A project-based approach to subtitler training

A project-based approach to subtitler training

Agnieszka Szarkowska

University of Warsaw, Poland


In this article I describe how a project-based approach can be implemented when training subtitlers. I present collaborative project-based work carried out at the Institute of Applied Linguistics, University of Warsaw, as part of an optional module on subtitling in the MA Programme in Translation. The project was conducted in cooperation with the Polish National Film Archive. As part of the project, students had to subtitle into English some Polish pre-World War II films. I describe the project, the structure and content of the course, the working methods, the assessment and the students’ feedback. Finally, I discuss the advantages and disadvantages of a project-based approach compared to traditional subtitling teaching models within the higher education framework.

Keywords: subtitling, subtitler training, translator training, project-based approach, social constructivism

1.    Project-based approach to translator training

In recent years, student-centred education models such as situated learning and project-based approaches (Corrius, De Marco, & Espasa, 2016; González-Davies & Enríquez-Raído, 2016; Kiraly, 2000, 2005, 2012b) have gained recognition among many translator trainers, paving the way for their application in other related areas, such as audiovisual translation (AVT) and media accessibility. In this article, following Greco (2018) and Romero-Fresco (2018), I adopt a wider view of media accessibility, which departs from the traditionally conceived notion of accessibility as serving solely people with sensory impairments and leads towards a more encompassing definition, including people with no impairments who need access to audiovisual products – for instance, owing to a language barrier. Thus, accessibility is understood here as “making an audiovisual programme available to people that otherwise could not have access to it” (Díaz Cintas, 2005, p. 4). As stated by Szarkowska, Díaz Cintas and Gerber-Morón (in press), “such conceptualisation of the term ‘media accessibility’ encompasses both traditional types of AVT like dubbing, voiceover or interlingual subtitling, and more ‘typical’ media access services”. This conceptualization therefore becomes a common denominator underpinning these practices (Díaz Cintas, 2005).

For decades, translator training was dominated by the traditional teacher-centred approach. In the traditional ‘transmissionist’ approach, the teacher is metaphorically seen as “a large vessel filled with knowledge that is doled out to the students — the relatively empty vessels” (Kiraly, 2003b, p. 28). This approach is sometimes referred to as “the teacher-centred chalk-and-talk technique” (Kiraly, 2012a, p. 85) and “who’ll take the next sentence (WTNS)” approach (Kiraly, 2005, p. 1100), reflecting typical classroom practices. The traditional approach is largely curriculum-based as it is designed to foster the acquisition of particular translation skills through short tasks and brief exercises. In contrast, the social constructivist approach to education departs from teacher-centred and leans towards learning-centred education. In Kiraly’s social constructivist approach, the teacher’s role is to “create learning situations within the institution that can provide students with authentic experience” (Kiraly, 2003b, p. 30), emulating real-life translation commissions. Within this framework, learning often takes place through situated projects.

The core of project-based education is learning by doing. Projects are understood as “multicompetence assignments that enable the students to engage in pedagogic and professional activities and tasks and work together towards an end product” (González Davies, 2004, p. 28). Within this framework, learners are provided with an opportunity to work on authentic, real-life assignments (Corrius et al., 2016; Kiraly, 2012a). An authentic assignment is understood here as “the collaborative undertaking of complete translation projects for real clients” (Kiraly, 2005, p. 1102). The project constitutes a key activity in the educational process, “with the students’ main focus being on tackling, experiencing and learning about the translator’s profession through real work” (Kiraly, 2012a, p. 84). Projects are thus seen as crucial to contributing to the development of translator competence.

One of the fundamental tenets of the social constructivist approach to education is the empowerment of students in their learning process, mainly through collaborative work (Kiraly, 2005, p. 1102). The concept of empowerment means that learners are invited to construct their own knowledge and to assume responsibility for their own learning. Noting that “true collaborative learning does not mean simply dividing up the work on a task, a mere division of labour” (p. 36), Kiraly (2000) emphasises that it is through interactive collaborative work that learners better internalize knowledge and learn how to learn. Teamwork and collaboration are seen as crucial components of the process, where learners share responsibilities and communicate to solve problems.

Among the advantages of the project-based approach to education Kiraly (2005) lists its highly motivating value, improving students’ self-confidence and preparing them better for market requirements. Yet, despite a recent rise in the popularity of the project-based approach to translator education, “relatively few attempts have been made to provide empirical evidence to actually evaluate the anticipated strengths and effectiveness of the proposed methods” (Li, Zhang, & He, 2015, p. 2). Kiraly (2012a) himself admits that “it is clear that much work is yet to be done to establish the viability of the approach beyond the scope of my own classes” (p. 93).

In this article I address the question of whether project-based collaborative learning, drawing on elements of Kiraly’s social constructive approach to the training of translators (2000, 2003b, 2005), can be successfully applied to the training of subtitlers in higher education institutions. Starting with an overview of the literature on project-based subtitler education, I move on to describe a subtitling course at my own university, including the course structure, workflow, assessment and students’ feedback. Finally, I discuss the advantages and disadvantages of this approach and suggest areas for improvement. Similarly to Kiraly (2005) and Kiraly (2012b), this article is mainly written in the first person, focusing on the perspective of the tutor, combining elements of an experiential report with those of a research article.

