UNIVERSITY OF EASTERN FINLAND PHILOSOPHICAL FACULTY
SCHOOL OF HUMANITIES English Language and Translation
Anna Heidi Kristiina Palosaari
TRANSLATION ERRORS IN THE FINNISH SUBTITLES OF THE BRITISH SOAP OPERA EMMERDALE
The usability of analysis on translation errors in the quality assessment of subtitles
MA Thesis November 2016
ITÄ-SUOMEN YLIOPISTO – UNIVERSITY OF EASTERN FINLAND
Tiedekunta – Faculty
Philosophical Faculty Osasto – School
School of Humanities Tekijät – Author
Palosaari, Anna Heidi Kristiina Työn nimi – Title
Translation errors in the Finnish subtitles of the British soap opera Emmerdale: The usability of analysis on translation errors in the quality assessment of subtitles
Pääaine – Main subject Työn laji – Level Päivämäärä –
Date Sivumäärä – Number of pages
English language and translation
Pro gradu -tutkielma x
20.11.2016 88 pages + appendix 2 pages Sivuainetutkielma
Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract
The aim of this study was to assess the usability of analysis on translation errors in the quality assessment of subtitles. This was done by applying a quality evaluation model based on counting and classifying errors to the Finnish subtitles of Emmerdale. Translation errors have not yet gained as much attention in subtitling research as they have in the research of other translated texts. “Amusing” mistranslations are noticed in subtitles but the purpose of this study was to find out what other kinds of errors appear in subtitles and how these errors affect subtitling quality. Furthermore, the recent changes in the audiovisual translation industry inspired me to study whether the subtitling quality of Emmerdale has changed since the commercial television channel MTV broadcasting the series outsourced its translation services to the multinational translation company BTI in October 2012.
The error analysis followed the translation error categorisation by O’Brien (2012). The errors were divided into linguistic, terminological, accuracy, stylistic and subtitling errors This error categorisation was based on quality evaluation models in active use in the translation industry. The linguistic analysis of the errors was mainly based on YLE subtitling guidelines (Vertanen 2007; Appendix 1). The main theoretical background on the current situation of audiovisual translation industry consists primarily of Abdallah’s (2007 and 2012a) and Laine’s (2007) accounts.
The material consisted of the Finnish subtitles of twenty episodes of Emmerdale. The first set of ten episodes were translated by former in-house and freelance translators working directly for MTV, whereas the other set of ten episodes were translated by freelance translators subcontracted by BTI. The analysis was divided into three sections. The quantitative analysis concentrated on counting the errors to see what types of errors are the most common ones in this set of data. The qualitative analysis focused on analysing the features of different error categories and evaluating the severity levels of the errors. The comparative analysis, in turn, compared the quantity and quality of translation errors in the subtitles by MTV to the errors in the subtitles by BTI.
The subtitles of Emmerdale contained a total of 167 errors. The results of the analysis showed that the most common type of translation error in this set of data was a linguistic error. Linguistic errors accounted for over 50 per cent of the total number of errors. The results also implied that the subtitling quality of Emmerdale has changed after the outsourcing. This can be seen as the increase in the quantity of translation errors. The subtitles translated by MTV contained 61 errors, while the subtitles translated by BTI contained 106 errors.
Still, the analysis indicates that this kind of quality evaluation model is not as it is a suitable method for quality assessment of subtitles. The technical restrictions of subtitling, the subjectivity of evaluation and not knowing the company-specific conventions set numerous challenges on classifying errors. In addition, it was found that this particular model does not take into consideration variables, such as genre, text or content type. There is a need for less subjective criteria and adjustments to the error categorisation to fit the data.
Avainsanat – Keywords
subtitling, subtitling quality, quality evaluation, error analysis, translation error
ITÄ-SUOMEN YLIOPISTO – UNIVERSITY OF EASTERN FINLAND
Tiedekunta – Faculty
Filosofinen tiedekunta Osasto – School Humanistinen osasto Tekijät – Author
Palosaari, Anna Heidi Kristiina Työn nimi – Title
Translation errors in the Finnish subtitles of the British soap opera Emmerdale: The usability of analysis on translation errors in the quality assessment of subtitles
Pääaine – Main subject Työn laji – Level Päivämäärä –
Date Sivumäärä – Number of pages
Englannin kieli ja kääntäminen
Pro gradu -tutkielma x
20.11.2016 88 sivua + liite 2 sivua Sivuainetutkielma
Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract
Tutkimuksen tavoitteena on arvioida käännösvirheanalyysin käytettävyyttä tekstitysten laadunarvioinnissa.
Virheanalayysin käyttöä laadunarviointimenetelmänä arvioitiin soveltamalla käännösvirheiden laskemiseen ja luokitteluun perustuvaa laadunarviointimallia Emmerdalen tekstityksiin. Tekstitysten käännösvirheitä ei ole vielä tutkittu samassa mittakaavassa kuin muissa käännetyissä tekstilajeissa. “Huvittavat” käännösvirheet kyllä huomataan tekstityksistä, mutta tämä tutkimus pyrkii selvittämään, millaisia muita käännösvirheitä tekstityksissä esiintyy ja miten ne vaikuttavat tekstitysten laatuun. Suomalaisten av-kääntäjien viime vuosina ennestään heikentyneet työolot innoittivat myös ottamaan selville, onko Emmerdalen tekstitysten laadussa tapahtunut muutoksia sen jälkeen, kun MTV ulkoisti käännöspalvelunsa monikansalliselle käännösyhtiölle BTI:lle lokakuussa 2012.
Tutkimuksessa käytetty käännösvirheluokittelu noudattaa O’Brienin (2012) mallia. Käännösvirheet on jaoteltu kielellisiin, terminologisiin, tyylillisiin ja täsmällisyysvirheisiin sekä tekstitysvirheisiin. O’Brienin malli pohjautuu käännösasiakkaina toimivien yritysten käytössä oleviin laadunarviointimalleihin. Virheiden kielellinen arviointi pohjautuu pääosin YLEn tekstityskonventioihin (Vertanen 2007; Liite 1). Suomalaisten av-kääntäjien työolojen tarkastelu perustuu suurilta osin Abdallahin (2007; 2012a) ja Laineen (2007) artikkeleihin.
Tutkimusmateriaali koostui kahdenkymmenen Emmerdale-jakson suomenkielisistä tekstityksistä. Ensimmäisten kymmenen jakson kääntäjinä ovat toimineet MTV:n oman käännösjaoston tekstittäjät, kun taas jälkimmäisten kymmenen jakson tekstityksistä vastaavat BTI:n freelance-kääntäjät. Analyysi jakautui kolmeen osa-alueeseen.
Määrällisen analyysin tavoitteena oli laskea tekstityksissä esiintyvät käännösvirheet, jolla pyrittiin selvittämään, mitkä käännösvirheet ovat tyypillisimpiä tässä aineistossa. Laadullinen analyysi taas pyrki selvittämään, millaisia virheitä kussakin käännösvirhekategoriassa esiintyy, ja arvioimaan virheiden vakavuutta. Vertailevan analyysin tavoitteena puolestaan oli selvittää, onko MTV:n ja BTI:n tekstityksissä esiintyvien käännösvirheiden välillä määrällisiä tai laadullisia eroja.
