RSS feed

Beyond the dialogue: A study of text-on-screen translation strategies in English-Spanish dubbed audiovisual products

Noemí Barrera-Rioja, Universitat de València

ABSTRACT

The use of text on screen (TOS) in audiovisual material is widespread. Due to technological developments, it is also expected to grow. Since TOS is commonly used as a narrative device in audiovisual products, translating it is crucial for the distribution of audiovisual products to foreign markets. However, this practice can be challenging for a range of financial and technical reasons. This article explores approaches to TOS translation and their impacts on the potential for viewers to appreciate translated audiovisual material. Drawing on examples from four television series, the range of translation strategies for TOS in English-Spanish dubbed programmes is examined whilst discussing the advantages, shortcomings and challenges of using different strategies from both translators’ and viewers’ perspectives. It is suggested that translators occasionally resort to a strategy (verbalisation) that is advantageous for the audiences of dubbed products but has not been previously studied by scholars. It is also argued that not all TOS translation decisions are determined by translators and that respecting both the content and visual aspects of TOS positively impacts the viewing experience. Finally, it is claimed that more research is required to improve TOS translation practice and thus the quality of the viewing experience.

KEYWORDS

Audiovisual translation, dubbing, text on screen, translation strategies, visual-verbal elements.

1. Introduction

In the 2000s, research approaches to audiovisual translation (AVT) remained predominantly linguistic (Gambier 2008). Currently, however, language is analysed along with other aspects of audiovisual texts, such as aesthetics and culture (Di Giovanni and Romero-Fresco 2019). Despite this shift in the perception and study of audiovisual texts, research on linguistic elements remains heavily focused on the acoustic-verbal and rather neglects the visual-verbal. As Giménez López (2013) argues, despite its challenging nature in AVT contexts, the amount of research conducted on written language seems scarcer than on spoken language.

Visual-verbal elements are not only increasingly present in audiovisual material but also tend to comprise information relevant to its comprehension (Giménez López 2013). However, translating such elements can sometimes be challenging for a range of technical (Fox 2016) and/or budgetary issues (Molerov 2015). Additionally, deciding whether a text element is important enough — and if so, whether it requires translation (Fox 2016) — can also be problematic. Therefore, it seems desirable to explore the causes of translation problems posed by TOS in audiovisual products, as well as their potential solutions.

Accordingly, the main objective of this article was to perform a critical assessment of the translation strategies1 employed to translate TOS in various television shows dubbed from English into Peninsular Spanish (henceforth Spanish). More specifically, this article examines the range of options available to translators to deal with TOS, the reasons to use a particular strategy when addressing the challenges posed by TOS translation, and the potential implications of these decisions for the target audience and their appreciation of television programmes. Since this is a descriptive study, it seems relevant to emphasise that the points made in relation to the viewing experience are based on assumptions about how viewers will react.

A qualitative analysis was conducted. The set of examples analysed comprises 40 instances of TOS extracted from four television shows chosen based on their genre, popularity and release date. Given the scope and purpose of this study, only instances of displays and inserts from English into Spanish are examined. These were extracted from the dubbed versions of these shows offered by the streaming service Netflix.

2. The concept of TOS

In most films, the visual channel does not only transmit non-verbal information but can also present visual-verbal information. Thus, it is essential to draw a clear distinction between written and spoken language in audiovisual material. As illustrated by Zabalbeascoa (2008: 23), audiovisual texts have four different components, depending on whether these are audio or visual and verbal or non-verbal. This clearly defines the difference between audio verbal elements, which he refers to as “[w]ords heard,” and visual verbal elements or “[w]ords read.”

The terms commonly used in Audiovisual Translation Studies to allude to ‘words read’ aretext on screen (TOS, Díaz Cintas and Remael 2014) and text element(Fox 2016)2. These refer to any written text that appears on audiovisual products (Delabastita 1989, Díaz Cintas and Remael 2014, Fox 2016, Giménez López 2013). According to Fox (2016), there are seven types of text elements: film titles, opening and closing credits, displays, captions, narrative texts, inserts, and subtitles3. The set of examples analysed in this paper only includes displays and inserts.

Adisplay is a “text that is an integral part of the picture” (Díaz Cintas and Remael 2014: 246) that is relevant to the plot and has not been added during editing but recorded by a camera (Ivarsson and Carroll 1998). One example of this is the INTERVENTION banners used in How I Met Your Mother (HIMYM). Inserts can also be regarded as a relevant part of a story; however, in contrast to displays, these are “superimposed” and “added in the post-production process” (Fox 2016: 256–257). Multiple deduction scenes in Sherlock are an example of this.

3. The translation of TOS
3.1. Challenges

While TOS translation has arguably not represented a major issue in translation practice, further research is desirable so that TOS is not only rendered but rendered adequately since this could play a vital part in offering a more viewer-friendly experience.

One of the reasons for this is that TOS normally carries information necessary for the comprehension of audiovisual products (Giménez López 2013). As will be explained, another reason is that the aesthetics of the target product can be significantly affected by the range of options available and the choice of one or another. Furthermore, it is currently highly unlikely to find TOS-free audiovisual material since written text is used with great frequency. For instance, Fox (2016) observed 1170 text elements in a corpus of 52 films. Additionally, their presence is likely to grow as technological developments favour the evolution and inclusion of TOS.

Therefore, to transfer an audiovisual product’s content, identity, tone and atmosphere to the target language (TL), translating the dialogue alone does not suffice; instead, also translating — and sometimes even recreating— relevant text elements is required (Fox 2016: 248–249). However, this appears to pose challenges given the wide variety of text elements, contexts and conditions in which they can be presented, as well as the complexity of the TOS translation process.

Dealing with TOS “represents a challenge that goes beyond strictly translation problems” and involves financial and technical matters (Molerov 2015: 13, author’s own translation). For example, although the translation and recreation of inserts are technically possible, “each graphic editing requires more time and the commissioning of a graphic studio,” which also increases the cost (Molerov 2015: n.p.). Therefore, when dealing with TOS, not all steps and decisions are made by translators, with some depending on external agents such as the client or studio. Thus, the translator must adapt.