2.    Project-based approach to subtitling training

Audiovisual translator education has come a long way since the first course on AVT offered at the postgraduate level by the University of Lille back in 1983 (Pérez González, 2003, p. 275). Following rapid technological developments and dynamic changes in the AVT industry, we have witnessed a steep rise in the number of AVT courses. As in the case of other types of translation, the social constructivist approach to translator education has also made inroads into AVT (Corrius et al., 2016; Kiraly, 2005, 2012b) and media accessibility (Chmiel, Mazur, & Vercauteren, 2019).

In one of the earliest accounts of project-based subtitler training, Brondeel (1994) describes the approach adopted in the Provinciale Hogeschool voor Vertalers en Tolken in Ghent, which focused largely on establishing “direct links with the world of professional subtitling” (p. 26). First, experienced professional TV subtitlers were invited to give introductory talks on subtitling, and later to evaluate the students’ output. After completing a subtitling course, the students were offered a chance to subtitle foreign films free of charge for the annual Ghent Film Festival. They were provided with subtitling equipment and viewed this as “a unique opportunity [...] to spend their [...] compulsory on-the-job-training period in a live subtitling experiment” (Brondeel, 1994, p. 27). For the project, the students worked in pairs to subtitle a film using “videotaped copies” (p. 27). Then, with their computers, they went to theatres and projected their own subtitles live, as budgetary limitations did not allow for burning subtitles onto copies of films. Given the limited festival budget, professional subtitlers were not invited to cooperate. As a result, the viewers were offered either the students’ translations or no translations at all, which prompted Brondeel to express “the reassuring thought that the students are not undercutting their future colleagues” (p. 27).

In what is probably the best-known account of a project-based approach to subtitler training, Kiraly (2005) describes a subtitling project he carried out with his students – despite the fact that, as Rundle (2008) put it, “neither Kiraly (by his own admission), nor his students or their client knew anything about subtitling” (p. 100). The project was carried out in the winter term of 2004–2005 with a group of 14 advanced students of Translation. Prompted by a request from a small production company in Germany to subtitle a German documentary into English, Kiraly agreed to take on the project with his students. Another tutor, who was familiar with the fundamentals of subtitling and with the freeware subtitling software Subtitle Workshop, introduced Kiraly and his students to subtitling during two 90-minute hands-on sessions. Then the group was divided into pairs, who were to prepare their part of the subtitles and review them. The full film was then reviewed by the entire group. The team had a generous deadline of 16 weeks in which to complete their work and agreed with the company on technical issues, such as the required subtitle file format and the maximum number of characters per line (32). The students then got down to work and discussed their problems and experiences in class. The tutor’s role, as the only native speaker of English in the class, was to provide assistance related to “vocabulary choice, idiomatic usage, stylistic infelicities and the like” (Kiraly, 2005, p. 1107). Halfway through the course, the group received feedback from a professional subtitler who visited the class. Eventually, the project “became the group’s raison d’être” and “developed a life of its own” (Kiraly, 2005, p. 1107). At the end of the course, it turned out that the text had to be condensed from the previously agreed 32 characters per line to 26, which meant revising 250 subtitles that had already been completed. As a result of technical issues, it was also necessary to enter the subtitles manually in the company’s software. Ten students volunteered to travel 130 kilometres to the company’s office, where they learnt to use the company’s film-editing software. Kiraly does not state who covered the cost of the trip; neither does he discuss whether and how the assessment for the course was undertaken, which I assume had to be done as the project was conducted within the framework of a higher-education setting at the University of Mainz.

A few years later, this time working within the institutional bounds of a different higher-education institution – the École supérieure d'interprètes et de traducteurs (ESIT) in Paris – Kiraly (2012b) supervised another subtitling project. Similarly to the previous project (Kiraly, 2005), a group of seven students undertook authentic translation work (this time from French into English) for a film production company, working on a short film about the last days of the life of Marcel Proust. The project spanned the period from mid-October until early December 2011, with a total of five three-hour sessions as contact time. As reported by Kiraly (2012b, p. 10), the students had to learn three major aspects as part of the course:

1)     the rudiments of functional translation on the basis of the Interpretive Theory approach as specified by the ESIT curriculum;

2)     the norms and basic techniques of subtitling, and

3)     how to create and modify subtitles using a subtitling software program.