Emmerdalen tekstitykset sisälsivät yhteensä 167 virhettä. Tulokset osoittavat, että tämän aineiston tyypillisin käännösvirhe on kielellinen virhe. Kielelliset virheet muodostivat yli 50 prosenttia kaikista aineistossa esiintyvistä käännösvirheistä. Lisäksi tekstitysten laadussa on huomattavissa muutoksia käännöspalvelujen ulkoistamisen myötä, mikä on nähtävissä käännösvirheiden määrän lisääntymisenä. MTV:n työsuhteisten kääntäjien tekstitykset sisälsivät 61 käännösvirhettä, kun taas BTI:n freelance-kääntäjien tekstitykset sisälsivät 106 virhettä.
Virheanalyysi ei sellaisenaan sovellu täysin tekstitysten laadunarviointiin. Tekstittämisen tekniset rajoitukset, arvioinnin subjektiivisuus ja yritys- tai kanavakohtaisten konventioiden puuttuminen asettavat useita haasteita virheiden luokitteluun. Lisäksi nousi esiin, että tämä laadunarviointimalli ei juuri ota huomioon muuttuvia tekijöitä, kuten genreä, tekstilajia tai sisältötyyppiä. Tekstittämisen erikoispiirteisiin vastaavan laadunarviointimallin
kehittämiseksi ja arvioinnin subjektiivisuuden vähentämiseksi käännösvirheluokittelu olisi muokattava arvioitavan materiaalin mukaan.
Avainsanat – Keywords
tekstitys, tekstitysten laatu, laadun arviointi, virheanalyysi, käännösvirhe
Sisällys
1. Introduction ... 1
2. Subtitling ... 5
2.1. Subtitling as a special form of translation ... 5
2.2. The role and function of subtitling ... 8
2.3. Subtitling industry in Finland ... 12
3. Subtitling quality ... 17
3.1. The definition of quality in the subtitling industry ... 17
3.2. Approaches to translation quality assurance ... 20
3.3. Quality and reception ... 23
3.4. Translation errors as tools for quality assessment of subtitles ... 27
4. Research material ... 34
5. Research method ... 38
6. Analysis ... 48
6.1. Quantitative analysis – the number of translation errors ... 48
6.2. Qualitative analysis – the characteristics of different error types ... 50
6.2.1. Linguistic errors ... 50
6.2.2. Accuracy errors ... 58
6.2.3. Stylistic errors ... 63
6.2.4. Subtitling errors ... 67
6.3. Comparative analysis – the differences between the two translator groups ... 68
6.4. Summary and discussion ... 73
7. Conclusion ... 79
References ... 82
Appendix 1. The YLE subtitling guidelines in Finnish ... 89
1
1. Introduction
The audiovisual translation industry in Finland has undergone dramatic structural changes in the past few years. A growing number of privately owned television channels have outsourced their translation services to multinational translation companies, and one of the most recent outsourcings occurred in October 2012 when the commercial Finnish television channel MTV (at that time known as MTV3) decided to outsource their subtitling activities to the
multinational translation company BTI Studios Oy (at that time known as Broadcast Text International Oy). BTI Studios Oy (from now on referred to as BTI) offered to keep the translators working directly for MTV, but at significantly lower rates. Consequently, the majority of the translators working for MTV refused the new terms and conditions of employment offered and resigned from their jobs. As a result, freelance translators subcontracted by BTI took over. In the wake of these changes I noticed that the subtitling quality of my favourite television programme, the British soap opera Emmerdale, changed.
My observations inspired me to study the changes and the possible reasons for them more closely.
The main focus of this study is to assess the usability of analysis on translation errors in the quality assessment of subtitles. This will be done by applying a quality evaluation model based on counting and classifying errors to the subtitles of Emmerdale. The quality evaluation model used in this study is based on models which are in active use within the translation industry (O’Brien 2012). Furthermore, the purpose is to establish what kinds of translation errors occur in the subtitles. Translation errors in subtitles have not yet gained as much attention as they have in other translated texts, such as journalistic translations (see for example Vehmas-Lehto 1989) or technical translations (see for example Schmitt 2002). As Jääskeläinen (2007: 117) points out, even though research on subtitles has become extremely popular especially among students, there is actually very little research on subtitling quality in general, let alone from the point of view of error analysis. As a matter of fact, most studies analyse subtitles only on the level of language concentrating on, for example, how culture- bound elements or humour are conveyed in subtitles. Jääskeläinen (2007: 118) notes that in these types of studies subtitles provide the research material but are not the actual object of research. In the present study, however, the subtitles and their quality are the focus of attention.
2 In the context of this study, subtitles and their quality will be evaluated by
identifying and analysing translation errors in the Finnish subtitles of Emmerdale. As mentioned in the previous paragraph, translation errors in subtitles have not gained much scholarly attention yet. In my opinion, one of the reasons for this might be the various
constraining elements the audiovisual environment poses on translating audiovisual material.
Condensation and omission are essential translation strategies in subtitling, which could make the researchers unwilling to tackle the issue of translation errors in subtitles as it might be difficult at times to distinguish translator’s personal choices from outright translation errors.
Additionally, the assumed unwillingness to acknowledge the possible existence of translation errors in subtitles might be due to the fact that appreciation of subtitling is not very high, and general attitude towards subtitles, especially among viewers, tends to be somewhat critical (Abdallah 2007; Tuominen 2013). Some viewers consider subtitles irrelevant and even claim not needing subtitles at all, but still their comments on subtitles give indications that subtitles are in fact read and noticed. Viewers’ comments tend to concentrate on mistranslations and subtitling blunders which are accepted either with a shade of irritation or with amusement (Abdallah 2007; Tuominen 2013). Despite the lack of appreciation, subtitles are indeed a significant support for understanding the audiovisual text and a method for learning foreign languages or even the mother tongue (Vertanen 2007; Ghia 2012; Tuominen 2013).
In addition to testing error analysis, I will observe the potential effects of outsourcing translation services on subtitling quality by examining whether there have been any changes in the quantity and quality of translation errors since the outsourcing in October 2012. The aim is to connect the subtitling quality of Emmerdale to the bigger picture of the recent changes in the audiovisual translation industry. The results of the study will be mirrored against the findings of other researchers to see whether my results support their views. As some researchers (Abdallah 2007, 2012a; Laine 2007) have noted, outsourcing translation services affects the working conditions of private sector translators, which might prevent the translators from doing their work properly and producing good quality. Both Abdallah (2007, 2012a) and Laine (2007) argue that the working conditions of freelance translators employed by multinational translation companies are poorer than those in the public sector. For
example, the pace of subtitle production in the private sector is often fast and the translators are working under very strict time limits. Furthermore, the translation fees are significantly lower than in the public sector which makes translating speed essential to earn a living.
