The Routledge Handbook of Translation and Pragmatics

Carlos de Pablos-Ortega

Introduction

The research presented in this chapter systematises the linguistic representation of the speech act of directives in audiovisual translation (AVT). The main objectives of the study are threefold. First, to investigate the linguistic representation of a specific speech act type, namely directives, and second, to analyse contrastively how directives are performed in British English and Peninsular Spanish, exploring their level of directness; and finally, to determine the extent to which the translation of these speech acts is closer to the source text (ST) audience or the target text (TT) audience in relation to the degree of (in)directness. Using the scripts from 24 films and their corresponding translation into subtitles, the study also takes into account the degree of fidelity¹ towards the ST at speech level.

This chapter provides an example of interdisciplinary research, combining AVT and speech act theory, using applied linguistics tools as a theoretical framework for the research methodology. The combination of speech act theory with film discourse and its translation creates an approach which, to date, has not been used widely (Pedersen, 2008). The applied linguistics framework helps inform the patterns used in the original film script and its translation, bearing in mind elements such as linguistic forms, politeness formulae or components to express (in)directness when performing speech acts. The findings of this research shed light on how specific linguistic elements i.e., speech acts, are represented in visual texts contrastively via subtitling. The chapter is divided into four sections. The first introduces previous research on directive speech acts (DSAs) with particular reference to studies on subtitling and the theoretical framework used in the analysis of directives, followed by the second section in which methodological aspects of the study are described. The final two sections present and discuss the findings of the study, concluding with a summary of key points and reflection on the significance of this type of research for scholars in translation studies and beyond.

1 Directives and subtitling

Following Searle’s classification of speech acts, requests are categorised as directives, the aim of which is to persuade the speaker to do something (Searle, 1979: 11). According to Brown and Levinson’s politeness model (1987), these speech acts are labelled as face-threatening acts, which are performed and take account of three main social variables: social power, social distance and level of imposition (see also Mapson, this volume). Contextual information is a key element which governs different performances of the speech act and, consequently, the level of (in)directness is a distinctive feature which is factored into the analysis.

From the linguistic perspective, a significant body of speech act research focuses on the examination of questions used for requesting in different languages, factoring in the relationships between the interlocutors (asymmetrical: social and power distance) or the object as a variable for the request (Lindström, 2005; Heinemann, 2006; Curl & Drew, 2008; Placencia, 2008; Craven & Potter, 2010; Rossi, 2012; González-Cruz, 2014; Takada & Endo, 2015; Zinken, 2015). Using data extracted from role plays, Discourse Completion Tests (DCTs) and participant observation notes, contrastive studies between different varieties of Spanish and English have confirmed the use of more indirect strategies in Spanish than in English when requesting (Placencia, 1996; Ballesteros Martín, 2001; Márquez-Reiter, 2002; Pinto & Raschio, 2007). The outcomes of these investigations have informed the interaction patterns in both Spanish and English and the aim of the present research is to explore whether similar patterns for requesting are present in film discourse and AVT, in the form of subtitles.

In the last ten years, research on speech acts has complemented other linguistically oriented research in AVT. Pedersen (2008), for example, has explored the extent to which speech act theory can be used for quality assessment in subtitling and other researchers have focussed on the representation of speech acts in both subtitled and dubbed scripts: expressing compliments (Bruti, 2009), giving advice (Pinto, 2010) and the formulation of greetings, leave-taking and good wishes (Bonsignori etal., 2011, 2012; Bonsignori & Bruti, 2015). Other work is connected with implicatures (Greenall, 2011; Desilla, 2012) and politeness (Gartzonika & Şerban, 2009). However, to date, not a great deal of research has been conducted on the theme of requesting.

One distinctive feature in the realisation of DSAs is their degree of (in)directness. In this respect, it is important to make a distinction between conventional and conversational indirectness. As Blum-Kulka (1987: 141) observes “conventional indirect requests realize the act by systematic reference to some precondition needed for its realization, and share across languages the property of potential pragmatic ambiguity between requestive meaning and literal meaning.” By contrast, conversational indirectness is the ability to produce and interpret indirect meaning in messages. This type of indirectness has two dimensions, according to the variability in which indirectness is manifested. The first type refers to the extent to which individuals look for indirect meaning in the remarks of others whilst the second dimension refers to an individual’s tendency to either speak indirectly (i.e., convey nonliteral meanings) or directly (Holtgraves, 1997).