Another challenge is the power of the image, which conveys meaning yet cannot be modified, thus commonly limiting the translator’s options (Martínez Tejerina 2012). Therefore, when TOS cannot be edited, the translator’s leeway will also be restricted. This could be the case for displays, which cannot be edited unless immense cost and effort are invested or the audiovisual product at hand is an animated film (Fox 2016: 255).

These are conceivably the most generalised TOS translation problems. However, other difficulties intrinsic to each strategy or specific to each text element can arise. These will be dealt with in upcoming sections.

3.2. Strategies

In audiovisual texts, “most of the time [translation] solutions are restricted to manipulations on the verbal plane alone” (Zabalbeascoa 2008: 33). Nonetheless, when referring to translation operations, a distinction between acoustic-verbal and visual-verbal elements should be made more frequently and clearly since, as previously stated, modifying the acoustic-verbal elements appears more habitual and feasible than modifying TOS from a technical and monetary perspective.

According to existing research, there are three options for translating a text element: subtitling it, offering a voiced-off rendition, and replacing it with its target text (Molerov 2015; Fox 2016; Martínez Tejerina 2016). In audiovisual products dubbed into Spanish, the translation of TOS is frequently presented in the form of either subtitles or voice-off (Martínez Tejerina 2016); however, it appears that substitution has occasionally also been employed, e.g. in German (Molerov 2015) and Spanish translations (forthcoming) of the BBC’s Sherlock.

Although barely mentioned in the literature about TOS translation, there is a fourth strategy: omission or no translation(Fox 2016). Translators must be able to detect the most relevant source text items in terms of content and function since this is vital to making “informed, context-sensitive, function-oriented, audiovisually-coherent decisions” (Zabalbeascoa 2008: 33). Applied to TOS translation, this would imply that an understanding of the function of each text element is necessary to adopt the most appropriate approach to their (non-)translation. As a result, translators may occasionally decide to leave specific instances of TOS untranslated.

4. Methodology

The present study is descriptive and qualitative given the nature of the research questions that it aims to answer, which are presented as follows:

    RQ1. Which strategies are employed to translate TOS?

    RQ2. What are the reasons to choose each strategy when translating TOS?

    RQ3. What might be the potential implications of these decisions for the target audiences?

For this purpose, a set of TOS instances encompassing displays and inserts was examined. This set of examples was limited to these two categories for three reasons: (1) the broadness of the research questions; (2) time and space limitations; (3) the presence of these two types of text elements in the analysed products being considerably higher than the remaining text element types.

Based on the aims, scope and approach of this study, out of the 82 examples identified, 40 were analysed. These were selected as follows: (1) by discarding those irrelevant to the analysis (primarily those with little or no narrative value since, based upon anecdotal observation, they are left untranslated in most cases due to this lack of narrative value and thus do not provide particularly relevant information for the discussion at hand); (2) by using a stratified sampling technique to avoid bias in choosing instances and ensure that the sample is varied (Saldanha and O’Brien 2014). For this purpose, two different strata were distinguished, as the instances were classified according to the type of text element. The instances were selected randomly by assigning a number to each and using an online random number generator.

The instances were extracted from four television shows available on Netflix that were originally produced in English and dubbed into Spanish. These included the following: Seasons 3 and 4 of the American romantic situational comedy How I Met Your Mother (2007–2009); seasons 1 and 2 of the British crime series Sherlock (2010–2012); seasons 8 and 9 of the American situational comedy Modern Family (2016–2017); seasons 4 and 5 of the British science-fiction anthology Black Mirror (2017–2019).

When compiling corpora in Translation Studies, the general aim is for “a balanced representation of the population” (Saldanha and O’Brien 2014: 73). In line with this, the choice of these television shows and seasons was based on the assumption that by considering a range of different types of programmes and seasons, a considerable variety of examples could be identified in terms of nature of the text elements (text messages, banners, etc.), as well as their timespan, genre, and the translation strategies adopted. Moreover, these programmes were selected due to their popularity (IMDb 2019) and fairly recent production. Thus, their use and translation of TOS may be more likely to have an impact on the television industry of the present and future than less popular or older programmes.

Consequently, the data set described provided a sound basis for discussion and was used to perform a qualitative analysis of the translation challenges normally presented by TOS, as well as of the advantages, disadvantages and potential impacts of the strategies typically employed.


 

5. Findings and discussion

Before delving into the qualitative analysis, it is important to provide some frequency figures for the use of the strategies that will be discussed.


Figure 1. Frequency figures for the use of TOS translation strategies.

Briefly, the most frequently used TOS translation strategies for the inserts and displays examined were subtitling and verbalisation (26.2%, 11 cases each), followed by substitution (19%, 8 cases). Thus, the most common ones in this case are those that provide the translation of the text element through the visual channel. The least frequently employed strategies were omission and voice-off (14.3%, 6 cases).

5.1. Subtitling

Given that “subtitles can be considered descendants of intertitles,” which are the earliest form of TOS (Giménez López 2013: 23, author’s own translation), and that subtitling is “cheap and fast” (Díaz Cintas 2013: 274) in comparison to other forms of AVT, it is unsurprising that this is the most common TOS translation solution. For example, a study by Fox (2016) showed that subtitling is the most frequently employed strategy used to render text elements of any type in dubbed content, with 52% of all instances subtitled. Similarly, as shown above, although less frequent (26.2%), this was also the most common strategy for the instances analysed in this paper.

Nonetheless, the target audience’s viewing experience would seem somewhat negatively affected by subtitles. Firstly, the audience is required to focus on the subtitles, thus often overlooking other visual information (Tveit 2009: 90). Moreover, as argued by Fox, since subtitles are also a type of text element, whenever employed to render TOS, they will simultaneously appear twice, i.e., in the source language (SL) and TL. This duplication is regarded as a “negative feedback effect” (2016: 261). Notably, TOS duplication implies that text appears twice in the visual channel. Thus, when the translation of a text element is provided through the auditory channel, it is not considered TOS duplication.

However, subtitling appears to have significant advantages for translators, producers and distributors. In dubbed material, subtitles are always a viable solution since there is no overlap with the translation of the dialogue. Additionally, no challenging or limiting tasks intrinsic to other strategies, such as recreation or lip-syncing, are required. Although subtitling is subject to spatiotemporal constraints (Díaz Cintas and Remael 2014), subtitling TOS is generally not severely affected by these since, based on the observation of this set of examples, text elements are frequently short.