Subtitling quality can be argued to be of major significance to viewers and readers especially in Finland as subtitles reach a large number of people and are a part of everyday
3 life. Almost all of foreign television programmes in Finland are subtitled (Vertanen 2007:
149), and in the light of Mäkisalo’s (2006) and Salmi’s (2010) studies, it is safe to say that subtitles are the most often read texts among some Finns. Mäkisalo (2006: 250–251, 254) collected weekly reading diaries from translator students, and the results showed that the largest proportion of texts read daily by the students consisted of subtitles. The respondents spent approximately 87 minutes daily reading subtitles, which accounted for 32.1% of the total daily reading time (Mäkisalo 2006: 255). Salmi (2010) had somewhat similar results.
Salmi (2010: 6) examined the reading habits of translator students as well, and the results indicated that subtitles constituted the second largest proportion of texts read daily by the students. An average of 56 minutes of the total daily reading time was spent on reading subtitles (Salmi 2010: 5–6). Hence, it is evident that subtitles should be considered to be of great importance and their quality is not irrelevant.
One of the challenges of this study is the subjectivity of assessment. Assessment of quality is often based on subjective views (Abdallah 2007; Tuominen 2013), and the fact that the object of research of the present study is the subtitles of my favourite television
programme might affect how I assess the subtitles and their quality. I have followed the series for over 10 years which, I think, makes me something of an expert on the series. However, when a viewer follows a series for that long, they might develop a personal attachment to the places, residences and characters in the series. According to Hietala (2007: 27), viewers can even develop parasocial relationships to the characters. This means that viewers might regard the characters as actual people, and they extend real and genuine feelings towards them (Hietala 2007: 27). Even though I do not regard my relationship with Emmerdale as parasocial, I do have some sort of personal attachment to the series which might make my subjectivity visible at times.
To sum up, the main focus of this study is to assess the usability of analysis on translation errors in the quality assessment of subtitles. Furthermore, the study aims to establish what kinds of translation errors occur in the Finnish subtitles of Emmerdale and to examine the possible implications outsourcing translation services might have on subtitling quality. The research material consists of the subtitles of twenty episodes of the British soap opera Emmerdale broadcast on MTV from autumn 2011 to the beginning of the year 2013.
The first set of ten episodes were translated by former in-house and freelance translators working directly for MTV, while the other set of ten episodes were translated by freelance translators subcontracted by BTI. The analysis is divided into three sections. In the analysis on translation errors both quantitative and qualitative methods are employed. Additionally, a
4 comparative analysis will be conducted to see whether there have been any changes in the quantity and quality of translation errors after MTV outsourced its translation services to BTI.
The main research questions of this study then are:
Is error analysis a suitable method in the quality assessment of subtitles?
Do the constraining elements of subtitling set any challenges on this quality assessment method?
What kind of translation errors occur in the subtitles of Emmerdale?
What are the most typical types of translation errors in this set of data?
Has the subtitling quality changed after MTV outsourced its translation services to the multinational translation company BTI?
Are there any differences in the quantity and quality of translation errors between the subtitles translated by former in-house and freelance translators working directly for MTV and the freelance translators subcontracted by BTI?
This thesis is divided into seven chapters. The next chapter discusses the special features and functions of subtitling, while focusing on the subtitling industry in Finland. The third chapter then deals with various aspects of quality in the context of subtitling and introduces the concept of translation error as defined by Hansen (2010) and Vehmas-Lehto (2005). Furthermore, the third chapter presents the quality evaluation model by O’Brien (2012) which serves as the basis of the error analysis. In chapters four and five the research material and methods are introduced. Chapter 6, in turn, presents and discusses the results, treating the three parts of the analysis in separate sections. Finally, chapter seven concludes this research by summarising the main findings of the analysis and highlighting the problems and questions this study raised.
5
2. Subtitling
This chapter concentrates on discussing subtitling as a special form of translation and the challenges the audiovisual environment poses on subtitling. Firstly, the definition of subtitling and the features that make subtitling a special form of translation will be introduced.
Secondly, the role and function of subtitling will be discussed. Finally, the subtitling industry in Finland will be described and some factors in the current situation that might be affecting subtitling and its quality will be addressed.
2.1. Subtitling as a special form of translation
Subtitling is a form of audiovisual translation which is defined as the act of translating any relevant information from the source language into a written form of the target language (Díaz Cintas 2010: 344). The underlying concept of subtitling does not, thus, differ from other forms of translated texts. The basic idea is to convey the message of the source text to the target text audience. However, the combination of sound, image and text makes subtitling a special form of translation. Subtitling does not only include translating the original dialogue uttered by the speakers but also translating other discursive elements that appear in the image, such as banners and inserts, and information that is transmitted aurally, such as songs and voices-off. Therefore, all subtitled programmes consist of three elements which are the
original spoken or written word, the original image and the added subtitles (Díaz Cintas 2010:
344). These three elements together form the audiovisual text.
In addition to this straightforward definition, Díaz Cintas (2010: 344) points out that the audiovisual environment sets certain constraints on subtitling which are necessary to take into consideration when discussing subtitling as a form of translation. Elements such as the size of the screen, the timing of the programme, the pace of the dialogue and visuality all have an effect on the content of the subtitles. In order to comply with these constraints, and to make it easier for the audience to follow the subtitles, some more or less universal subtitling norms and conventions are employed. The norms and conventions introduced here are a combination of universal (Díaz Cintas 2010) and national (see Appendix 1 for YLE subtitling guidelines; Vertanen 2007) guidelines. First of all, subtitles are generally placed horizontally on the lower part of the screen, and each subtitle can take up a maximum of two lines, most often with 35–40 characters per one line (Díaz Cintas 2010: 344). Secondly, subtitles should
6 appear in synchrony with the original image as well as the original dialogue, and thirdly, a full two-line subtitle should stay visible on screen for four to six seconds so that viewers have enough time to read it comfortably (Díaz Cintas 2010: 344–345; Vertanen 2007: 151).
While complying with the technical restrictions discussed above, the main purpose of subtitles is to “provide a semantically adequate account of the source language dialogue”
(Díaz Cintas 2010: 345) and to convey the style and atmosphere of the source text (Vertanen 2007: 150). However, considering the space and time limitations, it is quite obvious that subtitles cannot translate everything word for word. The translator must find ways to condense the content of the source text into its most essential parts. Vertanen (2007: 152–
153), who has worked as a translator and a journalist for Finland’s national broadcasting company YLE for some twenty years, lists some typical elements that can be omitted from the Finnish subtitles if they do not contain any relevant information for understanding the plot or are explicitly conveyed through the image. These include various leading clauses, such as
“minusta tuntuu, että” [I feelthat]and “luulen, että” [I think that], names and titles of the characters in case the characters have already appeared in the programme, attributive adjectives, place names and time adverbials. Although condensation and omission are the main strategies used in subtitling, the translators are advised to create subtitles that make sense as individual units of texts and use natural language while adhering to the grammatical rules of Finnish (Vertanen 2007: 152–153). According to Tuominen (2013: 32), the Finnish translators tend to use complete sentences and correct, varied language in subtitles even when condensation could be necessary. It could be argued that this tradition corresponds to viewers’
needs and preferences in Finland but, as Tuominen (2013: 32) rightly points out, it should be kept in mind that these kinds of opinions are culture-bound as subtitling conventions vary between countries.