Speech act theory was formulated by Austin ([1962] 1976) who observed utterances and divided them into three different categories: locutionary acts (the production of an expression with sense and reference), illocutionary acts (the performance of the act) and the perlocutionary act (the effect produced by the performance of the act). Searle’s development (1969) of speech act theory, in particular the notion of the illocutionary acts, led him to create a taxonomy of five speech act types: declarations, representatives, expressives, commissives and directives (see also Bruti, this volume). The type of speech act chosen for the analysis in this research is Directives, which are designed to induce the receiver take a particular action and include actions such as commanding, requesting and suggesting. No distinctions have been made between these for the purpose of this study.

The formulation of any speech act comprises two indicators: propositional content and illocutionary force (Searle & Vanderveken, 1985). The latter can be represented by any element and indicates a particular or a range of illocutionary forces. These are devices such verb tenses, punctuation, word-order in a sentence, intonation, etc., which can also serve as mitigators or hedging devices.

1.1 Components of directive speech acts

When performing a DSA, there are two elements that need to be taken into consideration: the relationship between the interlocutors,² the one who performs the DSA and the receiver, and the action for which the DSA is being carried out. In order to introduce the components present in a DSA, an example (1) taken from the English film Four Weddings and a Funeral is provided. In this example, the interlocutors are two guests at a wedding who have met recently. Charles requests information: the place where Carrie is staying the night of the wedding celebration.

(1) Charles asks Carrie:

I was wondering where you are staying tonight?

Two variables are identified for the realisation of the DSA. The first concerns the relationship between the interlocutors, which encompasses two additional elements: their social distance (whether the interlocutors know each other, i.e., family members, friends, acquaintances, etc.) and social power (the perceived power dynamic between the interlocutors). The second variable concerns the reason why the DSA is being performed; for example, an action or an object. In example 1 the DSA refers to a request for information about the location in which one of the interlocutors is staying overnight. Therefore, from a grammatical and syntactical viewpoint, different types of DSA are bound to be formulated differently, given the weight, importance and reason for their performance. Consequently, the construction of the DSA changes according to these variables, leading to the creation of utterances, which might include additional (in)direct elements. For example, when a higher imposition request is formulated (a large amount of money), indirect formulae are more likely to be used (Hartford & Bardovi-Harlig, 1996; Biesenbach-Lucas, 2006, 2007; Economidou-Kogetsidis, 2011; Félix-Brasdefer, 2012; Merrison etal., 2012; Savic, 2018).

From a cross-cultural perspective, the performance of the DSA may vary significantly and the illocutionary force of the act – being more or less (in)direct – might lead to a misinterpretation of an utterance. For instance, an indirect DSA in English can be understood differently by a non-native speaker of English, depending on the context, and the intended illocutionary force might therefore be lost. This is more likely to happen in languages where direct strategies are used more often in the formulation of a request. This is the case in the speech act of thanking, which can be used with the illocutionary force of a request. For example, in Britain a primary school teacher might say “thank you” to a noisy group of students in order to bring them to attention or infer that they must now be quiet. This speech act, allegedly indicating gratitude, is in fact the formulation of request, i.e., the teacher is thanking the students in advance of the action being performed. The illocutionary force in this case might be questioned by a non-native speaker of English, who will probably be puzzled by thanking someone for no apparent reason, and consequently, leads to confusion or misunderstanding.

In connection with the realisation of speech acts and their relation to (in)directness, Blum-Kulka and colleagues (1989) conducted a cross-cultural investigation of speech act realisation patterns using apologies and requests: Cross-cultural Speech Act Realization (CCSARP). The goal of their project was to establish the similarities and differences between native and non-native speaker’s realisation speech patterns in eight languages and varieties of English: Australian English, American English, British English, Canadian French, Danish, German, Hebrew and Russian (see also Mapson, this volume).