The following is an example of subtitled TOS:

Example 1

Instance

28

Strategy

Subtitling

Type of TOS

Display

Show

Modern Family, S8 E03

TCR

00:00:40 – 00:00:52

Context

Alex Dunphy is sitting on the couch in her living room, looking serious and showing the audience a piece of paper with some text written on it. While it is shown, Alex is on-screen but does not speak. Instead, she uses the piece of paper to communicate.
The subtitles used to translate this display are standard, white, written in capital letters and placed at the bottom centre of the screen.

Source TOS

Under doctor’s orders not to talk for 2 days. B/C sore throat B/C mono. The one person in this family with anything to say... And I can’t talk.

Target TOS

El médico me ha dicho que no hable durante dos días. Por el dolor de garganta de la mononucleosis. Soy la única de la familia con algo que decir… Y no puedo hablar.
[The doctor told me that I can’t talk for two days. Because of the sore throat from the mononucleosis. I’m the only one in the family with anything to say… And I can’t talk.]

Technically, voice-off could have been the only feasible alternative in this example. As will be explained later, substitution cannot be employed to translate displays, verbalisation requires an utterance from the character in the source product, and omission would have been inappropriate given the crucial nature of this text element to the story.

However, a voice-off could have been unsettling for the target audience due to the content of the text element since the character is explaining that she cannot talk; thus, hearing her (or any) voice could come across as rather incongruent. Additionally, hearing her voice while she is not speaking could seem rather illogical. Hence, despite the downsides of subtitling from the viewers’ perspective, this may be the only possible solution in such cases.

Interestingly, if adjusted to the atmosphere of the scene, subtitles can become part of the film’s identity (McClarty 2012; Fox 2016) and have a positive impact on the audience’s aesthetic experience. Some examples of creative subtitling in TOS translation can be found in the German version of Sherlock (see Instance 76 in Molerov 2015). As if it were a case of substitution, the font and colour of this display have been reproduced in the subtitle. Thus, even if there is still a duplication of TOS, the atmosphere of the scene is not spoiled.

In contrast, the Spanish version used plain, standard subtitles to translate this scene (see Instance 25 in the data set). Aside from the TL, the main difference between both instances is the medium in which they were broadcast. While the creative subtitles were broadcasted by ARD, the Spanish standard subtitles were broadcasted by Netflix. Neither ARD’s nor Netflix’s subtitling guidelines mention creative subtitles as one of their practices. However, while Molerov’s study shows several examples of them that were extracted from ARD, the set of examples analysed in this study (extracted from Netflix) does not include any. Although more research could be conducted, this initial finding would suggest that the decision (not) to adapt subtitles to the aesthetics of a show or scene appears to depend on the distributor of the target product. In line with the discussion above regarding creative subtitles, it should be highlighted that such decisions can have an impact on viewers’ experiences.

5.2. Substitution

Substitution replaces the source text element with its equivalent target text element in the target product (Chaume 2004: 283; Fox 2016: 261). This can be done by editing or by deleting and recreating the source element (Fox 2016) and has several advantages. Firstly, the duplication of TOS caused by subtitles can be averted (Giménez López 2013: 22). Secondly, substitution “prevents additional distraction from the image” and does not cover other on-screen items that may be relevant (Fox 2016: 261). Thirdly, when substituting TOS, the target text element is integrated into the scene and respects aspects of the source text element, such as animations, dramatic pauses and “surprise effects,” thus avoiding certain spoilers caused by subtitles (Masats Casassas 2014). Hence, it would seem that substitution is generally a more localised and viewer-friendly way of providing the target audience with access to the content of text elements.

Nonetheless, it also entails a long, complex and expensive process. According to Masats Casassas (2014), subtitling a 45-minute episode could cost about 200€, whereas a substitution process for inserts in the same episode could cost five times as much. In addition, highly complex inserts must be recreated by a motion graphics studio.

Moreover, it has been observed that substitution has only been employed when dealing with inserts, which appears to confirm Molerov’s (2015) remark that subtitling can be used to translate both displays and inserts, whereas substitution is limited to inserts. More precisely, as stated by Fox, all types of TOS — except displays — “could be substituted graphically” (2016: 261). Thus, when faced with displays, substitution cannot normally be considered a potential solution for technical reasons (compare Instances 13 and 14, or Instances 21, 22 and 23 in the data set).

Another disadvantage of substitution is that it could cause a feeling of strangeness in the target audience for two reasons. Firstly, since it is not always technically feasible, the translator cannot be consistent when employing it. Secondly, substitution denies the target audience access to the original text element (Molerov 2015), which could be problematic because audiences are accustomed to foreign TOS in foreign audiovisual products; thus, seeing texts in Spanish in a British show, for example, might be confusing. Additionally, the presence of SL text in a translated product is not necessarily negative. According to Venuti (2008: 20), maintaining aspects of the SL in the target text contributes to societies becoming less prone to fundamentalist ideologies, less ethnocentric and more accepting of other cultures.

Despite its downsides, the example below demonstrates how aspects such as the content, timing, position, graphics, animation or sound effects of a text element (Molerov 2015) may occasionally influence the choice of this strategy.

Example 2

Instance

20

Strategy

Substitution

Type of TOS

Insert

Show

Sherlock, S2 E01

TCR

00:20:42 – 00:20:52

Context

A middle-aged, elegant man is on-screen. He is wearing a black suit, a grey shirt, and a yellow and black striped tie.
The text element presents the information that this man’s clothing gives about him. Both in the SL and TL version, it is superimposed and floating on-screen, each phrase pops up at a different time and quickly fades, and the order in which the phrases pop up does not align with their position on-screen. The text is written in white and each word is capitalised, in both versions as well. While the text element shows, nobody speaks.

Source TOS

Dog Lover. Horse Rider. Public School. Early Riser. Left Side of Bed. Non-Smoker. Father. Half Welsh. Keen Reader. Tea Drinker.