Another special feature of subtitling is that it involves translating spoken language into written form. Even though subtitles appear in written form, Vertanen (2007: 153) advises that they should aim to create an illusion of spoken language. As the subtitles most often represent someone’s speech, they should not be “stripped too bare” (Vertanen 2007: 153).
However, many elements of spoken language, such as features of slang or dialect, can be easily understood when they are spoken but may be fairly difficult to convey to viewers in written form. Therefore, these elements should be translated in an easily readable style so that viewers are able to read and understand them in the few seconds the subtitle is visible on screen. Most often this means that translators are more prone to using easily understandable standard language in subtitles than play around with different language variants. Still, when
7 the translator decides to use some form of linguistic variation in subtitles, Vertanen (2007:
153) emphasises that they should use them consistently throughout the programme.
Both Vertanen (2007) and Díaz Cintas (2010) emphasise that subtitles should be easily understood on the first reading because viewers do not usually have the possibility to go backwards to read the subtitles again as they have with printed translations. Of course, the technology today allows, for instance, the possibility to pause and rewind when watching a television programme but pausing might distract the viewer and draw unnecessary attention away from the audiovisual text. Viewers should be able to concentrate on watching and listening to the programme as well as reading the subtitles without the subtitles drawing unnecessary attention. Accompanying sounds and images can help in understanding subtitles but the aim is that the subtitles support the viewing experience and understanding the source text, not the other way round. Effortless readability can be achieved, for instance, through spotting and line-breaking. These include marking the change of speaker in a two-line subtitle with a hyphen and marking a sentence continuing on the next line with a space and a hyphen (Vertanen 2007: 154). In addition, to enhance readability, words that are closely connected by logic, semantics or grammar should be written on the same line or subtitle if possible (Díaz Cintas 2010: 345; Vertanen 2007: 154).
According to Díaz Cintas (2010: 346), subtitling is in a “vulnerable situation”
because anyone with even the slightest knowledge of the source language is able to compare the source text to the target text. It is, of course, logical and necessary for the translators to assume that viewers cannot understand the source language, but particularly with a global language such as English it is typical to understand at least some of it. This “vulnerability”
can add to the challenges of subtitling as the translator has to bear in mind the customer opinion as well. In other words, what viewers consider being an acceptable translation (Tuominen 2013: 29). James (2001: 152) correctly argues that subtitling conventions are indeed “intended to provide guidelines to ensure consistent high quality” but as most viewers do not have a precise knowledge of the restrictions of the audiovisual environment or
subtitling conventions, dissent cannot be avoided. However, as Diaz Cintas (2001: 199) puts it, “in addition to time and space constraints, there are some metatextual factors that are at least as important in determining the quality of the end product”. These are concerned with the working conditions under which the translators are working. The working conditions of audiovisual translators are discussed in more detail in Chapter 2.3.
In conclusion, a good subtitle reproduces the original content of the source text while complying with the technical conventions of subtitling and using natural and grammatically
8 correct Finnish. As Kurvi’s (2013) study shows, these aspects seem to be what translators themselves consider being conducive to high-quality subtitling as well. Kurvi (2013) studied the working conditions of audiovisual translators in a survey and when the translators were asked to define high-quality subtitling, five themes were highlighted above others. The translators appreciated most of all that the subtitles appear in synchrony with the original dialogue, remain faithful to the style and atmosphere of the source text and condense the original content of the source text successfully. In addition, the translators emphasised that the language of the subtitles should be vivid and natural and it should not contain grammatical errors or misspellings (Kurvi 2013: 32). Due to the various challenges and restrictions, it is nevertheless obvious that subtitles cannot remain as faithful to the source text as other
translated text types, such as literary translations, can. Instead, as Tuominen puts it (2013: 26–
27), subtitles represent the translator’s view of what was most relevant in the source text.
Translators are expected to follow general rules and conventions and make decisions based on contextual constraints but the subtitle that appears on screen always reflects the translators’
personal vision of what they think is the most important and relevant information for the viewer (Tuominen 2013: 27). This could be an issue with the present study as evaluation of an error as a translation error depends on the evaluator’s subjective views on what they consider to be an acceptable translation. In some cases this might to lead to questioning the translator’s personal vision.
2.2. The role and function of subtitling
It could be argued that subtitling is at its best when the viewer does not even notice reading the subtitles (Mueller 2001: 147; Jääskeläinen 2007: 119). In an ideal situation, as Tuominen (2013: 30) correctly argues, “subtitles flow easily within the multimodal text and support the viewing experience by providing necessary information at a suitable pace”. In other words, subtitles should become part of the programme flow, appearing in synchrony with the original dialogue and the image (Díaz Cintas 2010). Skilfully created subtitles are read without even noticing them which might lead the viewer to believe that they have understood the source text itself, instead of having read the, at times, significantly condensed version (Tuominen 2013: 28; Vertanen 2007: 151). As discussed in the previous chapter, subtitles should reproduce the original content of the source text while complying with the technical constraints governing subtitling as well as the grammatical rules of the target language.
However, Vertanen (2007: 150–151) makes clear that merely providing the essential meaning
9 of the source text does not fulfil the function of subtitles properly, but they should also
convey the style and atmosphere of the source text. This requires at times real creativity from the translators in order for them to produce easy-to-read and comfortable texts for the target audience and still remain faithful to the source text.
Jääskeläinen (2007: 119) notes that viewers themselves seem to be under the impression that subtitles are irrelevant and not actually even needed since most of them are able to understand the source text, or at least some of it, without the subtitles. But if the translator succeeds in creating subtitles which form a coherent whole with the original dialogue and image, they are read without paying particular attention to them. This
characteristic of subtitling together with a relatively good command of English helps to create the illusion of better knowledge of the language than what the viewer actually has
(Jääskeläinen 2007: 119). Research by Tuominen (2013) gives clear indications that subtitles are “a significant support for understanding” in the viewing experience. Tuominen (2013) studied the reception of subtitled films through focus group discussions. Even though some of the viewers in Tuominen’s (2013: 270–278) study denied reading subtitles at all, in reality, most of them were unable to avoid reading the subtitles while watching a foreign language film. Based on the discussions, Tuominen (2013: 270) distinguished four strategies that the viewers use as they process the source text and the subtitles when the source text is in English. In the first strategy the viewer only listens to the source text, whereas in the second strategy the viewer concentrates only on reading the subtitles. The third strategy involves the viewer listening mainly to the source text and supporting this with reading the subtitles. The fourth strategy, in turn, involves the viewer reading mainly the subtitles and supporting this with the source text. Tuominen (2013: 271) found that, contrary to viewers’ own perceptions, none of them were able to concentrate solely on the source text. That is, none of them were able to rely on the first strategy alone. All of the informants had noticed the subtitles and received some information through them. On the basis of the results, Tuominen (2013: 316) correctly argues that “subtitles are a tool, a supportive text that helps in absorbing the audiovisual message”. They are intended to be read quickly and without paying particular attention in the multimodal flow of the audiovisual text (Tuominen 2013: 36), and if they fulfil their function, they are read without even noticing them which leaves the viewer with the impression that they have understood the source text itself.