The instrument used for data collection in the investigation by Blum-Kulka and colleagues was a discourse completion test, which consisted of incomplete discourse sequences, representing socially differentiated situations. This pioneering investigation, which produced a vast amount of data, has enriched our knowledge about requests, unveiling important findings with regard to requesting (Blum-Kulka etal., 1989: 198). Drawing on the findings, these researchers created a taxonomy for the categorisation of the indirectness strategy types based on three main groups: direct/impositives, conventionally indirect and non-conventionally indirect/hints.

In order to analyse the DSAs extracted from the film scripts and their translation into subtitles, the taxonomy from the CCSARP’s coding manual (Blum-Kulka etal., 1989) was used. The DSAs were explored by focusing on three elements: 1) the head act, as the minimal unit that performs the DSA; 2) the alerters, as opening utterances that precede the head act and 3) the supportive moves, as external units to the DSA. These supportive moves are also known as hedging, mitigators or downtoners. Example 2 illustrates a DSA, which includes three elements: I could have is the head act, which is introduced by an alerter (I’m so sorry to disturb you) and preceded by a mitigator (I wonder).

(2) I’m so sorry to disturb you but I wonder if I could have your autograph?

Another variable included in the analysis is the degree or level of directness in the realisation of the DSA which was carried out, bearing in mind the grammar categories of the DSA that appear in the head act and the supportive moves. With a view to creating a consistent analytical approach, a categorisation taxonomy (Table 11.1) was adapted from the work of other researchers: Searle (1979), Blum-Kulka and colleagues (1989), Márquez-Reiter (2002), DeCapua and Dunham (2007) and Pinto (2010).

Table 11.1 Categorisation taxonomy (adapted) of directive speech acts with examples from English films

The head act group for Strong Direct DSAs includes bare imperatives, which lack visible subjects (“You”), adverbs (“Out now!”) and nouns, which are sometimes used for requesting (“Boarding pass”). The second category is Weak Direct and includes imperatives with personal pronouns (“You”), imperatives followed or preceded by a polite marker (“Go out, please” or “Please go out”), a gerund or present participle in an interrogative form (“Going out?”) and other constructions expressing obligation (“I want you to go out” or “I say go out”). Instances where the first person plural imperative is used (“Let’s go out”) are also part of this category as well as the use of polite formula, such as “Please” or “Thank you”. As mentioned in the previous section, it is interesting to note the illocutionary force of thanking as a DSA, which is uncommon in other languages.

The third category of the categorisation taxonomy of DSA is labelled as Conventionally Indirect and has been subdivided into two further categories: 1 and 2. The first one refers to more direct forms of the speech acts and all modal verbs are part of this category (can, could, may, might and will). The use of the modal verb turns the head act into an indirect DSA, providing hedging for the performance of the action, usually presented in the form of a question: “Can you go out?”, “Could you pass me the salt?”, “May I have your boarding pass?”, etc. However, it might also be present in statements like “You must go out”, where the DSA provides a sense of obligation.

The second group of Conventionally Indirect DSAs includes other modal verbs (would, should) and expressions with verbs whose lexical meaning denotes indirectness of the DSA. Some examples of these constructions are: “Would you like to go out?”, “I wonder if you would mind going out?”, “Do you mind going out?”, “I’m sure you would like to go out”, “You should probably go out”, etc. The final category has been labelled as Non-conventionally Indirect or Hints and contains structures whose illocutionary force is not intended as a DSA. However the contextual and extra-linguistic elements (relationship between the interlocutors, situation, etc.) make the speech act function as a directive. For example, if two interlocutors are talking indoors and one remarks: “It’s really nice outside”, this utterance can be interpreted as a hint or off record statement (Grainger & Mills, 2016: 2), implying that one of the interlocutors would like to go outside. The sentence in this example, which is a statement of fact, has in fact the illocutionary force of a DSA. Therefore, different linguistic structures, which might appear to perform other types of speech acts (compliments, apologies, thanking, etc.) if used in isolation, can behave with different illocutionary force given contextual and extra-linguistic factors. It is important to bear in mind that the variables of social and power distance between the interlocutors when the DSA was performed are beyond the scope of the research objectives of this investigation.