Target TOS

Le gustan los perros. Monta a Caballo. Colegio público. Madrugador. Lado Izquierdo de la Cama. No Fumador. Padre. Medio Galés. Ávido Lector. Toma té.
[Likes dogs. Rides horses. Public school. Early riser. Left side of the bed. Non-smoker. Father. Half Welsh. Avid reader. Drinks tea.]

Here, the graphics of the original product are largely respected since only the font size is different and the animation has been recreated. Although some typographic details were overlooked in most examples (see Instances 13, 15, 16, 17 and 24 in the data set), the animations appear to have been respected in all of them. An interesting aspect of this and Instance 16 (in the data set) is that the animation presents the items conforming to the text element in an order that does not align with their on-screen position. For instance, Tea Drinker is the last phrase to appear, yet it is located at the top left, which would instinctively make it the first to be read.

Consequently, deciding the order in which the components of the text element will appear in subtitles can be challenging. Thus, substitution seems more appropriate to avoid confusion among both translators and viewers. Therefore, this option could be advantageous in cases where the position and animations of TOS are irregular. Despite this, the Catalan version of Sherlock rendered these and other similar scenes through subtitling to comply with the broadcast network’s style (Masats Casassas 2014). Thus, translators seem constrained by the fact that distributors do not allow the use of certain strategies — even when these might be more advantageous in certain contexts.

In this instance, a potential alternative to substitution and subtitling could be voice-off because the scene is silent, thus allowing the addition of a voice track with the translation as the inserts pop up on-screen. However, the timing seems to be an obstacle. The speed at which the phrases fade in and out does not enable the inclusion of a voice track since voices-off cannot be too fast, and pausing is required to allow the audience to process the information (Knott 2020).

5.3. Omission

Since omission suppresses the message (Martínez Tejerina 2012: 172), the target audience is exposed to an untranslated TOS and generally deprived of information. Therefore, in principle, this strategy may not seem appropriate. In fact, omissions in TOS translation are highly likely to be noticed by the audience, which causes uneasiness because they receive an incomprehensible message (Martínez Tejerina 2012: 172). An example of this is provided as follows:

Example 3

Instance

33

Strategy

Omission

Type of TOS

Display

Show

Modern Family, S8 E17

TCR

00:07:37 – 00:07:40

Context

Gloria and Manny Delgado, mother and son, are in a shopping centre and stop to look at a publicity board.
The text is on that publicity board and is the title of a film. It is at the top left of the board, written in white and in capital letters.
The background is red, and there is a woman in a wedding dress and a child in a tuxedo holding a red rose and looking shocked. Gloria and Manny both look shocked as well. While the text element is shown, Gloria and Manny are on-screen but do not speak.

Source TOS

Oops, I married my mom

Target TOS

(Not translated)

Here, omission comes across as a rather strange choice. Firstly, the circumstances in which the display is presented (especially timing) arguably allow the inclusion of subtitles. Secondly, this text element is highly relevant to the humorous effect of the scene. Since this show is a situational comedy, achieving acceptability as a funny product should be the translator’s primary concern (Zabalbeascoa 1996: 247). This case is surprisingly similar to one found in The Big Store (1941), from which Martínez Tejerina (2012: 172) concluded that translating the text element was imperative to avoid the loss of humour. As explained by Martínez Tejerina (2012: 172), that film also presented a sign that showed a plot-pertinent joke (in that case, a wordplay), which technically could also have been translated because the circumstances (timing and silence) allowed it.

Nevertheless, “[t]he existence of a text element does not necessarily imply the need to act” (Fox 2016: 260). While viewers should not be deprived of a necessary translation, they should also not be presented with an unnecessary one (Fox 2016). Hence, there are cases in which omission appears suitable. For example, when there are iconographic symbols involved (Chaume 2004), the text can easily be recognised by the target audience (e.g., a stop sign, Fox 2016), or it conveys information of little or no importance to the scene or plot (Martínez Tejerina 2012: 166). This idea is illustrated as follows:


Example 4

Instance

5

Strategy

Omission

Type of TOS

Display

Show

HIMYM, S3 E10

TCR

00:12:03 – 00:12:32

Context

Ted Mosby, Barney Stinson and Marshall Eriksen arrive at a Victoria’s Secret party. They are standing at the entrance, looking around the room, which is full of women. The decoration is extravagant. The wallpapers are patterned in black and white and there are screens that show bright colours.
The text is written on a banner, which is at the entrance right behind the characters. While it shows, Ted, Barney and Marshall are on-screen. Barney speaks, but he does not read the banner out loud or refer to it. Before this scene shows, Barney has talked to his friends about this event.

Source TOS

VICTORIA’S SECRET

Target TOS

(Not translated)

On this occasion, the target audience — in this case, Spain — is almost unequivocally familiar with the content of this text element. Moreover, the context in which it is presented makes its meaning easily inferable (Fox 2016). Therefore, although appropriate from a technical perspective, subtitling this display could have had an unnecessarily negative impact on the audience’s experience as it would have led to a somewhat redundant target version. In fact, “[g]raphically speaking, not translating is obviously an ideal case” (Fox 2016: 260) since nothing is interfering with the image and typographic identity of the product.

Another reason to omit the translation of a text element could be to avoid repetition. For example, the instance below shows a text element that had already been translated in previous scenes. Consequently, rendering it again would have arguably been repetitive and somewhat graphically inappropriate.

Example 5

Instance

22

Strategy

Omission

Type of TOS

Display

Show

Sherlock, S2 E01

TCR

01:21:48 – 01:21:51

Context

The lock screen of a mobile phone is shown, and the text is on that screen. The first two and the last words are written in white capital letters and read ‘I AM LOCKED’. Between them, there are four white squares, in which the password must be introduced. The password, ‘SHER’, is written in black capital letters and each letter is placed in one of the squares.
While the text element shows, there is nothing and nobody else on-screen. Seconds earlier and later, the translation of this display is shown (see Instances 21 and 23).