Unlike in his previous paper, this time it was not reported what software (professional or freeware) the students were working on and whether it was made available to the students through ESIT. In the first session, the students were introduced to the idea of working on the project and received a copy of the screenplay. Kiraly (2012b) reports that “[t]hat first class comprised a lively discussion about how to go about the project, with me sharing what I had learned from numerous authentic projects” (p. 10). The second session focused on norms: Kiraly distributed a number of texts to the students and an MS PowerPoint presentation related to translation norms, which the students were to study at home. They also worked on their translations of the screenplay and decided to translate another short film to practise working on subtitling (as they had not yet received the video file with the film they were supposed to translate). In the third session, the group spent part of the time learning how to use the subtitling program and revising the subtitles of the other short film “in terms of punctuation, chunking, and timing” and aspects of translation (Kiraly 2012b, p. 12). Session 3 is the only one in which the students were learning together how to use the subtitling program. Then they divided the work between themselves and started working autonomously, uploading their subtitles to the educational platform used at ESIT. The overall structure of the course consisted of initial stages, which focused on establishing contact and “instructional moments” (Kiraly 2012b, p. 11); further stages, in which the students worked outside of the class; and two final class sessions, during which their subtitles were completed and prepared to be sent off to the client.

Interestingly, because “subtitling work was not well suited to the university’s examination requirements (500 words to be translated on paper in 90 minutes)”, the group received a completely different non-subtitling translation task on sustainable tourism to translate as their summative assessment for the course. This reveals potential problems with fitting project-based work into a university curriculum and with assessing the students’ work created during such a project. On the one hand, as noted by Kiraly (2012b), at the end of the course his voice as a tutor “had clearly become one among equals” (p. 14) and the students had found their own voice. Yet, on the other hand, bound by the strict university regulations on assessment, the tutor had to step out of his current role and revert to the traditional tutor role in the examination session, where the students were assessed on work that was not based on the subtitling project they had done.

Regarding the quality of the subtitled products created in these subtitling projects, Kiraly (2005) argues that “the quality of the final subtitles was clear evidence of the competence we had developed as a team” (p. 1108) and that “the subtitles were submitted on time to the full satisfaction of our discerning client. […] From the client’s perspective, the outcome of the group’s project work was both competent and professional” (Kiraly, 2012b, p. 15). However, Rundle (2008) expresses reservations regarding the quality of such project output, given the lack of subtitling experience and professional subtitling software (in Kiraly 2005):

So although this was a real job with a real client, it was not done to professional    standards using professional methods (though they did benefit from the advice of a  professional subtitler), but was instead carried out to the best of their collective abilities and making the best of the inappropriate tools that their client had supplied (p. 100)

Such a critical statement on Rundle’s part concerning the quality of subtitles in Kiraly’s (2005) project may be entirely unfounded. Yet, in neither of the two papers on subtitling projects does Kiraly (2005, 2012b) mention some key technical aspects of subtitling, particularly those related to spotting – for instance, shot changes, chaining or reading speed – which are important indicators of quality in professional subtitling. Compared to a professional subtitling program, working with freeware software may result in less precision and, in consequence, in a lower-quality output. Admittedly, in his work Kiraly focuses primarily on the emergence of translator competence, not translation competence (Kiraly, 2012b), which may explain the omission of aspects related to spotting, software and other technical issues. It is also worth noting that Kiraly’s projects (2005, 2012b) were not courses in subtitling per se, but rather general translation courses where subtitling was simply used as a translation method. However, in this article it is my contention that the overall quality of subtitling, including spotting, is of primary importance and needs to be considered in any project-based subtitling course.

For his part, Rundle (2008) reports on the Subtitle Project developed at the University of Bologna in Forlì. The goal of the project was for its participants to learn about the subtitling industry in Italy, given that many students were writing their master’s dissertations with a view to becoming subtitlers. Another goal was to provide students with an opportunity to work as a research team. The project meant to contribute to the development of knowledge about the newly emerging subtitling industry in Italy, which has traditionally been a dubbing country. In this team-based research project, a group of students conducted interviews with subtitling professionals in Italy. As part of their work placements, the students worked for Cineteca di Bologna, an arthouse cinema and film archive which had a policy of subtitling – rather than dubbing – its films. The university agreed with the Cineteca to provide five students per year to work on the annual Human Rights Nights Film Festival. In the first editions of the project, each student first subtitled four feature films for a festival and then projected the subtitles live in the cinema, using “pre-formatted PowerPoint slides” (Rundle, 2008, p. 99). The students working for the Cineteca could later use the materials for their master’s dissertation if they wished. The university later entered into similar agreements with other film festivals, providing more work-placement opportunities for its students. Rundle (2008) argues that the project empowered students in their role as novice subtitlers and as AVT researchers. It is also important that project results are openly available on a dedicated website (http://subtitleproject.agregat.net/). In Rundle’s view, the Subtitle Project enhanced the AVT teaching practices by building on connections with the industry. The project also received recognition from the university, which resulted in the introduction of a module on subtitling at the second-level degree in Translation. Thanks to the insights gained through the project, the teaching of subtitling now prioritizes “perfect synchronisation in subtitle delivery” (Rundle, 2008, p. 107) rather than the perfect accuracy of the translation.

