RSS feed

Professional identity and training of translators in the context of globalisation: The example of subtitling

Dionysios Kapsaskis, Roehampton University, London, UK


This paper argues that the pressures placed on translators working in a globalised context amount to a professional identity crisis which should also concern academic translator training programmes. Focusing on the example of subtitling, the paper first shows that the role of subtitlers within the subtitling workflow has been fragmented because of structural changes in the industry, such as outsourcing and the use of template files. The issue of translation quality is key in this context. An analysis of the French and Greek subtitles of an episode of the The Sopranos, which were based on a template file, will suggest that quality may suffer as result of the centralisation of the subtitling process, which has become less reliant on the specialist competence of subtitlers. Broadening the discussion, it will be argued that translators in general are often called to play an extremely flexible and less specialist role in the industry. A pertinent response by translator training courses would be to shift the emphasis from specialisation to the development of a core translation competence, as a means of redefining and empowering the professional identity of translators.

Translator identity, translator training, quality, translation competences, subtitling, globalisation.

The relationship between translation and globalisation has attracted plenty of attention in the past ten years from professional translators and scholars. Michael Cronin (2003) notably set the terms of this discussion by showing how translation has historically been a powerful tool for globalisation and how our perception of translation is being transformed as a result of the globalising process. The association between globalisation and translation should be evident in many ways: within various definitions of economic and cultural globalisation there are common conceptual threads that lead more or less directly to translation. In his more recent study on the same topic, George Ho (2008: 55-56) quotes various definitions of globalisation, including the “interdependence of countries and cultures,” the “increasing volume of cross-border communications,” the “global exchanges of people, commodities and ideas” as well as the “diffusion of technology,” all of which point to the relevance of translation within an increasingly globalising context. Yet what is readily accepted by translators and translation scholars seems to be less obvious to globalisation theorists. As Cronin (2003: 64) has pointed out,

Though, in view of the spread of its interests and the cogency of its concerns, the discipline of translation would seem to be powerfully equipped to have a substantial impact on intellectual debates centred around globalization, its impact on other academic disciplines investigating the topic, with certain notable exceptions, has been limited.

This argument was recently reiterated by Bielsa and Bassnett (2009: 18), who suggest that current theories of globalisation need to be interrogated with regard to “their lack of attention towards translation as a key process in the development of global connectedness.” According to the authors, present approaches to globalisation (those of Lash and Urry, and Appadurai are especially discussed) place emphasis on the mobility of goods and information and fail to account for “the productive processes that shape and make current developments possible” (ibid: 23). As one of these processes, translation reverts the focus to the spatio-temporal, material and mediated nature of communication, which theories of uninterrupted mobility and flows tend to obscure. The authors also note that theories that do examine production forces in the globalising context, such as Castells’ theory of network society, still tend to overlook translation, by assuming the advent of universal monolingualism—for example a lingua franca—or of the global language of technology (ibid: 22-26). This case of theoretical blindness is indicative of translation’s predicament in the age of globalisation. It is today possible to argue that not only the translator, but translation itself is becoming invisible (Venuti, 1995). At this socio-economic juncture, where production processes are expected to be time and cost-effective at an unprecedented scale, translation is under pressure to conform to the rules of instantaneity, transparency and invisibility (Bielsa and Bassnett, 2009: 28-29).

In this paper, I am concerned with the consequences of market globalisation on translation as a professional practice and, secondly, on translator training. I will begin by focusing on the example of audiovisual translation and the way changes in industry have had a profound influence on the role and identity of subtitlers over the past few years. Subsequently, I will discuss how quality in subtitling has been a casualty of these changes, specifically as a result of outsourcing and the introduction of template subtitling files. Finally, I shall be asking how these developments could affect decisions relating to translator training at university level, insofar as issues of specialisation, competence and skills are concerned. The overall purpose of the paper is to assess critically the perceived weakening of the status of professional translators in a globalised setting and to examine ways in which University translator training programmes can respond to this situation.

The example of subtitling

The advent of new media, the need to manage centrally an increasing amount of information and the ever shorter deadlines are some of the factors accounting for the fragmentation of the translation process. Within the past ten to fifteen years, a number of new skills and sub-competences have emerged. In his manual for new translators, Daniel Gouadec (2007: 120) mentions some of them:

The ‘new’ translator must in fact be ready […to become] an information management expert, technician, terminologist, phraseologist, translator, adapter, proof-reader, reviser, quality control expert, post-editor, editor, graphic design expert and Web page designer, technical writer, Web site designer, Web page integrator, file manager, macro-command writer and in some cases IT specialist, all rolled into one.