However, the previously described form of conventional subtitling that strives to comply with the technical and linguistic constraints while creating easy-to-read and
unnoticeable subtitles does not satisfy all researchers. Nornes (2004, quoted in Pasanen 2015:
10 16), for one, suggests that all subtitles are violent and conventional subtitling is corrupt. On one hand, Nornes (2004, quoted in Pasanen 2015: 16) argues that subtitles which promote effortless readability, follow the target culture’s linguistic norms and strive to become part of the programme flow are corrupt. According to this view, conventional subtitling, which seems to be the norm in Finland, is corrupt. On the other hand, Nornes (2004, quoted in Pasanen 2015: 16) claims that subtitles which attempt to direct the audience to the source text by maintaining and emphasising the foreignness of the source text are abusive. Corrupt subtitles, thus, are an unnoticeable part of the audiovisual text, whereas abusive subtitles do not try to remain invisible. Whereas corrupt subtitles employ fixed conventions such as the ones discussed above, abusive subtitles employ unconventional means but still remain faithful to the source text and the author. Even though conventional subtitling, that is corrupt subtitling according to Nornes (2004, quoted in Pasanen 2015), also aims to remain faithful to the source text, adapting the source text to follow target language norms and subtitling conventions and omitting parts of the source text is what makes conventional subtitling especially violent. Nornes’ (2004, quoted in Pasanen 2015) theory of abusive subtitling concentrates on the translators’ visibility and fidelity to the source text and the author and he suggests that translators should make themselves visible and not hide the act of translation.
The opinion of Nornes is that abusive subtitling would actually bring the viewer closer to the source text.
McClarty (2012, quoted in Pasanen 2015: 26) agrees with Nornes (2004) by admitting that subtitling conventions are useful in quality control but stating that they have resulted in translations that are alike. McClarty (2012, quoted in Pasanen 2015: 27), in turn, introduces her view of creative translation which emphasises immersion and aesthetics and suggests that subtitles should be created as part of the film or the programme. It is the view of McClarty (2012, quoted in Pasanen 2015: 27) that creative subtitles would be achieved through variations in font, text size, colour, position and integration to the background. Both abusive and creative translation underline the notions of experimentation, visibility and subjectivity but they raise the question of how these kinds of new forms of subtitling would be received in a subtitling country such as Finland which is used to watching foreign television programmes with subtitles that are uniform and follow specific set of rules and conventions. As Pasanen (2015: 29) points out, one issue that comes to mind with creative subtitling is subtitle positioning. In creative subtitling, the subtitles are not always placed horizontally at the bottom of the screen which might make it difficult for viewers to predict where the subtitle will appear next. Viewers might miss the subtitles altogether or not have
11 enough time to read them completely if they do not pay enough attention. In turn, trying to predict where the subtitle will appear next and in what colour might draw attention away from the programme itself. Alternatively, in abusive subtitling, choice of style may cause confusion among viewers. This could especially be the case with viewers who can understand both the source and target language. As the style chosen by the translator might differ from the source text, Pasanen (2015: 21) correctly argues that it could therefore raise the audience’s interest by making them wonder what it was in the source text that made the translator choose this particular style. There is a need for reception studies to see how the audience would response to these kinds of new modes of subtitling. While Nornes (2004) argues that abusive subtitling could eventually be applied to all genres, McClarty (2012) admits that creative subtitling would not suit all genres but the style should be chosen individually for the film in question. I agree with McClarty (2012) in that abusive or creative subtitling would not suit all kinds of genres but the “appropriate” way of subtitling depends on the programme. For example, art films might be appropriate contexts for creative subtitling, whereas children’s programmes might benefit more of the conventional form of subtitling because children’s perceptual ability might not be developed enough to follow creative translation. Soap operas, such as the research material of the present study, might be an ideal genre to employ abusive subtitling.
Soap operas contain exciting plot twist, colourful characters from different backgrounds and features of a foreign culture and, thus, abusive subtitling, might bring the viewer closer to the source text.
Subtitles can also be seen as a learning method. Vertanen (2007: 149) daringly claims that Finnish children’s great literacy skills result partly from subtitling. In my opinion, this might well be true as most Finns are used to viewing subtitled programmes and movies already from the beginning of their childhood, and as Mäkisalo’s (2006) and Salmi’s (2010) studies indicate, subtitles are the most often read texts among some Finns. Vertanen’s (2007) view is supported by Tuominen’s (2013) findings. Her study (2013: 174–176; 319–320) shows that viewers often see subtitles as a method for learning not only the source language but also the target language. The subtitling tradition in Finland to use grammatically correct language and complete sentences, as previously discussed, could be explained by the fact that subtitles have an important role as a learning method. Ghia’s (2012) study supports
Vertanen’s (2007) and Tuominen’s (2013) views as well. Ghia (2012: 101) studied the foreign language acquisition through subtitles by showing British and American television series subtitled in their native language to 22 Italian viewers who were learning English as a foreign language over a three-month period. The viewers’ syntactic skills were assessed through tests
12 before and after the exposure to the foreign language programmes. Despite the relatively small sample, the results indicated that a foreign language programme subtitled in the
viewer’s mother tongue can improve learning the syntax of the foreign language. Ghia (2012:
108) notes that the improvement in syntactic skills was evident in “the arrangement of main constituents in clauses and phrases, the formation of interrogative structures and the
production of higher-complexity patterns, such as tags and subordinate clauses”. Ghia’s (2012: 111) results suggest that subtitles have a positive effect on the development of general competence and syntactic skills in the foreign language.
Kurvi’s (2013) study on the working conditions of audiovisual translators, in turn, shows that translators consider the role of subtitling important as well. When discussing the role and importance of subtitling, the translators highlighted two themes. For one, the translators considered subtitling a support for understanding the source language and the programme. At the same time, the translators see the correct and creative language of subtitles as an effective way to improve the native language skills as well as the foreign language skills of the viewers (Kurvi 2013: 25). Thus, Kurvi’s (2013) findings agree with both Tuominen’s (2013) conclusions on subtitling being a support for understanding and with Vertanen’s (2007) and Ghia’s (2012) arguments that subtitling might have an effect on improving the general literacy skills of viewers in both the target and the source language. All of these studies are clear reasons why subtitling quality should not be compromised and quality requirements should be set high.