Source TOS

I AM
SHER
LOCKED

Target TOS

(Not translated)


 

5.4. Voice-off

Voice-off refers to substituting a voice track found in the source products with the target text when the speaker is “temporarily off-camera” (Kozloff 1988: 3) but present in the diegesis of the film (Doane 1980). However, in the context of TOS translation, this term refers to a voice track that does not exist in the source product but has been added to the target product to translate a text element, and which is inexistent to the characters of the show (Doane 1980). This strategy can only be used if the text element appears during a silent scene; otherwise, the voice track with the translation would overlap with the dialogue.

An example of this is provided as follows:

Example 6

Instance

1

Strategy

Voice-off

Type of TOS

Display

Show

HIMYM, S3 E04

TCR

00:13:27 – 00:13:30

Context

Lily Aldrin is showing Robin Scherbatsky a drawing of a woman – who represents Robin – and a child. The text is written on the piece of paper, above the woman and the child. It is handwritten in light blue and each word is capitalised.
While it is shown, Lily is on-screen but does not speak and Robin is only partially visible and does not speak either. In the source version, the text is not read out loud.

Source TOS

My New Mommy

Target TOS

Mi nueva mamá
[My new mum]

As will be shown, given the lack of dialogue, verbalisation — i.e., a character reading the (translated) text element out loud as part of the diegesis of the film — cannot be considered as a possible solution here. Substitution also cannot be considered since this text element is not an insert or omission because this display is significant to the story and is not as likely to be understood by the target audience as Instance 5 (see 5.3), for example. Therefore, subtitling and voice-off are the only possible alternatives. Nonetheless, the latter seems more advisable because it prevents the duplication of TOS arising from subtitling, as well as a distraction from and interference with the image. This is relevant here because later in the episode, an item within the drawing will be referred to. Hence, covering it or distracting the target audience with subtitles should ideally be avoided.

5.5. When no strategy is needed: Verbalisation

When attempting to categorise some instances of the data, the available options did not seem accurate. This is the case of instances in which the text element is read out loud by a character in the source version. As Díaz Cintas and Remael (2014: 60) noted, sometimes, “when [text elements such as] a letter or newspaper [are] filmed in close up, a character will often read the text out loud, causing visual-verbal and acoustic-verbal channels to coincide.”

In such cases, when translating the product for dubbing, the translation of the text element is inserted into the existing audio track — unlike voice-off, which requires adding an audio track to the target product. Based on Molerov’s4 terminology, this practice has been labelled verbalisation. Molerov defined this term as “a cross-mode verbalisation, embedded in the dialogue, of the translation of the insert” (2015: 10, author’s own translation).

Thus, whenever possible, this seems to be the most natural way to translate TOS in dubbed products. However, its application is restricted by the source product since it is not possible unless a character reads the text element out loud or refers to it. Additionally, as will be explained, this solution must necessarily be implemented when a character reads the text element out loud or refers to it in the source product, thus verbalisation seems incompatible with the rest of the strategies.

An example of verbalisation is presented as follows:

Example 7

Instance

29

Strategy

Verbalisation

Type of TOS

Display

Show

Modern Family, S8 E03

TCR

00:01:32 – 00:01:37

Context

Alex Dunphy is showing her family a notebook in which she has written a message for them. The text is thus written in a notebook. It is handwritten in black capital letters.
While it shows, Alex is on-screen but does not speak. Phil Dunphy, who is not on-screen but present in the scene, reads the text out loud in both the original and translated version.

Source TOS

Is that why you’re dressed like a hookah?

Target TOS

¿Por eso vas vestida de putilla?
[Is that why you’re dressed like a little slut?]

Since one of the characters reads the text element aloud, the translation can be provided through the auditory channel, which seems the most natural and uncomplicated solution for a dubbed product. Although technically possible in this case, subtitling has arguably been discarded because it would cause the target product to come across as redundant, which can be observed in Instance 40 below:

Example 8

Instance

40

Strategy

Subtitling and verbalisation

Type of OST

Display

Show

Black Mirror, S5 E01

TCR

00:24:10 – 00:24:11

Context

The image shows a neon pink, futuristic pinball machine. The text is inside of it, written in red neon lights and embossed. While it shows, there is nobody on-screen. The text is read out loud by the machine.
The subtitles used to translate this display are standard, white, written in capital letters and placed at the bottom centre of the screen.

Source OST

Game over

Target OST

FIN DE LA PARTIDA
[END OF THE GAME]

Hence, in these cases, since the translation had to be provided as part of the diegesis of the film, employing any of the previously discussed translation strategies would lead to repetition of the translated TOS, while not verbalising it would imply modifying or omitting part of the source text, thereby causing inaccuracies in the dialogue. Therefore, since the target audience would be negatively affected, it could be argued that choosing another TOS translation strategy when verbalisation is available might be rather inappropriate, as can be seen in Instance 40.

However, it should be noted that such decisions are normally not up to the translator. For instance, regarding Sherlock’s Catalan translation, Masats Casassas (2014) explained that subtitling TOS was the distributor’s decision and that in that case, the distributor did not even inform the translator of that choice before they started translating. As a result, certain text element translation solutions in the target product come across as unexpected or even questionable (e.g., unnecessary subtitles); however, the translator had little to no decision-making power. This could explain why some text elements are simultaneously verbalised and subtitled, such as in Instance 40.

Another example of verbalisation is presented as follows:

Example 9

Instance

10

Strategy

Verbalisation

Type of TOS

Display

Show

HIMYM, S4 E07

TCR

00:10:09 – 00:10:13

Context

Barney Stinson is showing his colleagues the merchandising for a holiday that he has created. He is at the centre of the screen and two colleagues are looking at him from behind. They are in a conference room at work and Barney has a carton box in front of him, out of which he is pulling a grey T-shirt.
The text is printed in red on the T-shirt. While it shows, Barney is on-screen and reads it out loud.

Source TOS

Who’s NOT Your Daddy?

SL dialogue

There’s ‘Who’s NOT your daddy?’ T-shirts.

TL dialogue

Camisetas de ‘¿Quién NO es tu padre?’
[‘Who’s NOT your father?’ T-shirts.]

The only distinction between this and Instances 29 and 40 is that, in this case, the character is on-screen whilst reading out loud the text element, whereas the characters were off-screen in the previous examples. The analysis of this example in terms of implications and possible alternatives is thus very similar to that of Instance 29, with the sole difference being that in cases such as Instance 10, lip-syncing may have to be considered when translating the text element.