Situated at the other end of the spectrum from a project-based approach to subtitler training is a classic, curriculum-based approach, where students are first taught theoretical aspects of subtitling and then they move on to hands-on work by completing different tasks aimed at developing different subtitling skills, such as spotting and text condensation. Based on this traditional approach to subtitler training, Díaz-Cintas (2008) enumerates the necessary components of a subtitling course: condensing dialogue, matching the register to different speakers, striking the right balance between the written to oral transfer and developing an awareness of the cultural and linguistic issues characteristic of subtitling. He claims that “in the initial sessions of the course students should be given a general introduction to subtitling, placing this practice within the world of translation in general and distinguishing it from other forms of audiovisual translation” (p. 92). In his view, the best progression in a subtitling course is to start with such a general overview of subtitling, to move on to technical and then linguistic dimensions, and to finish off with “a detailed overview of the profession and the subtitling market” (p. 92). Díaz-Cintas (2008) believes that students should be exposed to hands-on subtitling tasks at an early stage, which will enable them to move from “the traditional declarative knowledge-driven course (know-what) to a more skills, procedural knowledge-based approach (know-how)” (p. 92). This, in his view, will enhance students’ performance, foster responsibility, promote autonomous work and provide a sense of empowerment.

The traditional approach to translator education and the project-based approach may be complementary and, as shown in this article, can in fact be combined in one course. As noted by Kelly (2005), it may be “simply a question of level/stage of training, that is of sequencing” (p. 116). In other words, the traditional task-based approach may be more appropriate at the early stages of translator education, with the project-based education following at a later, more advanced stage. In fact, this was already shown by both Brondeel (1994) and Kiraly (2005, 2012b), where the learners were first introduced to the fundamentals of subtitling before undertaking collaborative work and embarking on authentic projects.

The subtitling course described in this article combined the two approaches outlined above. In line with recommendations by Díaz-Cintas (2008) and similarly to Kiraly (2005, 2012b), we started in a traditional way by mastering the fundamental subtitling skills, both the theory and the hands-on practice with subtitling software. Unlike Kiraly, however, we spent much more time (the first few classes together lasting for about a month) on mastering key subtitling skills, working with a professional subtitling programme (EZTitles) available to students at the university. The key subtitling skills included both linguistic aspects such as text condensation, line breaks and chunking and technical aspects such as familiarity with the software, spotting, shot changes and reading speed. In other words, our initial focus was largely on acquiring translation (i.e., subtitling) competence. Only after learning the subtitling fundamentals did we proceed to work on an authentic project and to acquire translator competence (for more on translation and translator competence, see Kiraly 2003b and 2012b).

3.    Subtitling pre-World War II films for the Polish National Film Archive

This project was carried out at the Institute of Applied Linguistics (Instytut Lingwistyki Stosowanej, ILS), University of Warsaw, the oldest translator training institution in Poland. The institute offers both undergraduate (BA) and postgraduate (MA) courses in Translation. The Subtitling course described here was offered as an elective face-to-face course in the two-year MA programme taken by students in their first year. It was worth three ECTS credits and consisted of 30 contact hours, spread once a week over the course of the winter term and amounting to 15 meetings of 90 minutes each. The cohort consisted of 20 students who had Polish as their mother tongue and English as their B or C language.

Coincidentally, before the beginning of the academic year, I was approached by the Polish National Film Archive (Filmoteka Narodowa, FN) with a request to subtitle into English some Polish films from their pre-World War II collection. The Polish FN is a state cultural institution that aims to protect the national heritage in cinematography and the dissemination of film culture. At that time, they were working on the NITROFILM Project, which involved the digitization and restoration of 43 pre-war feature films. The films were to be screened at a retrospective film festival. Seven films were selected for the project: ABC miłości (1935), Ada, to nie wypada (1936), Dwie Joasie (1935), Dziewczyna szuka miłości (1937), Przez łzy do szczęścia (1939), U kresu drogi (1939) and Ludzie Wisły (1938).

3.1    Structure of the course

Similarly to other project-based courses in subtitling (Brondeel, 1994; Kiraly, 2005, 2012b), our course began with a general overview of subtitling, followed by a few hands-on sessions aimed at mastering the professional subtitling software (EZTitles). Only then, equipped with those basic skills and fundamental technological competence (EMT, 2017), were the students ready to embark on a real-life project.

To begin with, I invited a representative of the National Film Archive to come to class. He told us about the project, its aims and potential problems. He presented the NITROFILM Project and offered help if there was anything we did not understand. It needs to be noted here that the language of the film dialogues was archaic compared to modern Polish. Actors were also using highly theatrical pronunciation, which was often difficult to understand. After the first attempt at subtitling the films, we realised it was sometimes very difficult to make out what the actors were saying, which was due to both the poor quality of the sound and the old-fashioned variety of Polish. We therefore requested dialogue lists and the FN promptly provided us with them. In some cases, we also received glossaries containing explanations of some old-fashioned terms and expressions.

Apart from classwork in a computer lab, the students could also use a self-study computer room in our department where full EZTitles licences are also installed. Alternatively, they could work from home using a demo version of the software, which was fully functional up to 25 subtitles. We used Moodle as our virtual learning environment, where we stored all the course materials, such as the EZTitles manual, our subtitling settings and readings.

3.2    Project workflow

The course was conducted during the winter term and lasted from October until the end of January. We received the films on DVD in November and the deadline of mid-January for delivering subtitle files (in .stl format) was agreed on.