2.3. Subtitling industry in Finland
The field of subtitling industry in Finland is divided into two distinct sectors: the public and the private sector. The former sector consists of translators working for the Finnish public broadcasting company YLE. YLE has an in-house translation department but it employs subcontracted translators as well (Abdallah 2007: 287; Forssell 2016). What sets YLE apart from private sector companies is that YLE employs most of its translators directly, without translation companies as intermediaries (Abdallah 2007: 273). YLE uses a negotiated
contract, the collective Yhtyneet agreement, as the basis of employment which guarantees the translators it employs favourable working conditions and secure and financially stable
working environment. However, in the recent years, YLE has also outsourced some of the translation work to translation companies but most of the translation work, at least 81 per
13 cent, is still performed by YLE’s in-house and freelance translators working directly with the company (Forssell 2016).
The latter sector consists of translators hired as subcontractors by national or multinational translation companies. Here, I will focus on multinational companies because some of the research material used in this study is produced by translators subcontracted by a multinational translation company. There are two significant multinational translation
companies operating in Finland. These are BTI and SDI Media Oy (Laine 2007: 265).
Abdallah (2012a: 173) observes that multinational translation companies have emerged when lead firms, such as pay television providers and privately owned television channels, decided to outsource the services that were previously performed by in-house or freelance translators working directly with the lead firms. Contrary to public sector translators, translators working as subcontractors to multinational translation companies do not have any direct contact with the customer, rather they are only linked to the customer through the translation companies (Abdallah 2012a: 173). Competition for translation contracts in the private sector is fierce, and as both Abdallah (2007: 277) and Laine (2007: 268–269) point out, remuneration for translation has fallen, resulting in lower fees for the subcontracted translators.
As a result, the terms and conditions of employment of subcontractors in
multinational translation companies have become undoubtedly poorer than those in the public sector (Abdallah 2007, 2012a; Laine 2007). The pace of subtitle production in the private sector is often fast and translators are working under very strict time limits. Sometimes translators are even forced to translate directly from the soundtrack on to the screen without a script (Abdallah 2007: 276). Furthermore, the translation fees are significantly lower than in the public sector. Study by Kurvi (2013) supports Abdallah’s (2007; 2012a) and Laine’s (2007) arguments. Kurvi’s (2013: 30–31) survey of audiovisual translators’ working
conditions showed that most of the respondents who considered their fees to be too low were at the time of the survey working for BTI or SDI Media Oy, that is, the two major
multinational translation companies operating in Finland. The respondents also pointed out that low fees make translation speed essential to earn a living (Kurvi 2013: 30). Additionally, Abdallah (2012a: 177) argues that in order to keep the translation costs low, it is common for the translation companies to employ the cheapest labour possible. The companies have a habit of recruiting translation students still in training who might not have enough experience and skills to negotiate a suitable compensation for their work (Abdallah 2012a: 177). Lång (2013) further argues that inadequate fees do not offer professional translators, or the inexperienced
14 ones either for that matter, enough incentive to stay in the industry and the translators tend to move on to different jobs quite quickly.
Evidently, the emergence of multinational translation companies has raised a great deal of critical discussion. Especially one of the most recent outsourcings where MTV
outsourced its translation services to BTI in October 2012 gained much attention in the media as well as among translators, researchers and other translation professionals. MTV previously used the same negotiated contract, the collective Yhtyneet agreement, as YLE and what concerned most of the people involved in the field of subtitling was that BTI refuses to participate in negotiations of collective bargaining, pays only a fraction of the translation fees formerly paid by MTV, forces the translators to become entrepreneurs and claims the
copyright of the subtitles (Sorsa 2012). All of these factors were suspected to result in subtitling quality suffering (Pääkkönen 2012). Subtitling quality will be discussed more in Chapter 3.1.
Consequently, all of the factors mentioned above might prevent the more
inexperienced, as well as the experienced, translators from doing their work properly and the subtitling quality can suffer (Abdallah 2007: 277). Kurvi’s (2013: 30–31) findings agree with this, as the translators admitted that honing the subtitles into perfection does not “pay off”
with the current translation fees. It should be noted, though, that working under such conditions does not automatically mean that the subtitling quality is poor. However, all of these factors together put pressure on the translators to complete their work quickly, and the translators might be more likely to make mistakes. Hietamaa’s (2012) study expresses a similar view. Hietamaa (2012) analysed the subtitles of the show Star Trek: Deep Space Nine which was translated by two different groups. The first four seasons of the show were
translated by a single translator who worked directly as a freelancer for the channel while the latter three seasons were translated by a total of thirteen translators who worked for a
multinational translation company. The approach of the study is somewhat similar to the approach of my comparative analysis which will be further discussed in chapters 4 and 5. The purpose of Hietamaa’s (2012: 16–17) study was to compare how well the latter seasons of the show comply with the terminology and register which had been established by the translator of the first four seasons. The results of the study show that the latter three seasons are a chaos of mixed solutions which deteriorate the quality of the target language texts (Hietamaa 2012:
66). As Pedersen (2011: 215) and Hietamaa’s (2012) note, television series are often subtitled by a team of translators which makes collaboration on term usage and other translation issues that might arise important in order to maintain consistent high-quality. Hietamaa combined
15 the qualitative analysis with ethnographic interviews with the translators involved in the subtitling process of this particular programme. From the interviews, Hietamaa (2012: 50) draws the conclusion, as Abdallah (2007, 2012a) and Laine (2007) do, that the poor working conditions indeed lead to translations of poor quality. Subtitling quality will be discussed in more detail in Chapter 3. Additionally, it can be argued that the current working conditions of audiovisual translators do not encourage translators to experiment with new forms of
subtitling, such as abusive subtitling (Nornes 2004, quoted in Pasanen 2015) and creative subtitling (McClarty 2012, quoted in Pasanen 2015) discussed in Chapter 2.2. I agree with Pasanen (2015: 25–26) in that the translators working under strict deadlines with insufficient pay cannot be expected to produce highly creative subtitles. Employing fixed and familiar conventions might be the easiest and fastest form of subtitling and thus, the form that helps earn the most.
Improving the working conditions of audiovisual translators is crucial as the subtitling industry will not function without the translators (Abdallah 2007: 286). As
Tuominen (2013: 40) points out, there is light at the end of the tunnel. Translators working as subcontractors have begun to organise themselves and negotiate collectively in order to provide translators more stable working conditions. This might well be a result of generation Y entering the working life. Generation Y is a group commonly agreed to have been born between 1982 and 2000 and they seem to be different from other generations (Kultalahti &
Viitanen 2014). According to Meier at al. (2010), this generation is confident, independent and goal-oriented, and Kultalahti and Viitanen (2014) further emphasise that they have, for instance, clear ideas about work ethics, work-related goals and what they expect of working life. Meier et al. (2014: 71) suggest that salary, regarding pay and benefits, may have the greatest impact on Generation Y employees. Generation Y does not have any problems moving on to another job that will offer them what they want. In other words, if the company does not offer decent pay and benefits, they might have problems attracting young employees in the future (Meier at al. 2014: 75). This generation seems to be more aware of the demands of working life and their rights. This suggests that it is not the employees who have to
measure up to the employer’s standards, rather the other way round. In my opinion, it seems that Generation Y might be bringing much needed changes into the working world and the field of subtitling might benefit from it as well.