2 Methodology

For the purpose of this investigation, a corpus of 24 films (Table 11.2), divided into 12 British and 12 Spanish were used. The criteria for the selection of these films were two-fold: the film genre (comedy) and the year in which they were released, from 1975–2013 (Table 11.2). The choice of comedy genre was based on the wide range of speech acts identified in these films during a preliminary phase of analysis. These speech acts are directly related to the dynamic nature of the films, which is represented by the actions and interactions between the characters. The selection of comedies was carried out based on their popularity and their accessibility on DVD format with subtitles in both languages. This study does not take into account the longitudinal perspective (i.e., changes in the subtitling practice in the 70s and those currently used), therefore for the purpose of this investigation, the release year is not a key variable.

Table 11.2 List of films used in the study

For the purpose of the analysis, the Spanish and English films used as corpora were selected on their DVD formats, including subtitles in English and Spanish respectively. The DSAs were extracted from all the films and included in an Excel spreadsheet, from both the original film script and their translation into subtitles. These were then classified, following the proposal for the categorisation taxonomy, and subsequently quantified, using percentages which were worked out from the total number of DSAs found in the corpora. Selected findings are presented in the next section.

3 The representation of directives in film scripts and subtitles

The findings of the research are presented following the structure of the analysis: the results with regard to the representation of DSAs in the film scripts in both languages: English and Spanish contrastively. This distribution is in turn split into direct and indirect DSAs by film in each language. The representation of the DSAs also takes into consideration the classification of the speech acts according to the level of directness: Strong Direct, Weak Direct, Conventionally Indirect Type 1, Conventionally Indirect Type 2 and Unconventionally Indirect.

The analysis then explores the shifts³ in translation from the original script to the subtitles. The results in these cases are presented bearing in mind the change in the translation process: either direct/literal translation of the DSAs or, in the cases where there is a change in the translation, shifting from the direct to the indirect form of the DSAs. Finally, the changes in the translation are introduced in both corpora, taking into account the translation from English into Spanish, for the English corpus, and from Spanish into English for the Spanish. As mentioned in the methodology section, the DSAs in all the combinations described above have been quantified by means of percentages in order to systematically present the findings and draw conclusions.

The total number of DSAs found in both corpora is 1966, broken down into 828 DSAs for the English film scripts and 1138 DSAs for the group of Spanish films. The total length of the English films is 22 hours and 21 minutes compared with 19 hours and 9 minutes for the Spanish films.

3.1 Directives in the English and Spanish film scripts

Once the films had been analysed, the DSAs found in the scripts were quantified according to film, and generally speaking, the largest number of DSAs in both groups of film scripts corresponds to the direct type (strong and weak). However, differences are found between the corpora.

Figure 11.1 shows that half of the English films present a higher number of DSAs (direct type) in their original scripts, between 91 per cent and 84 per cent, namely Billy Elliot, The Full Monty, The Life of Brian, The Holy Grail and Bridget Jones. The graph also illustrates that the remaining six films have a slightly lower representation of DSAs (direct type), showing higher percentages of the indirect type of DSAs: Four Weddings and a Funeral (52 per cent), Love Actually (46 per cent), About Time (45 per cent), Best Exotic Marigold Hotel (43 per cent), Calendar Girls (41 per cent) and Notting Hill (37 per cent). The results, with regard to the percentages of directness, reveal that the film genre is a relevant factor for the inclusion of a specific type of DSAs (i.e., direct or indirect). As a consequence, comedies, such as the Monty Python films (The Life of Brian or the Holy Grail), include a higher number of direct type of DSAs, whilst the group of romantic comedies (Four Weddings and a Funeral, Notting Hill or About Time) tend to use more of the indirect types of DSAs. This finding demonstrates that specific features, such as the subtleties and nuances of the language, are used in the creation of film scripts in relation to the genre (in this case, romantic comedies), hence the need for scriptwriters to mitigate against using more direct strategies in the dialogue (Kozloff, 2000; Mernit, 2001; Desilla, 2012).

The exploration of DSAs in the Spanish film scripts (Figure 11.2) indicates regular use of the direct type. All films in this group reveal a higher percentage (77 per cent or above) in the use of the direct type of DSAs. In three of them the percentage of the indirect type is very low: Crossing Borders (6 per cent), Cousinhood (6 per cent) and Common Wealth (8 per cent). These results show that the performance of DSAs in the Spanish language tends to involve more direct than indirect formulae. This contrasts with the English corpus, where it is apparent that the general tendency is to use indirect types of SDAs, more specifically in the comedy films. These findings corroborate the previous contrastive applied linguistics research in the use of more direct strategies when requesting in Spanish than in English (Placencia, 1996; Ballesteros Martín, 2001; Márquez-Reiter, 2002; Pinto & Raschio, 2007).

