speech defects;
• 6.dislocation: differing expression, adjusted content, e.g., musical/visualized language-specific items);
• 7.condensation: condensed expression, concise rendering, e.g., mid-tempo speech with some redundancy;
• 8.decimation: abridged expression, reduced content, e.g., fast speech; low redundancy speech;
• 9.deletion: omitted expression, no verbal content, e.g., fast speech with high redundancy);
• 10.resignation: deviant expression, distorted content, e.g., incomprehensible or ‘untranslatable’ speech.

Ventola et al. (2004: 159) adopt only reducing strategies, namely the seventh, eighth and ninth. This is quite reasonable, as subtitles on screen have to be condensed due to spatial and temporal constraints (Al-Adwan, 2019). In fact, there is no change in the mode in voice-over (i.e., spoken into spoken), in contrast to subtitling (i.e., spoken into written). In Gottlieb (2004: 219) terms, ‘isosemiotic translation’ “uses the same semiotic channel – i.e. channel of expression – as the original, and thus renders speech as speech and writing as writing.” It is perhaps futile to adopt only reducing strategies in voice-over, as non-reducing strategies may work pretty well. In this regard, Khuddro (2018: 4) argues that “the application of effective strategies, such as shortening, condensation, or compactness […] can be used in subtitling, but not at the expense of clarity. One cannot compromise clarity for the purpose of compactness and condensation.” Hussain and Khuddro (2016: 19) further add that “a standard duration of a subtitle is about 3–5 s but could be up 7 s; whereas reading one line of TT (about 12 words) for a voice-over recording would take roughly 5 s. This means that the translator has the job of cutting down all the unnecessary words.” For example, in Text 1 TTe and TTf, one may go for the strategies of expansion, transfer of full expression and imitation to bring something like من سهول فلوريدا الخضراء الفسيحة شرقا (lit. ‘From the everglades of Florida in the east’). Another example is in Text 4 TTb, where an expansion strategy has been used. In Text 2 TTa and Text 5 TTa, it is convenient to opt for formal translation through the imitation strategy. To cut it short, we can set a claim, insofar as the data of the study are concerned, that reducing strategies–cum–non-reducing strategies seem to have potential in voice-over.

8. Concluding remarks

The foregoing analysis has shown that voice-over problems can be technical or include unnatural translation. The task of the voice-over translator is fraught with peculiar perils. Franco, Matamala, and Orero as cited in Sepielak (2016: 1058) propose ‘literal synchrony’, referring to “a literal translation of the audible words”, which may lead to “word-by-word translation [resulting] in unconventional phrasing or alien syntax that, instead of strengthening the illusion of authenticity, could have the opposite effect; therefore, it is sometimes not preferred.” In our data, examples include lack of authenticity, as in Text 3, TTa, where there is a gender mismatch between القسم (lit. ‘part’ masculine) and the pronoun هي (lit. ‘she’ feminine), and in the ‘alien syntax’ of run-on Arabic sentence in TTa, Text 2.

As implications for translators, students should be trained to be public speakers. Insofar as our sample is concerned, students may be given ample time to read a text (preferably a technical one) aloud in front of the class or at home several times before starting voice-over. Besides the need to practice some vocal and presentation skills, there has been a need to develop the software competency, working towards a standard variety and applying correct translation techniques. Franco et al. (2010: 139), for instance, speak of three levels of specialization: “the subject matter, the format (audio and visual) and finally technical specificities which support the audiovisual material (software programs, audiovisual formats, translation templates and clients’ requests).”

Academically, voice-over is a neglected AVT mode that still needs to be streamlined in the Arab world. Attempts are being made to give it a jump-start at Hamad bin Khalifa University and Al-Quds University in the hope that the lifeline to the job market could be promising.

Finally, the current study is limited to the intricacies of voice-over of an English documentary entitled Great NW translated by graduate students into Arabic, so the findings can only be confined to the Arabic context.

9. Endnotes

• 1.Available online at: =http://aplicacionesua.cpd.ua.es/tra_int/usu/buscar.asp?idiomaen [visited on January 4, 2021]
• 2.2. Available online: in https://www.effatuniversity.edu.sa/English/Academics/Graduate/CoSH/Pages/default.aspx [visited on January 4, 2021].

Declarations

Author contribution statement

M. Thawabteh: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

A. Al-Adwan: Conceived and designed the experiments; Analyzed and interpreted the data; Wrote the paper.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

Data included in article/supplementary material/referenced in article.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.

Appendix.

From the Everglades of Florida in the east, to the great forests and mountains of the west, this continent has many rich and varied landscapes. But it is the south-western part of the United States that is perhaps most distinctively American. Here are the deserts and canyon-lands of the "Wild West", just famous for so many of the country's most extraordinary natural wonders. This is one of the hottest spots on Earth. It's also one of the lowest and the driest. Death Valley fully deserves its awesome reputation as one of the most forbidding places on the planet. But it's also a place of startling beauty. During the gold rush of 1849, the American pioneers travelling west to California met this unforgiving desert. They gave its shimmering landmarks very telling names: Starvation Canyon, Death Man's Pass, Funeral Mountains, Coffin Peak, Hell's Gate. Today, a century and a half later, this same inhospitable landscape attracts thousands of visitors. Death Valley is now one of the United States many national parks. A concept that the Americans were the first to devise and put into practice.