Another situation in which verbalisation can be used is presented as follows:

Example 10

Instance

30

Strategy

Verbalisation

Type of TOS

Display

Show

Modern Family, S8 E06

TCR

00:02:15 – 00:02:19

Context

Phil and Luke Dunphy, father and son, are at the country club. They are standing next to a board which advertises an event. The text is written on that board. The background is burgundy and the text is written in elegant fonts, the first word in yellow and italics, and the rest in white and Roman type. All words are capitalised.
While the text is shown, Phil and Luke are on-screen, and both speak. Phil asks about the event and Luke describes it.

Source TOS

Tonight: Prospective Members Cocktail Party

SL dialogue

Phil: Oh, how’s that work?
Luke: You find a member to sponsor you, come have some drinks with the committee, and they vote a few people in.

TL dialogue

Phil: Ah, ¿y cómo funciona?
Luke: Te buscas un socio que te presente, te tomas una copa con el comité y votan para ver quién puede entrar.
[Phil: Oh, and how does that work?
Luke: You find a member to introduce you, you have a drink with the committee, and they vote to see who can get in.]

This differs a little more from the previous examples because the actors do not read the text out loud; instead, they comment on it. Consequently, although the text element itself is not translated, the target audience can infer its meaning — and thus understand the scene — thanks to the character’s explanation.

In this respect, it should be noted that since the intention is not to leave viewers in the dark about the meaning of the text element, this cannot be regarded as a case of omission. Cases like this are not even considered TOS translation cases stricto sensu. Despite this, presenting and discussing examples such as this seems relevant because they are a reminder that audiovisual texts have multiple ways of transmitting meaning and that not all need translating.

In this case, even if the text element is not actually translated, the target audience has access to it through other means (the TL utterance), hence making it unnecessary or even inconvenient to add a more ‘literal’ translation. This is because the repetition may be detrimental to the viewing experience, as previously explained. Since TOS translation decisions do not generally depend on translators, it seems crucial to raise awareness about cases like this among those with decision-making power over these matters (e.g., commissioners or distributors).

6. Discussion: Implications for translators and audiences

Ten criteria were considered to evaluate the feasibility and advantageousness of the examined strategies. Four of them concern the translation team, thus facilitating or hindering the translation process (RQ2), while the other six affect the target audience by conditioning their reception of the product (RQ3).

These criteria were established as a result of the previous analysis and the literature on TOS translation.


Figure 2. Criteria to assess TOS translation strategies
from the translation team’s perspective.


Figure 3. Criteria to assess TOS translation strategies
from the target audience’s perspective.

Firstly, it has been perceived that omitting text elements with a narrative function is normally not arbitrary; however, there is a reasonable justification for this. For instance, a very similar or identical text element may have been translated previously (see Instance 22 in Section 5.3), or there might be a perfectly clear context surrounding the text element that contributes to inferring its meaning (see Instance 5 in Section 5.3 and Instance 9 in the data set). Despite being presented with an untranslated text, the target audience’s viewing experience in these cases is conceivably positive because the target product remains unaltered whilst offering the information necessary to follow the story. Therefore, it seems that whenever possible, this may be one of the most convenient strategies for dealing with TOS. Despite this, and even if it greatly facilitates the translation process, it must be used carefully. Otherwise, the target audience might be deprived of important details of the story (see Instance 27 in the data set and Instance 33 in Section 5.3).

Secondly, when dealing with dubbed products, visual TOS translation strategies (i.e., subtitling and substitution) never come into conflict with the acoustic-verbal elements. Nonetheless, they are not always viable—especially substitution, which is seriously limited by the type of text element and the financial and technical means available. However, substitution offers smoother and more visually pleasing translations than subtitling since subtitles “influence the audience’s overall aesthetic experience” (Fox 2016: 245), an influence that tends to be perceived as negative. Despite this, it has been observed that subtitling has been adopted in numerous cases in which more suitable options (e.g., verbalisation or voice-off) were possible (see Instances 2, 6, 12, 14, 31, 39 and 40 in the data set).

In contrast, substitution is likely to provide a positive aesthetic experience because (1) none of the disadvantages of subtitling materialise in substituted text elements and (2) TL text elements tend to reproduce the visuals of the SL ones. Although this recreation process is significantly more frequent when resorting to substitution, it is not necessarily restricted to it (creative subtitling is an example). Therefore, TOS recreation, be it for subtitling or substitution, involves tasks for which audiovisual translators may need to develop new skills in case this practice is fostered in the future. In Fox’s words:

[e]xamples such as Sherlock show that new translation expertise is required to deal with the various aspects of typographic film identity. Only if existing layout strategies and designs are understood, and text elements and their value recognized, can they be recreated in a way that does not disturb the film’s atmosphere and tone (2016: 251).

Additionally, “[t]he choice between replacing a text element and subtitling seems similar to the choice between dubbing and subtitling” (Fox 2016: 261) since dubbing seeks to “emphasize semiotic authenticity, boosting the domestic language and smooth content mediation (in other words: viewer-friendly and localized versions of foreign productions)” (Gottlieb 2005: 25), which substitution achieves. Hence, for dubbing audiences, the substitution would seem more appropriate than subtitling. In fact, Molerov (2015) argued that encouraging this strategy could be beneficial for target audiences, especially in dubbing countries precisely because it resembles this AVT modality. Nonetheless, substitution presents major limitations and difficulties; thus, contrary to subtitling, it is not always feasible.

Thirdly, auditory solutions (i.e., voice-off and verbalisation) are greatly affected by the conditions of the source product and the context in which the source text element finds itself since a space in the voice track to insert the translation is required. Thus, from a technical perspective, these appear to be reserved for dubbed audiovisual material5. Moreover, since these also require a character’s involvement, they seem restricted to displays since other text elements are normally “not visible to possible ‘readers’” (Fox 2016: 263).

Despite this, when translating for dubbing, auditory translation strategies seem advantageous, especially from the audience’s perspective, but also for translators. Therefore, it seems advisable to take advantage of them whenever possible. Nonetheless, there seem to be cases in which even though the source product allowed their use, the translator used another strategy instead (see Instances 2, 6, 12, 13, 14, 15, 16, 31 and 40 in the data set). This could be due to the broadcast network’s preferences or the more limited body of research on this when compared to more widely used and investigated strategies (e.g., subtitling).