A more in-depth analysis of the type of DSAs was subsequently carried out of both corpora following the categorisation taxonomy (Figure 11.3). In six out of the 12 of the English films, Strong Direct DSAs (bare infinitives, adverbs or nouns) are used in more than 50 per cent of cases (The Holy Grail, The Life of Brian, The Full Monty and Billy Elliot). Weak Direct DSAs (imperatives including the pronoun you or imperatives followed by polite markers, like “please”) are used in 33 per cent or less of the whole sample. These findings indicate that when using a Direct type of DSAs, the preferred type is Strong versus Weak, adding emphasis to the directness of DSAs.

The film scripts in English with a higher percentage use of Conventionally Indirect type of DSAs, show a preference for type 2 (modal verbs and expressions with verbs whose lexical meaning provides indirectness), between 25 per cent and 35 per cent were present in Four Weddings and a Funeral, Love Actually, The Best Exotic Marigold Hotel and Notting Hill. Conventionally Indirect type 1 (modal verbs such as can, could, etc.) are found to be over 20 per cent in only two films: About Time (22 per cent) and Calendar Girls (29 per cent). Unsurprisingly, Non-conventionally Indirect forms only appear in two films showing very low percentages: Calendar Girls (4 per cent) and A Fish Called Wanda (2 per cent). This may be due to the fact that when performing directives, this type of subtle nuance, often indicated by non-conventionally indirect/hints (section 2), seem to be uncommonly used in the film scripts chosen for this investigation.

Closer analysis of the DSA type in the Spanish corpus (Figure 11.4) shows a preference for Strong Direct forms (between 70 per cent and 91 per cent) in seven of the 12 films: Crossing Borders, Common Wealth, Rated X, Family United, Eight Dates, You Are My Hero and Women on the Verge of Nervous Breakdown. Only three film scripts reveal a high percentage of Weak Direct type of DSAs: The Other Side of the Bed, Cousinhood and As Luck Would Have it. The use of Weak Direct type of DSAs is more frequent in the English corpus, as nine of the 12 film scripts present 20 per cent or higher of this DSA type. The same number of film scripts (nine) use Conventionally Indirect forms, which is lower than in the English corpus and shows 10 per cent or less of this type of structure. In the Spanish film scripts, Non-Conventionally Indirect/Hints are not found, which is a significant factor in relation to the representation of (in)directness in the Spanish language.

3.2 Directives in the English and Spanish subtitles

The analysis of the subtitles indicates variability in the approach to the translation of the DSAs. In this section, the findings are presented in relation to both corpora, paying close attention to the instances where translation shifts are observed. In these instances, a more detailed exploration takes place and the type of DSA that is presented in the corresponding subtitle of the film is highlighted.

Generally speaking, the translation of DSAs in the subtitles shows little change in both corpora (Figure 11.5 a, b). In 89 per cent of the cases, a direct/literal translation of DSAs is the preferred option in the Spanish subtitles of the English films, and in 90 per cent of the cases, the same happens in the English subtitles of the Spanish films.

In instances where a shift in the translation occurs (Figure 11.6 a, b), in both corpora the shift tends to be from the indirect to direct type of DSAs. For the English films, the shift is from indirect to direct type in 75 per cent of the cases, and for the Spanish films in 84%. From a technical (subtitling) perspective, this choice is coherent given that the reduction in the number of characters is a key element in audiovisual translation. Subtitlers need to find strategies for the reduction of characters, through omission or deletion techniques, in order to guarantee the readability of the subtitles and also comply with the appropriate reading speed (Díaz-Cintas & Remael, 2007: 72).

In the English corpus, the analysis also encompassed the use of particular strategies when a change of DSA type takes place in the translation of the film scripts into subtitles (Figure 11.7). The most common strategies are the shifts from Conventionally Indirect type 1 to Weak Direct and from Weak Direct to Strong Direct. Both cases are shown in 18 per cent of the analysed instances. The following examples help to illustrate the shifts or changes from the original script into the subtitles.