It was decided that the students would be working in groups of three, subtitling and reviewing up to 30 minutes of film per person. The files would then be combined into one for the entire film subtitled by the team. Finally, the files were to be verified by the tutor and sent to the client.

We started in-class work in an attempt to identify similar problems in different films, pool resources, collect feedback from everyone and help one another with technicalities. However, the first class meeting was not entirely successful because some students had not watched their films before coming to class. Other students – although they had watched the films – had not started translating yet, so again we could not discuss any problems they faced. This shows that the students had not taken responsibility for their own learning process.

Regarding the division of work, the students decided to have equal roles in their projects, with no role of project manager who would be responsible for each film as a whole. In each group, the films were divided into parts. Everyone had to spot their fragment and to translate it. Then the students were supposed to review one another’s files in groups, emulating the professional quality control process, before finally submitting them to me. In practice, as there was no project manager, some groups submitted their final subtitle files without verifying them as a whole. This led to numerous inconsistencies, particularly in terms of vocabulary and character names.

3.3    Assessment

Given that we were operating within the bounds of a higher-education institution, as the tutor I was obliged to assess my students’ performance at the end of the course and to give them grades. Corresponding with the requirements of professional subtitling, the grading criteria were based on three groups of quality aspects of subtitling: translation, timing and text. Translation included the choice of an appropriate translation strategy and text condensation; timing related to synchronisation, reading speed and shot changes, and text was about appropriate line breaks and text segmentation – both within and between subtitles, spelling and punctuation.

The deadline for submitting the final subtitling file was set two weeks before the end of the term. However, some students were late with delivering their files, which made it impossible for their groups to complete the task on time. In consequence, as the tutor, I had less than two weeks to mark the seven feature films from this project. This turned out to be a very demanding and time-consuming task. Time considerations, however, were not the trickiest issue I had to deal with.

Once I received what the students considered to be the final files, I was faced with the following choice: should I indicate the problems in a comment box and send the files back to the students to revise, or should I correct the files myself? Normally, in a traditional non-project-based course, the assessment would be done using the comment box and the subtitling files would not be revised once they had been graded. In this case, however, the files were later to be used by a real client, so we had to deliver a complete, good-quality final product. I took the decision – in what turned out to be only the first round of revisions – that I would mark the problems and ask the students to correct them, as only in this way would they learn more and receive some feedback.

By the time the students revised the files, the deadline for submitting the grades in the university system had passed, so my marking had to be done on the students’ first submissions. At this stage, students were no longer formally obliged to continue working on the project. However, some of them felt responsible for completing the project and volunteered to revise the files after the course ended. All in all, we ended up meeting outside of classes, in our free time, for the next few weeks, revising the subtitles together to the best of our ability.

3.4    Students’ feedback

After the end of the course, I was curious as to how the students had perceived the experience of working on the project and whether they thought the extra effort we all had to put into the project was worth it in the end. I therefore asked them to complete anonymous feedback forms, distributed as an online survey.[1] Below I report on their feedback, grouped into main themes using the method of inductive coding (Bazeley, 2013).

Real-life task. When considering the fact that we had worked on a real-life project instead of a traditional ‘artificial’ assignment, the students expressed an overwhelmingly positive opinion of the course. One person simply wrote: “Loved the final assignment!” Someone else said it was “a very, very good idea, especially the fact that we knew that our translations will be used later”.

Some students stressed the more challenging nature of real-life project-based assignments compared to traditional university courses, as noted in the following quote:

In my opinion it was quite difficult and demanding, yet I like the idea of creating something that has real use and will be helpful to someone.

Despite the extra challenge and work, the course was generally perceived in a positive light, as expressed by this student:

I liked it, it was challenging. I’m glad I could take part in a larger project, out of university and to gain experience in real subtitling. I really appreciate that you planned the final assignment in such an interesting way, though of course it required a lot of work.

Time. Many students stated that the project was very time-consuming and challenging – much more so than the traditional course. One person wrote: “It was great fun, but at the same time very difficult. It took me ages to translate my part. It was very demanding, but, all in all, I enjoyed it.”

Translation. Among the major translation problems the students faced were outdated vocabulary, dated forms of address, songs and rhymes. We discussed these issues in our in-class meetings, focusing on how to create the target text in a way in which it would satisfy the needs of contemporary audiences and at the same time reflect the old-fashioned nature of the language used. The students also complained about technical issues related to the film quality, such as poor image visibility or missing utterances due to the incomplete nature of the archive material.

Work management. As the time for completing the task spanned the Christmas break, some of the students complained that it was difficult for them to work on the assignment as the university was closed for Christmas. While a few of them complained that they did not have enough time to complete the task, some of them admitted that “we didn’t manage to plan the work optimally”.

Group work. Perhaps somewhat surprisingly, the largest number of complaints related to the nature of the group work. Some students noted that it was necessary for the team members to rely on one another and that this did not always work well. One person argued: “The drawback of the assignment was the necessity to work in a group and taking responsibility for other people’s work, which was quite problematic.” Another student expressed concern at the necessity to work in groups, which apparently was quite common at that time in our department:

It taught me more than all the home assignments combined. It was hard and really time-consuming but, at the same time, interesting and satisfying. However, I really don’t like working in groups because there’s not always a chance to choose someone you trust to work with and ILS forces students to work in groups constantly.