However, this has its downsides as well. Generation Y is presumably more used to the fast-paced life which, at the same time, means that it becomes bored more easily. They tend to move on to different jobs quite quickly if they are not happy with the working
16 conditions. This poses a risk that also Tuominen (2013: 39) addresses. Subtitling will no longer be considered as a lifelong career, rather as an entry-level job into the field of
professional translation (Tuominen 2013: 39). It is clear, however, that there will be an even higher demand for subtitling as the number of television channels and audiovisual releases keep rising (Tuominen 2013: 38). At the same time, however, when the number of releases and channels rises, the audience is exposed to foreign languages even more which could improve their understanding of foreign languages and reinforce their opinions about the irrelevance of subtitles. Still, I agree with Tuominen (2013: 38) in that “it is unlikely that subtitles will lose their significance in the Finnish media”. Thus, it is time to start paying attention to the working conditions of private sector translators in order to ensure high quality in the future as well.
17
3. Subtitling quality
This chapter concentrates on identifying what quality in the subtitling industry means. Firstly, the differences in quality definitions between the academic and the economic environments will be discussed. Secondly, some relevant approaches to subtitling quality assurance will be introduced. The third chapter deals with quality in relation to reception. Finally, the concept of translation error as a tool for quality assessment will be introduced.
3.1. The definition of quality in the subtitling industry
The Oxford Dictionary defines quality in terms of “the degree of excellence of something”.
This dictionary definition does not differ all that much from the definition of quality in Translation Studies where the term “quality” has traditionally been used to refer to the degree of excellence of a translation (Abdallah 2007; Chiaro 2008). This traditional view
concentrates on comparing translations to their source texts or target language parallel texts while emphasising translators’ expertise and responsibility to produce acceptable translations (Abdallah 2007: 275). As Jääskeläinen (2016) and Abdallah (2007) correctly argue, in Translation Studies quality tends to be discussed only in terms of the quality of the products.
However, Abdallah (2007: 275) rightly points out that evaluating quality of the product is largely based on the evaluator’s subjective criteria and views on quality. Abdallah (2007: 275) calls this widely recognised way to define quality in Translation Studies product-centred quality.
The academic translator training at universities concentrates on the product-centred quality standards as well by giving students normative advice and emphasising the translation product and its quality while disregarding the real life working conditions (Abdallah 2007:
276; 2012a: 181). Product-centred quality standards aim at translational excellence.
Nevertheless, Abdallah (2007: 275) notes that this mind set enables the evaluators to divide the translators into good and bad translators on the basis of their translations alone, without taking into consideration the current changes in the industry and how they affect the working conditions of translators and their ability to produce good quality translations.
Abdallah (2007; 2012a) goes on to argue that in the economic environment where privately owned television channels outsource their translation services to multinational translation companies, the product-centred quality standards do not quite apply. Whereas
18 translators themselves as experts, as well as the wider public in general, understand quality to mean good quality, the business definition of quality often means the degree to which the product quality co-varies with price (Abdallah 2007: 280; 2012a: 179). This basically means that the subtitling quality a client gets is only as high as the price they pay. Another expression Abdallah (2012a: 179) uses for the definition of quality in the economic
environment is fitness for purpose which, in turn, means that quality should fulfil the client’s specifications, needs or desires. These two definitions of quality indicate that the basic
principles of product-centred quality translators have internalised at university might not necessarily be valid in the economic environment (Abdallah 2007: 276–277; 2012a: 182) where “demands on quality may often be in conflict with the time and resources available”
(Jääskeläinen 2016). This means that, for instance, short deadlines and lack of sufficient source material, which seem to be typical of private sector subtitling as discussed in Chapter 2.3., might prevent translators from doing their work according to the standards and norms acquired during university training. Translators might want to do what they have been taught is their responsibility but do not have enough support or resources to do the right thing (Abdallah 2012a: 181–182).
Evidently, the definition of quality among translators differs from that of the translation companies and as Chiaro (2008: 244) correctly argues, there is a distinct gap between the real life working conditions and what is discussed in the academic community.
As discussed in Chapter 2.3., competitive bidding among translation companies has led to low fees, very tight deadlines and the hiring of students as translators (Abdallah 2012a: 183).
Abdallah (2012a: 183) draws attention to the fact that “in Translation Studies, such
parameters are not considered to be conducive to good quality”. It should be noted, though, that hiring a professional translator does not automatically lead to higher-quality products.
Jääskeläinen (2016) reports that studies on the quality of the translations produced for
research purposes have revealed that professionals do not always produce high quality and are sometimes outperformed by students. While some translation companies try to keep the costs at the lowest level possible, the amount of work load per translator increases and the deadlines become shorter, so the translators might not be able to do their work properly (Abdallah 2007:
277; Laine 2007: 268–270). Companies might even disregard quality standards in order to retain their position as the cheapest service provider in the industry (Abdallah 2012a: 181). As a result, the translators might not able to work according to the standards and norms learned during academic translation training. Abdallah (2007: 277) argues that translators are put in a very difficult position. They either have to compromise what they have previously learned or
19 to produce good quality no matter what it takes. As in Abdallah’s case study (2012a: 181), this might result in translators suffering from professional ethical stress when good quality has to give way to “anything will do” principles.
Abdallah (2007: 273) rightly points out that outsourcing translation services has increased the number of actors involved in defining quality. Both Abdallah (2007: 273;
2012a: 181) and Kurvi (2013: 49) suggest that oftentimes the quality criteria used in lead firms and translation companies are in conflict with the understanding of quality the translators have acquired during translation training at the university. Lång’s study (2013) however, does not entirely support Abdallah and Kurvi’s arguments. Lång (2013: 53) compared the subtitling conventions and guidelines used by some of the biggest suppliers of subtitles in the Finnish broadcasting market. The four companies that provided Lång their subtitling guidelines or quality manuals for his study were YLE, MTV, SDI Media Oy and Pre-Text. Somewhat surprisingly, Lång discovered (2013: 54–59) that the translation companies provide their translators extremely detailed subtitling guidelines and pay close attention to quality. While there was great variation in the length of guidebooks, the actual content of the conventions varied very little among the four companies YLE, MTV, SDI Media Oy and Pre-Text. The longer guidebooks provide the translators very detailed
instructions on, for example, typographical techniques or the use of punctuation, whereas the shorter ones concentrate on general topics. Even though YLE is known for the quality of subtitles it broadcasts, Lång’s results, nevertheless, indicate that YLE provides its translators only broad guidelines for subtitling. Lång (2013: 58) suggests that the lack of more specified subtitling guidelines could be explained by the interaction between the translators at YLE.
The more experienced translators familiarise the new ones with the working methods and subtitling conventions of the company. This kind of mentoring is only possible when the translators are working under the same roof (Lång 2013: 58). The freedom the translators of YLE are given shows that they are trusted and respected for their work and probably have good possibilities for influencing, for example, translation quality assessment and deadlines.