Taken from the film Notting Hill, example 3 illustrates the first strategy where the use of the modal verb (could) turns into an imperative form preceded by please (por favor, avísame) in the subtitle.

(3) ST/original dialogue: Could you get a message to him that I’ll be a little late?

Subtitle: Por favor, avise que llegaré tarde [Please, warn them that I’ll be late]

In example 4, from A Fish Called Wanda, the DSA in the original script uses the polite formula please, which is left out in the subtitle in Spanish.

(4) ST/original dialogue: Please do take your time

Subtitle: Tómese su tiempo [Take your time]

As shown in Figure 11.7, the second most common strategy used in the translation of the subtitles is the shift from Conventionally Indirect type 2 to Conventionally Indirect type 1 (13 per cent), to Weak Direct (12per cent) and to Strong Direct (12 per cent).

Example 5, from The Best Exotic Marigold Hotel, illustrates one of these shifts where the indirect formula for requesting in English (Would you mind if . . .?) turns into the use of a modal verb in the Spanish subtitle (Puedo [Can]).

(5) ST/original dialogue: Would you mind if I show something?

Subtitle: ¿ Puedo enseñarte algo? [Can I show you something?]

The shift from Conventionally Indirect type 2 to Weak Direct is presented in example 6, from Calendar Girls, where the head act, the verb “mind” in the present tense in English, turns into the simple present tense of the verb, which is used for the actual request: “¿Me lo prestas?” [Will you lend it to me?]

(6) ST/original dialogue: Ted, do you mind if I borrow this?

Subtitle: Ted, ¿me lo prestas? [Ted, will you lend it to me?]

Example 7, from Notting Hill, represents the second most common strategy in the translation of the film script into subtitles which shows the change from Conventionally Indirect type 2 to Strong Direct. The English formula for requesting “If you would like to come with me” turns into an imperative form in the Spanish subtitle “Venga conmigo” [Come with me].

(7) ST/original dialogue: If you would like to come with me, we can just rush you through the others

Subtitle: Venga conmigo para continuar con los otros. [Come with me to continue with the others]

Figure 11.7 shows that in 9 per cent of the cases, the shift in the translation of the DSAs occurs from Strong Direct to Weak Direct forms, and in 6 per cent of the cases from Conventionally Indirect type 1 to Strong Direct and from Weak and Strong Direct to Conventionally Indirect type 2.

Example 8, from The Holy Grail, illustrates this change where the auxiliary verb “will” is used to express a request turns into an imperative form (Preguntad [Ask]) in the Spanish subtitle.

(8) ST/original dialogue: Will you ask your master if he wants to join my court at Camelot?

Subtitle: Preguntad a vuestro señor si quiere unirse a mi corte de Camelot. [Ask your master if he wants to join my court at Camelot]

Example 9 shows the change from a Weak Direct type, an imperative sentence (Let me just have), to a Conventionally Indirect type 2, using the conditional tense in a question to indicate the request (¿Me dejariáis? [Would you leave me?]).

(9) ST/original dialogue: Let me just have a little bit of peril!

Subtitle: ¿ Me dejariáis solo un poco de peligro? [Would you leave me a little bit of peril?]

The translation strategies used in the Spanish corpus, shown in Figure 11.8, reveal that in 55 per cent of the cases where there is a shift in the translation from the original script, the preferred strategy of the subtitler is Weak Direct to Strong Direct. Example 10 (from As Luck Would Have It), illustrates this change in the translation: the directive form in the original script in Spanish includes an imperative form followed by a polite marker (por favor [please]), which is left out in the subtitle in English, thereby leaving the imperative form on its own.

(10) ST/original dialogue: Por favor, ¡déjennos pasar! [Please, let us through!]

Subtitle: Let us through.

A second group of translation strategies, shifting from Conventionally Indirect type 2 to Conventionally Indirect type 1, and from Conventionally Indirect types 1 and 2 to Strong Direct take place in 12 per cent, 10 per cent and 9 per cent of the instances respectively. Example 11, from The Other Side of the Bed, shows the change in the directive form where the request (¿Te importa? [Do you mind?]) to the use of the modal verb (Could) in the English subtitle.