Overall, the general feeling relating to the project was overwhelmingly positive. Only one person commented on the unpaid nature of the work: “doing stuff useful for someone for free feels stupid.” Some other suggestions for improvement included shorter film fragments – for instance, up to 15 minutes instead of 30.

4.    Discussion

By adopting the project-based, learning-by-doing approach, I was hoping to provide my subtitling students with an opportunity to engage in “authentic professional practice” (González-Davies & Enríquez-Raído, 2016, p. 2). Departing from the belief that the acquisition of fundamental technological competence (EMT, 2017) is indispensable to becoming a professional subtitler, we started by learning the basics of subtitling, including the subtitling software. This was also in line with previous work (Kiraly, 2005, 2012b). Without such an introduction, the students would not have been able to work on any subtitling project. The experience reported here demonstrates the viability of applying the project-based approach to the training of subtitlers at a higher-education institution, albeit not without problems. Although the project in general was successful at many levels, a number of problematic areas emerged in the process, including role assignment, group work, time management, the quality of the final product, and assessment.

Among many hurdles I stumbled upon on the way to course completion was the fact that the students were accustomed to a trainer-centred didactic environment and a transmissionist approach (see Kiraly, 2003a) rather than a project-based approach. In the traditional model, the teacher’s amassed knowledge is passed on to learners, who need to absorb it; knowledge is “handed down from a better-stocked mind to less knowledgeable ones” (Kiraly, 2003a, p. 8). The teacher takes a privileged position and “is privy [...] to the right answers, that is, to truth, and the learners are there to find out what those answers are” (Kiraly, 2003a, p. 5). In contrast, in the project-based approach, the teacher is supposed to support the learners in their learning process, but it is ultimately the learners who need to take responsibility for their own learning process. Teachers are facilitators of knowledge whereas:

empowered learners are autonomous thinking individuals whose capacity for meaning-making is no less valuable than that of the instructor, thus the traditional      hierarchies of power which place an undue and unaccountable degree of authority in the hands of the instructor are subverted. (Varney, 2009, p. 29)

However, in my subtitling course, I had the impression that the traditional power dynamics were quite strong, for two reasons. First, most of the students perceived me from the traditional perspective and were not mentally ready for any subversion of the hierarchy. Particularly during the early stages of the course, they seemed wary of the new type of teacher–student interaction offered by the social constructivist model, which “varies radically from traditional classroom discourse” (Kiraly, 2000, p. 47), with the teacher holding the position of power. The learning autonomy offered by the social constructivist approach depends on “the development of the ability to reflect on one’s own work” and requires “reflective capacity” (Kiraly, 2000, p. 48), which in my view the students were struggling to achieve. Secondly, my traditional role in the power dynamics was further strengthened by the necessity of my marking their final performance as we were operating within the bounds of a higher-education institution. Although social constructivists have proposed alternative solutions to traditional assessment practices (Kiraly, 2000) – which include, for instance, the integration of portfolios into translator education assessment – it is the traditional summative assessment model that is still deeply entrenched in the highly regulated assessment practices at higher-education institutions. This is the reality that subtitler trainers also have to face, including Kiraly himself, who, when reporting on his subtitling projects, either did not mention assessment (Kiraly, 2005) or said he had to resort to the traditional summative assessment required by his institution (Kiraly, 2012b).

To recap the discussion so far, the project-based approach to subtitler training at a higher-education institution marks a departure from the traditional teacher-centred transmissionist approach. Yet, in terms of summative course assessment, it seems to be situated closer to the transactional approach, where the teacher “still has the final answer to the problems set” (González Davies, 2004, p. 14) rather than the transformational approach, where teachers become facilitators “in a transformational environment that favours real life professional practice and hands over the final responsibility for the translation to the students” (González Davies, 2004, p. 225) (for a detailed discussion of the approaches to translation training see González Davies (2004)).

For me as a tutor, working on the project turned out to be much more demanding than a regular course: I had considerably more film footage to mark, which later encroached on my free time as I had extra work in the following term. Compared to the traditional syllabus I normally went through with my subtitling students in a non-project-based course, spending an entire course on one project felt limiting as it did not allow the students to experience a wide range of other translation problems. As we worked with pre-World War II films, the project was not fully representative of the present market situation and the students could not face some of the problems typical of contemporary productions. On the positive side, the students were highly motivated. They also had to deal with a number of technical issues, which is a frequent occurrence in professional subtitling, so in this sense the project met the requirement of being true to life and authentic.