7. Conclusions

This paper has examined the range of options used when translating TOS in dubbed audiovisual products translated from English into Spanish. As a result, in response to RQ1, although previous research on this matter presents four possible TOS translation strategies, this study has demonstrated the existence of another, namely verbalisation. Therefore, the analysis of the examples compiled for this paper suggests five TOS translation solutions: subtitling, substitution, omission, voice-off, and verbalisation. Notably, the latter is deemed incompatible with the rest of TOS translation strategies. This is because the use of both — verbalisation and another strategy — generally makes the target text redundant and thus conceivably affects the viewing experience negatively.

Regarding the reasons for choosing each strategy (RQ2), this study allows us to conclude that the selection of TOS translation strategies is mainly influenced by three factors: financial aspects; each broadcast network’s preferences; technical considerations. The latter encompasses the technical means available to implement a strategy, as well as the conditions surrounding the text element in the source product (a character’s intervention, the comprehensibility of the source text element, the type of text element…). Based on these factors, subtitling may normally be the easiest or most realisable strategy apart from omission, whereas substitution can be the most challenging. Moreover, the choice of a particular strategy in each case may also be conditioned by difficulties intrinsic to each strategy or specific to each text element. Based on data observation, these can include TOS duplication, overlap with the source text element, interference with the image, timing, or impossibility to insert the translation in the voice track, to name a few. Since these were detected by analysing this particular set of examples, others may have been overlooked on this occasion but could perhaps be identified by examining different materials.

Concerning the impact of these five strategies on the reception of the target product (RQ3), the main conclusion drawn is that voice-off and verbalisation are likely to provide the smoothest viewing experience, whereas subtitling would perhaps be the least viewer-friendly strategy. In any case, reception studies are required to confirm whether the target audience’s views and needs align with these findings. Additionally, based on an examination of this set of examples and previous research, it can be claimed that while rendering text elements with a narrative function is crucial for the target audience to fully grasp the story, visual and aesthetic aspects also serve an important role in their viewing experience and should thus be considered whenever possible.

While the four translation strategies presented and discussed by scholars to date appear sufficient and appropriate, this article has shown that there is one strategy — verbalisation — that makes the rest unnecessary when available, which seems crucial to highlight. In fact, since a couple of cases of overlap (verbalisation and subtitling) have been identified by observing this set of examples, it can be argued that this seems to have been overlooked in research despite occurring in practice and being inappropriate. This corroborates the claims made in this paper regarding the need for further research on TOS translation.

Overall, there still seems to be room for improvement in terms of the practice of TOS translation. In this respect, a more exhaustive study of TOS translation strategies can be beneficial because this allows us to discuss and reveal the potentialities, limitations and incompatibilities of each solution. This is relevant because by being aware of these, the most advantageous approach among those available is more likely to be selected. As a result, this and other studies on this matter can have a powerful impact on the practice of TOS translation and positively influence the quality of translated audiovisual products, from which audiences can greatly benefit.

Therefore, the starting point of this improvement is research. However, as previously noted, several factors influencing the choice of translation strategy are not determined by translators. In fact, their performance can be seriously limited by producers and broadcast networks. Thus, awareness also needs to be raised among producers and broadcast networks since some of their decisions with respect to TOS translation can ultimately affect the reception and enjoyment of their products in foreign markets.

I would like to finish by suggesting some ideas for future research based on the limitations of this study. Firstly, all the examples analysed were extracted from the same distributor. Conducting similar studies with TOS instances from other sources seems desirable to obtain further information about the TOS translation strategies implemented by other distributors. Likewise, similar studies could be conducted on other shows, seasons or language combinations, and even on types of OST other than displays and inserts. Furthermore, due to the scope of the paper, a conversation with translators or distributors was not possible; however, these conversations are essential to learn about the practice of TOS translation from the perspective of professionals who are involved in it.


 