(11) ST/original dialogue: Oye, Javier ¿ te importa que veamos esto un segundito? [Listen, Javier. Do you mind that we see this for a second?]

Subtitle: Javier, could we watch this here?

Conventionally Indirect types 1 and 2 turned into Strong Direct forms in 19 per cent of the cases. In a few of these examples, question forms used as requests in Spanish, like “¿Me vais a ayudar a separarlos?” [Are you going to help to separate them?] are changed into imperative forms in the subtitles in English (“Help me split them”).

In less than 5 per cent of the cases, other changes in the translation take place but these do not represent a significant finding with regard to the translation strategies used across the Spanish corpus.

As mentioned earlier, the shift in the translation from indirect to direct forms of the DSAs is more common than the shift from direct to indirect. This finding is unsurprising given that from a technical viewpoint, the reduction in the number of words (i.e., fewer characters per line) is needed in order to facilitate the audience the reading of the subtitles. However, in some cases, in the English films, direct strategies are used in the original script which are turned into indirect strategies in the subtitles in Spanish. This is illustrated in example 12, taken from the film Notting Hill, where there is change from a Strong Direct type of directive to a Conventionally Indirect type 2. It seems that in this case, the subtitler has decided to show more indirectness in the realisation of the DSA and, as a consequence, the character in the film is portrayed more politely, given the reformulation of the request in the subtitle.

(12) ST/original dialogue: Give me 5 minutes

Subtitle: ¿ Me das 5 minutos?

4 Discussion

Generally speaking, the analysis carried out in both corpora, English and Spanish films, reveal significant findings for the DSAs in both languages. Although there are many types of directives, this study focuses only on the issue of (in)directness. This section summarises the most relevant results following the analysis of the film scripts individually by language, and contrastively. It also discusses the most important findings about the translation of the film script into subtitles, with the aim of providing an insight into linguistic strategies used for the creation of the subtitles, and more specifically for the translation of the DSAs. A final section discusses the extent to which the three objectives of the study have been met, namely: 1) to investigate the linguistic representation of DSAs from 24 film scripts and their subtitles; 2) to analyse contrastively how DSAs are performed in British English and Peninsular Spanish, exploring their levels of directness; and 3) to determine the extent to which the translation of DSAs is closer to the ST audience than to the TT audience as well as to identify the degree of fidelity towards the ST at speech level.

The representation of direct and indirect types of DSAs indicates a clear preference for the use of the direct type: found in 70.4 per cent and 86.3 per cent of cases in English and Spanish films respectively. However, the percentage of the indirect type of DSAs in the English film scripts is slightly higher (29.58 per cent) when compared to that of the Spanish film scripts (13.60 per cent), thus indicating a preference for the indirect type of DSAs in the English corpus. A possible explanation for this might be that the use of DSAs is connected with the film genre, especially in the English comedies. The English films Holy Grail, Life of Brian, A Fish Called Wanda, Full Monty and Billy Elliot use the direct type of DSAs more frequently than other films, such Notting Hill, About Time, Love Actually, Four Weddings and a Funeral and The Best Exotic Marigold Hotel, which use more indirect types of DSAs in their scripts. The last group of films can be classified as romantic comedy genre, not simply comedies. This result is in line with previous research, which confirms that romantic comedies make more use of indirect language (Kozloff, 2000; Mernit, 2001; Desilla, 2012). In addition to the film genre, it may also be important to highlight an idiosyncratic aspect that might have influenced the use of indirect requests: the fact that Richard Curtis is the scriptwriter of the five romantic comedies used for the analysis.

The findings of this research reveal that in 89 per cent of cases relating to Spanish subtitles and 90 per cent relating to the English subtitles, the direct/literal translation is the preferred strategy. Nonetheless, where the findings indicate a change in the translation from the script into the subtitles, differences are observed between both corpora. The subtitles in English, produced from the Spanish film scripts, show a slightly higher percentage (84 per cent) of shift in the translation from indirect to direct DSAs than the subtitles in Spanish for the English films (75 per cent). This finding confirms a preference for a more direct type of DSAs in the TT (English subtitles) for the Spanish films. This leads us to conclude that the representation of Spanish culture via the audiovisual text leans towards the use of a more direct approach when performing DSAs and poses the question of the extent to which subtitles reinforce cultural stereotypes of Spanish people by English native speakers. As a consequence, this may lead to the creation of a negative perception of Spanish native speakers, as directness is commonly seen as an invasion of personal space by speakers from England (De Pablos-Ortega, 2010 and 2015).