The project discussed here relied largely on collaborative learning in an attempt to benefit from the synergies arising from such work. As stated by González Davies (2004):

the whole is not only the sum of the parts, i.e., positive team work – in which each member of the team strives towards attaining the best collective performance – can contribute to reduce peer pressure, improve communicative and social skills, bridge linguistic and cultural diversity, and increase group cohesion, thus resulting in more effective learning. (p. 13)

Another advantage of providing subtitling students with an opportunity to work in teams is that this helps them realise the work done by many agents in the subtitling workflow, most importantly, spotter, translator, QC-er[2] and project manager. As pointed out by Díaz-Cintas (2008), subtitling is indeed“the end result of teamwork and students should learn about the different stages that are needed from the commission of the work until the broadcast, screening or distribution of the subtitled programme” (p. 102).

In our project, however, several students complained about the nature of the teamwork. Their reservations related mainly to the unreliability of other team members – it being impossible for them to have any influence over their fellow students – and the problem of assessing work completed by a few people, whose contribution was not of equal value to the team. All these problems show that group work, a key element in the social constructivist approach, is a complex process that may not be met with enthusiasm by all learners.

The question for the future is therefore how much teacher intervention should be allowed in such project work? In the social constructivist view, the teacher is supposed to be a facilitator of learning. Yet, as argued by Marco (2016), sometimes the teacher “is forced to make radical interventions in the face of obvious mistakes and the need to ensure continuity of collaboration” (p. 41). Marco (2016) stresses that the teachers’ intervention in real-life projects may be necessary, as “not all students are ready to assume an expert role” (González-Davies & Enríquez-Raído, 2016, p. 6). At the beginning of the project, I allowed the students to assume the roles of subtitlers, that is, both spotters and translators, as well as reviewers, without there being a project manager in each team. I believed that performing all the roles would be more beneficial for them as it would give them more varied experience of the different stages of the subtitling workflow. With the benefit of hindsight, however, my recommendation would be that all the roles should have been precisely allocated from the outset, with the responsibilities of each clearly defined.

Management issues in our project were related not only to the assignment of roles, but also to time, workflow and quality assessment. Kiraly (2005) argues that “The market requires that the final product be of impeccable quality” (p. 1109). In our project, however, the question was: What if the product quality is not impeccable at the end of the course? What should the tutor’s role be? As I had not posed these questions at the early stage of the course and had no contingency plan in place, feeling responsible for the final project outcome, together with a handful of students, I felt obliged to complete the project, which meant extending its original duration and assuming the role of the QC-er. The experience of working in this project has taught me to allocate much more time to the revision stage. My recommendation would also be to set the rules on how decisions will be made when the final product is ready to be sent to the client. In hindsight, perhaps it would have been better if, instead of me, a student had been the main contact with the client. This may have made the students assume more responsibility for the project and its final outcome.

The essence of project-based work is cooperation with companies, institutions or organisations from outside academia in an attempt to provide subtitling students with real-life experience. However, cooperation between a higher education institution, on the one hand, and professional translation companies or language service providers (LSPs), on the other, may be difficult at many levels. First of all, it is unlikely that many large LSPs would be willing to “outsource projects to inexperienced translators-to-be” (Bolaños García-Escribano, 2016, p. 24). Not only may they be worried about the quality of the students’ output, but the deadlines and turnaround times in the subtitling industry tend to be much shorter than the timespan of university courses. Secondly, it would be good if students were remunerated for the work they do in the project, but organising such payments within the remit of a higher education institution “would present intricate administrative challenges” (Bolaños García-Escribano, 2017, p. 16). Other issues when working with companies include confidentiality and copyright when using authentic material.

Last but not least, a recurrent topic in project-based work is the question whether students working on authentic commissions are undermining the work of professional translators. However, it is often the case in such project work conducted by university students that the companies or institutions in need of translation do not have the financial means for engaging professionals. In effect, the translation is either done by students free of charge or it would not be done at all (Brondeel, 1994). This was also the case with our project. In a sense, it was a win-win for all the stakeholders involved in the project: the Film Archive received their translation, which they otherwise would not have had, and the students had an opportunity to work on an authentic project.

5.    Conclusion

All in all, the project-based approach to translator education offers unique opportunities for learners to work on authentic assignments and gain professional experience. However, as shown in this article, when conducted within the bounds of a higher education institution, the approach also entails a number of risks and challenges, both for the tutor and for the students, particularly in terms of time, role management and the workload involved in completing the project.

As translator trainers and translation researchers, we continue to strive to pinpoint the successive stages in the development of translator competence and to find methods through which this competence can best be developed. The project-based approach presented in this article constitutes a step towards establishing a way in which translator competence is developed through the interplay of authenticity and emerging autonomy. In the words of Kiraly (2005), qualitative case studies on situated translation projects “can contribute significantly to the triangulation of claims concerning the enactive nature of cognition and learning” and in this manner can lead to a deeper understanding of the nature of translator competence.


Many thanks to Alejandro Bolaños García-Escribano for his assistance with the materials on situated learning and project-based education, to the anonymous reviewers and to the special issue editors for a thorough reading of the article and many pertinent suggestions.


Bazeley, P. (2013). Qualitative data analysis: Practical strategies. London: SAGE.

Bolaños García-Escribano, A. (2016). The pedagogical potential of cloud-based platforms: Towards an integrated model for the didactics of subtitling in translator training environments (Unpublished master’s thesis). University College London, London, United Kingdom.