References
  • Chaume, Frederic (2004). Cine y traducción. Madrid: Cátedra.
  • Delabastita, Dirk (1989). “Translation and mass-communication: Film and T.V. translation as evidence of cultural dynamics.” Babel 35(4), 193–218.
  • Díaz Cintas, Jorge (2013). “Subtitling. Theory, practice and research.” Carmen Millán and Francesca Bartrina (eds) (2013). The Routledge Handbook of Translation Studies. London/New York: Routledge.
  • Díaz Cintas, Jorge and Aline Remael (2014). Audiovisual Translation: Subtitling. London/New York: Routledge.
  • DiGiovanni, Elena and Pablo Romero-Fresco (2019). “Are we all together across languages? An eye tracking study of original and dubbed films.” Irene Ranzato and Serenella Zanotti (eds) (2019). Reassessing Dubbing: Historical approaches and current trends. Amsterdam: John Benjamins, 125–144.
  • Doane, Mary Ann (1980). “The Voice in the Cinema: The Articulation of Body and Space.” Yale French Studies 60, 33–50.
  • Fox, Wendy (2016). “’Should she really be covered by her own subtitle?’ Text elements in film and their graphical translation.” Translation Spaces 5(2), 244–270.
  • Gambier, Yves (2008). “Recent developments and challenges in audiovisual translation research.” Delia Chiaro, Heiss, Christiane and Chiara Bucaria (eds) (2008). Between Text and Image. Amsterdam: John Benjamins, 11–35.
  • Giménez López, María (2013). “Texto en pantalla, un reto para la Traducción Audiovisual”. María Teresa Veiga Díaz and Marta García González (eds) (2013). Traducción multimedia: diversas pantallas, enfoques diversos. Vigo: Universidade de Vigo, 9–38.
  • Gottlieb, Henrik (2005). “Multidimensional Translation: Semantics turned Semiotics.” Heidrun Gerzymisch-Arbogast and Sandra Nauert (eds) (2005). Proceedings of the Marie Curie Euroconferences: MuTra: Challenges of Multidimensional Translation, Saarbrücken, 2-6 May 2005, 1–29. Amsterdam: John Benjamins.
  • Ivarsson, Jan and Mary Carroll (1998). Subtitling. Simrishamn: TransEdit.
  • Knott, Ryan (2020). How To Do Voice Overs Like a Pro: The Complete Guide. https://www.techsmith.com/blog/voice-over/ (consulted 20.06.2022).
  • Kozloff, Sarah (1988). Invisible storytellers: Voice-over narration in American fiction film. Berkeley: University of California Press.
  • Martínez Tejerina, Anjana (2012). “La interacción de los códigos en doblaje: juegos de palabras y restricciones visuales”. MonTI 4, 155–180.
  • Martínez Tejerina, Anjana (2016). El doblaje de los juegos de palabras. Barcelona: UOC.
  • Masats Casassas, Maria (2014). La traducción de textos superpuestos en Sherlock. Undergraduate thesis. Universitat Pompeu Fabra Barcelona.
  • Mayoral, Roberto, Kelly, Dorothy and Natividad Gallardo (1988). “Concept of Constrained Translation. Non-Linguistic Perspectives of Translation.” Meta 33(3), 356–367.
  • McClarty, Rebecca (2012). “Towards a Multidisciplinary Approach in Creative Subtitling.” MonTI 4, 133–155.
  • Molerov, Dimitar (2015). Schriftlicher Text im Film und seine Übersetzung: Die Inserts der BBC-Serie Sherlock (2010–12). Master’s thesis. Johannes Gutenberg Universität Mainz.
  • Molina, Lucía and Amparo Hurtado Albir (2002). “Translation Techniques Revisited: A Dynamic and Functionalist Approach.” Meta 47(4), 498–512.
  • O’Sullivan, Carol (2008). “Multilingualism at the multiplex: a new audience for screen translation?” Linguistica Antverpiensia, New Series – Themes in Translation Studies 6, 81–97.
  • Saldanha, Gabriela and Sharon O’Brien (2014). Research Methodologies in Translation Studies. London/New York: Routledge.
  • Tveit, Jan-Emil (2009). “Dubbing versus Subtitling: Old Battleground Revisited.” Jorge Díaz Cintas and Gunilla Anderman (eds) (2009). Audiovisual Translation: Language Transfer on Screen. New York: Palgrave Macmillan, 85–96.
  • Venuti, Lawrence (2008). The Translator’s Invisibility: A History of Translation. London/New York: Routledge.
  • Zabalbeascoa, Patrick (1996). “Translating Jokes for Dubbed Television Situation Comedies.” The Translator 2(2), 235–257.
  • Zabalbeascoa, Patrick (2008). “The nature of the audiovisual text and its parameters”. Jorge Díaz Cintas (ed.) (2008). The Didactics of Audiovisual Translation. Amsterdam: John Benjamins, 21–37.
Websites
Filmography
  • Bays, Carter and Craig Thomas (Creators). (2005–2013). How I Met Your Mother [TV Series]. 20th Century Fox Television.
  • Brooker, Charlie (Creator). (2011–). Black Mirror [TV Series]. Endemol Shine UK.
  • Gatiss, Mark and Steven Moffat (Creators). (2010–2017). Sherlock [TV Series]. BBC One.
  • Lloyd, Christopher and Steven Levitan (Creators). (2009–2020). Modern Family [TV Series]. 20th Centrury Fox Television.
  • Reisner, Charles (Director). (1941). The Big Store [Film]. Metro Goldwyn Mayer.
Data availability statement

The complete set of examples is available on the OSF open repository and can be accessed via the following link: https://osf.io/mfw7c/?view_only=3282295826a24cf9b35b118fb88a10db

Biography

Noemí Barrera-Rioja holds a BA in Translation and Interlinguistic Mediation from Universitat de València (2019) and an MA in Translating for Business and International Institutions from Aston University (Birmingham, UK, 2020). Funded by the Spanish Ministry of Education, she is currently writing her doctoral thesis whilst working as a teaching fellow at Universitat de València. Her research mainly revolves around audiovisual translation, the translation of humour, and media accessibility.

ORCID: 0000-0002-7237-9748

E-mail: noemi.barrera@uv.es

Notes

Note 1:
Translation decisions “that affect micro-units of the text” are labelled “techniques” (Molina and Hurtado Albir 2002: 507–508). Despite this, to remain consistent with the terminology employed in academic works focussing entirely on TOS translation (e.g. Fox 2016; Giménez-López 2013), the term “strategy” is used instead, as it is used with that meaning in such works.
Return to this point in the text

Note 2:
“Insert” can also be synonymous with “text on screen” and “text element” (i.e., an umbrella term used to refer to any written text appearing on an audiovisual product (e.g., Chaume 2004, Díaz-Cintas and Remael 2014). Nonetheless, in this article, insert refers to a type of text element since one of the objects of study are inserts as defined by Fox (2016) and Molerov (2015) (i.e., as a type of TE instead of an umbrella term).
Return to this point in the text

Note 3:
“Captions” are “texts that tell the audience when and/or where a scene is taking place or, in programmes of a more documentary nature, the name of a speaker and perhaps [their] position and title” (Ivarsson and Carroll 1998: 97). Fox considered “narrative texts” as a subcategory of caption; they differ in that they are longer, more relevant to the plot and “tend to appear as prologues, epilogues, and like title cards between major chapters of a film” (2016: 256). Last, by “subtitles,” Fox referred only to those that are “created during the post-production of a film, usually for one or more additional language(s) in the film” (2016: 257); that is, part-subtitles (O’Sullivan 2008).
Return to this point in the text

Note 4:
Molerov (2015) briefly mentioned the possibility of using the dialogue to provide the translation of a TE. Nevertheless, this has not been analysed in greater depth despite its advantageousness, especially for dubbing audiences.
Return to this point in the text

Note 5:
In subtitled products, rendering TOS through subtitling is not always possible due to overlap, whilst auditory solutions are unavailable. Therefore, there seems to be a greater range of TOS translation options available when dealing with dubbed products. Hence, interestingly, the idea that translating for dubbing poses more restrictions than for subtitling (e.g., Mayoral et al. 1988) may not apply to the translation of TOS. In any case, more research is required to assert whether this is true with reasonable certainty.
Return to this point in the text