The range of strategies used in translating the film scripts into subtitles indicates the use of a wider variety of resources in the English corpus. A high percentage of the DSAs in the English film scripts, which use Conventionally Indirect strategies types 1 and 2, turn into Direct forms (Weak and Strong) in the Spanish subtitles. Contrastively, the higher percentage of DSAs in the Spanish film scripts are Weak Direct, which are changed into the Strong Direct in the English subtitles. These findings confirm a wider representation of direct type of DSAs in the Spanish corpus than in the English. The shift in the translation presents only one change in the directive type, for example from Weak Direct to Strong Direct, from Conventionally Indirect 1 to Weak Direct or from Conventionally Indirect 2 to Conventionally Indirect 1. The percentages where there are two changes in the directive type, from Conventionally Indirect 2 to Weak Direct, or from Conventionally Indirect 2 to Strong Direct for example, are much lower. This finding indicates the absence of significant shifts in the translation strategies with regard to the level of (in)directness.

The analysis of directives in the English and Spanish scripts reveals that from a linguistic viewpoint, a remarkable number of linguistic structures for directives are found in both corpora. It is apparent that in the English language, more indirect forms of requesting are used, compared with Spanish. As far as translation strategies are concerned, it seems that the most common form is to maintain a faithful translation of the film scripts when subtitled. Where a change in the translation into subtitles occurs, the Spanish subtitles present a variety of both direct and indirect forms of the DSAs, whilst the English subtitles for the Spanish films focus mainly on the use of direct forms.

In both language directions, the shift in the translation from indirect to direct forms of the DSAs is more common than the shift from direct to indirect type. This pattern may be due to the fact that there are some technical aspects of the subtitles which are connected with the reduction of the text: space limitation and reading speed. It can also be argued that in some instances the subtitler’s knowledge of the target culture might influence his/her translation choice. As Colina observes (2015: 95)

[the] translation that does not consider the translation brief and context of the situation (including the characteristics of the audience) runs the risk of transferring only the illocutionary force (word-by-word) and therefore of failing to capture the necessary illocutionary force.

For this investigation, the film scripts and the subtitles of the chosen films, extracted from the DVDs, present several potential limitations. Additional information regarding translation procedure, i.e., whether the subtitlers worked with the film scripts of the video recordings, and the translation strategies used in the creation of subtitles would have provided additional information for the analysis of the translated subtitles. As a consequence, a holistic approach, that takes account of the translation procedures, strategies the target audience and the intricacies, nuances and representation of the language, would be recommended for any future analysis of this type.

Concluding remarks

This chapter has provided an account of the linguistic representation of DSAs in the film scripts and the subtitles of 24 films. Given the size of the corpora, the findings shed light on and reach significant conclusions with regard to the representation of this specific group of speech acts. From the audiovisual translation perspective, the size of the corpora and the significant amount of instances identified in the analysis allow us to corroborate the evidence that the subtitles of the films are kept closer to the ST (source text) audience than to the TT (target text) audience, leading to high fidelity in the translation of DSAs towards the ST audience at speech level.

Finding an appropriate theoretical framework in order to carry out interdisciplinary research, however, is challenging. Blum-Kulka and colleagues’ linguistic taxonomy (1989) has made it possible to ascertain how the speech act of requesting is performed in audiovisual discourse. Despite the necessary adaptation and, in some cases, simplification of the taxonomy, I encourage other researchers to test this framework using different speech acts in AVT. There is also scope for further research to develop a broader view of the linguistic representation of speech acts in audiovisual texts and AVT.

Notes

1 The general term used to describe “the extent to which a TT can be considered a fair representation of a ST according to some criterion” (Shuttleworth & Cowie, 2014: 57).

2 This variable has not been included in the analysis as it is beyond the scope of this investigation.

3 Note that the terms “shift” and “change” are used interchangeably when referring to a change in the translation from the ST to the TT.