Bolaños García-Escribano, A. (2017). The didactics of cloud-based audiovisual translation: Towards an integrated model for online education in audiovisual translator training. Upgrade report. Centre for Translation Studies, University College London: London, United Kingdom.

Brondeel, H. (1994). Teaching subtitling routines. Meta, 39(1), 26–33. doi:10.7202/002150ar

Chmiel, A., Mazur, I., & Vercauteren, G. (2019). Emerging competences for the emerging profession: a course design procedure for training audio describers. The Interpreter and Translator Trainer, 13(3), 326–341. doi:10.1080/1750399X.2019.1656408

Corrius, M., De Marco, M., & Espasa, E. (2016). Situated learning and situated knowledge: Gender, translating audiovisual adverts and professional responsibility. The Interpreter and Translator Trainer, 10(1), 59–75. doi:10.1080/1750399x.2016.1154343

Díaz Cintas, J. (2005). Audiovisual translation today: A question of accessibility for all. Translating Today, 4, 3–5.

Díaz-Cintas, J. (2008). Teaching and learning to subtitle in an academic environment. In J. Díaz Cintas (Ed.), The didactics of audiovisual translation (pp. 89–103). Amsterdam, The Netherlands: John Benjamins. doi:10.1075/btl.77.10dia

EMT. (2017). European Master's in Translation: Competence framework 2017 [Report]. Retrieved from https://ec.europa.eu/info/sites/info/files/emt_competence_fwk_2017_en_web.pdf

González Davies, M. (2004). Multiple voices in the translation classroom: Activities, tasks and projects. Amsterdam, The Netherlands: John Benjamins. doi: 10.1075/btl.54

González-Davies, M., & Enríquez-Raído, V. (2016). Situated learning in translator and interpreter training: Bridging research and good practice. The Interpreter and Translator Trainer, 10(1), 1–11. doi:10.1080/1750399x.2016.1154339

Greco, G. M. (2018). The nature of accessibility studies. Journal of Audiovisual Translation, 1(1), 205–232. Retrieved from https://www.jatjournal.org/index.php/jat/article/view/51/10

Kelly, D. (2005). A handbook for translator trainers: A guide to reflective practice. Manchester, England: St. Jerome.

Kiraly, D. (2000). A social constructivist approach to translator education: Empowerment from theory to practice. London, United Kingdom: Routledge. doi:10.4324/9781315760186.

Kiraly, D. (2003a). From instruction to collaborative construction: A passing fad or the promise of a paradigm shift in translator education? In B. J. Baer & G. S. Koby (Eds.), Beyond the Ivory Tower: Rethinking translation pedagogy (pp. 3–27). Amsterdam, The Netherlands: John Benjamins. doi: 10.1075/ata.xii.03kir

Kiraly, D. (2003b). From teacher-centred to learning-centred classrooms in translator education: Control, chaos or collaboration? In A. Pym, C. Fallada, J. R. Biau, & J. Orenstein (Eds.), Innovation and e-learning in translator training: Reports on online Symposia (pp. 27–31). Tarragona, Spain: Universitat Rovira I Virgili.

Kiraly, D. (2005). Project-based learning: A case for situated translation. Meta, 50(4), 1098–1111. doi:10.7202/012063ar

Kiraly, D. (2012a). Growing a project-based translation pedagogy: A fractal perspective. Meta, 57(1), 82–95. doi:10.7202/1012742ar

Kiraly, D. (2012b). Skopos theory goes to Paris: Purposeful translation and emergent translation projects. mTm. A Translation Journal, 4, 119–244.

Li, D., Zhang, C., & He, Y. (2015). Project-based learning in teaching translation: Students’ perceptions. The Interpreter and Translator Trainer, 9(1), 1–19. doi:10.1080/1750399x.2015.1010357

Marco, J. (2016). On the margins of the profession: The work placement as a site for the literary translator trainee’s legitimate peripheral participation. The Interpreter and Translator Trainer, 10(1), 29–43. doi:10.1080/1750399x.2016.1154341

Pérez González, L. (Ed.). (2003). Speaking in tongues: Language across contexts and users. Valencia: University of Valencia.

Romero-Fresco, P. (2018). In support of a wide notion of media accessibility: Access to content and access to creation. Journal of Audiovisual Translation, 1(1), 187–204.

Rundle, C. (2008). The subtitle project: A vocational education initiative. The Interpreter and Translator Trainer, 2(1), 93–114. doi:10.1080/1750399x.2008.10798768

Szarkowska, A., Díaz Cintas, J., & Gerber-Morón, O. (in press). Quality is in the eye of the stakeholders: What do professional subtitlers and viewers think about subtitling? Universal Access in the Information Society.

Varney, J. (2009). From hermeneutics to the translation classroom: A social constructivist approach to effective learning. Translation & Interpreting, 1(1), 27–43.


[1]     As the course was conducted in English, the survey was also in English. All the students’ comments are quoted verbatim as they were written in English.

[2]     QC stands for quality control.