BTI provided only the table of contents of their quality manual for the study, which then had to be excluded from the analysis (Lång 2013: 53). According to Lång (2013: 53), the table of contents showed, however, that there are nearly a hundred pages in BTI’s audiovisual
translation guidebook, and the company does assure on its website that subtitling quality is achieved by the company’s high standards and by employing the best qualified linguistic translators in the industry (BTI Studios: http://www.btistudios.com/translation-subtitling.html, accessed 10 April 2016). Lång’s results suggest that the differences in subtitling quality must
20 lie elsewhere than in the contents of the quality manuals. This again implies that there is a conflict between theory and practice. This conflict might partly result from the current working conditions which do not enable the translators to follow the quality manuals and subtitling guidelines as they might lack the time and resources to do so. As has been emphasised previously, it should be noted that working under such conditions does not
automatically mean that the subtitling quality is poor. It nevertheless means the translators are more likely to make mistakes, which could affect subtitling quality.
In conclusion, there are three distinct ways to define quality. In the scope of the present study, it could be assumed that the business definition of quality applies to the present situation, where MTV has quite recently outsourced its translation services to a multinational translation company. However, the business definition is in conflict with the understanding of quality the translators have acquired during translation training at the university.
3.2. Approaches to translation quality assurance
Abdallah (2007) points out that due to the recent structural changes in the field of subtitling discussed in Chapters 2.3. and 3.1., the industry has to find a new way to define quality in order to ensure high-quality subtitles in the future. When television channels outsource their translation services to translation companies, the number of actors involved in production of subtitles increases and more actors are now involved in defining quality and, additionally, in defining what constitutes a “good” translation (Abdallah 2012a; Kurvi 2013). Therefore, Abdallah (2007; 2012b) suggests that the field of subtitling, and the audiovisual translation industry in general (for more see Jääskeläinen 2016), should define quality collectively and adopt a wider definition of quality in order to align the quality standards applied in translation quality assurance.
Abdallah (2007) introduces a three-dimensional quality model which is divided into product quality, process quality and social quality. Product quality is what Translation Studies, and usually the target audience as well, understands as translation quality. Product quality refers to the features of the end-product, the translated text, which in this study means the subtitles appearing on the screen. Process quality refers to the translation tools that are used to do the actual work and to the quality of the source material. Social quality, in turn, refers to the working conditions of translators and the translators’ degree of professionalism.
The degree of professionalism refers to who produces the subtitles appearing on the screen:
for instance, a professional translator, a student still in training or an amateur who subtitles for
21 fun. Abdallah (2007: 284; 2012b: 45) specifies that all of these three dimensions of quality are linked to each other; the invisible third dimension, social quality, affects process quality, which then affects the visible product quality. The purpose of the model is to not only transfer focus from product quality towards process and social quality and make the various actors involved in the subtitling process aware of the fact that quality of the end-product depends on process and social quality but, more importantly, to align the differing quality standards between the actors. By adopting this multidimensional model, it might become easier to ensure consistent translation quality as the translators would know how quality is defined in each commission which gives them the opportunity to proportion the time and effort used to complete the job (Abdallah 2007: 285). A collective definition of quality might also help monitoring whether rules, such as the quality manuals and subtitling guidelines discussed in the previous chapter, are followed and measures of quality control are carried out (Abdallah 2012b: 37).
Alternatively, Chiaro (2008: 248–249) presents a theoretical quality management tool called Total Quality Management (TQM) which could be utilised to ensure high quality.
With this tool, quality is monitored throughout an organization at every stage of production so that problems in quality are noticed and corrected before they develop any further. Moreover, Chiaro (2008: 249) reports that in TQM viewers’ opinions are “constantly sounded out and taken into serious consideration by the organizations”. TQM seeks to improve product quality through on-going improvements in response to continuous feedback from the viewers (Chiaro 2008: 249). Even though both Abdallah’s (2007) and Chiaro’s (2008) models aim at
improving and ensuring quality of the end products by drawing attention to process quality, the focal points of the models differ from each other. Whereas Abdallah (2007) emphasises the importance of a collective definition of quality by the various actors involved in
production of subtitles, Chiaro (2008) underlines viewers’ role in ensuring and improving translation quality.
O’Brien (2012), in turn, suggests that the translation industry should create a more dynamic quality evaluation model which would cater better for the changing industry and different text types. O’Brien (2012) studied quality evaluation models in active use in the sphere of professional translation and also three publicly available quality evaluation models:
the LISA QE model, the J2450 and EN15038. The study shows that most of the models evaluate translation quality from the point of view of errors and apply a “one size fits all”
approach to all kinds of translation (O’Brien 2012). O’Brien (2012: 55) notes that the current models do not pay enough attention to variables, such as content type, communicative
22 function, end user requirements, context or who or what creates the translation. Thus, O’Brien (2012: 65–74) introduces a more dynamic quality evaluation model which consists of eight different quality evaluation methods. The evaluator of the translation should decide which method is the most appropriate one for a particular translation after identifying the
communication channel and the content profile and having rated the content according to utility, time and sentiment. In identifying the communication channel the evaluator should consider who is producing the content and to whom; is the content of the text directed from business to consumer, from business to business or from consumer to consumer. In
identifying the content profile the evaluator considers what kind of content has been produced, for instance, whether the content is marketing material, website content, online help, or training material. Finally, the evaluator should rate the text for utility, time and sentiment. Even though O’Brien’s (2012) model seems to agree with Abdallah’s (2007) view in that it defines quality according to each translation commission and takes into
consideration contextual factors as well, it seems rather complex and raises the question whether companies have the time and resources to put a model like this into operation and whether it would be suitable for quality assessment of subtitles.
In addition to these rather theoretical approaches to translation quality assurance, James (2001) introduces a more practical form of quality control. He calls this “preview”
which refers both to the preliminary reading of the subtitled text and to the preliminary viewing of the subtitled programme. On the one hand, preliminary reading of the subtitled text means that a person, other than the translators, reads the subtitles without the distraction of the soundtrack and the screen, which gives the “evaluator” a chance to focus more intently on the linguistic contents of the subtitles (James 2001: 153). In the preliminary reading of the subtitled text, the “evaluator” is more able to detect, for example, instances of ambiguity, grammatical or spelling mistakes and inconsistency of terms. On the other hand, preliminary viewing of the subtitled programme means the “evaluator” familiarises himself with the contents of the programme as well as the subtitles. This gives the “evaluator” a chance to detect oversights which may result in obstructing the natural flow of the subtitled programme (James 2001: 153). These two forms of quality control together are more likely to produce a satisfactory outcome than no quality control at all. What is important with these two forms is that both the preliminary reading of the subtitled text and the preliminary viewing of the subtitled programme should be conducted by someone other than translator. As Chiaro (2008:
247) points out, translators do not usually have enough psychological distance to be able to judge the quality of their own work. When others review the end product, quality can be more