Gouadec complements this list with further roles, such as those of the subtitler, overtitler and localiser. Professional translators would readily add further entries. While the length and diversity of this list serves as a testament to the key role played by linguistic mediation in a globalised setting, it also hints at a radical fracture that has occurred within translation as a profession. To possess many of these skills and to be ready to develop even more following market demand is a formidable challenge, of course. At the same time, such compartmentalisation of knowledge and expertise leaves the translator disoriented, when it comes to shaping a professional identity.

This is not only a theoretical issue, it also has practical repercussions. To give an example of the new market reality and thus move on to audiovisual translation, the Media Consulting Group report (2007: 8) on the dubbing and subtitling industry stresses that:

[I]nternationalization of the technical industries activity is now the key factor conditioning development of sector businesses. This is mainly due to the development of international customers selling their programs in several countries. Based on discussions with surveyed companies, economies of scale, notably as regards subtitling, will only be possible if the volume of work handled is increased.

The increase in volume, however, does not necessarily mean a better deal for professional translators or indeed for the end clients (the public). As a more recent report by Yves Gambier (2009) on behalf of the European Master’s in Translation expert group finds:

Professions in multilingual communication have developed rapidly over the last 20 years, both under pressure from technological changes and as a result of the transformation of markets, linked to globalisation, outsourcing and flexibility. The consequences of this twofold development are felt in the division of work and in the relationships between translation volume, quality control and price.

By raising questions of quality and, especially, by pointing out differences in the division of work brought about by the process of globalisation, Gambier provides a crucial lead for assessing the professional status of audiovisual translators. With regard to interlingual subtitling for DVD, changes in the division of work have specifically affected the subtitling workflow as well as the subtitler’s identity in two interrelated ways. On the one hand, these changes are of a temporal order, in the sense that they affect the time and the sequence of actions through which audiovisual translation (AVT) is carried out. On the other hand, these changes are of a spatial order, as they refer to the place in which these actions are accomplished. I will discuss briefly these types of changes before moving on to the quality issues that are raised as a result.

On a temporal level, the new division of work is expressed through the introduction of template subtitling files and the ensuing restructuring of the subtitling process. This development is well known, but an attempt will be made here to interrogate it here with a view to spelling out its implications for the profession. Until the early days of DVD subtitling, the subtitling process consisted in two major tasks: a technical task, namely the timing of the subtitles, which involved no linguistic intervention, and the translation, directly from the audiovisual material. Often the same person would carry out both tasks, thus fully “originating” a subtitling file. Georgakopoulou (2006) describes this process in some detail and points out that, thanks to the development of dedicated subtitling software “subtitlers could […] spot the film themselves and then write their translations so as to fit the time slots they had spotted. They could also simulate their subtitles on their workstations and alter the wording and the timings as necessary.” As Díaz Cintas and Remael (2007: 30) point out, this is an ideal situation. This is because translators have the chance to make informed decisions about reading speeds and the segmentation of the dialogue into subtitles, in such a way as to optimise the translation of humour, hesitation and surprise. They can also decide whether some subtitles need to be omitted due to redundancy or whether any editing interventions are required by the stylistics of their native language and the specificities of their target audience. However, in response to the need for centralised management of multilingual projects, translation companies have brought back the distinction between an initial timing phase and a subsequent translating phase, with two important differences. Firstly, that the timing phase involves extensive linguistic processing of the source text (the audible dialogue plus any text on screen), leading to the creation of a template subtitling file in the source language (SL) that observes centrally specified editing rules and technical conventions; and, secondly, that the two phases are structurally rigid, in the sense that translators working with the template file may no longer alter the timings when they would rather do so, and are generally compelled to follow the editing choices by the template file creator. The template file (or “master file”) became the internationalised version of the audiovisual source text. As it shall be argued in more specific terms below, the compulsory use of that version has drastically restricted the translator’s role within the process of subtitling and has had visible effects on the quality of the localised product.

There has been some literature on the introduction of template files (Georgakopoulou, 2006; O’Hagan, 2007: 162; Pym, 2010: 126), but its consequences for the professional identity and self-perception of audiovisual translators have not been spelled out. Arguably, one of the immediate effects of this development is the weakening of the subtitler’s status. Whether a monolingual template file creator or a text translator, the subtitler becomes less of a specialist, a more expendable partner and a more vulnerable one to financial pressure. To use a phrase by Gottlieb (1996: 284), interlingual subtitling as a specialisation risks becoming, “just a matter of translating some lines from a script and shaping them into neat blocks.”

Division of work happens on a spatial level too. Under pressure for minimising costs, translation companies, following other sectors, have embarked on projects of outsourcing. As far as subtitle providers are concerned, this process was facilitated by the use of the template, easily emailed to translators living in the so-called “territories” (Carroll, 2004). Outsourcing has been acutely felt by a generation of London-based subtitlers in the early 2000s. It affected almost all specialisations, including template file creation and, occasionally, simulation and proofreading. Above all, it affected interlingual translators, who suddenly saw most of the available work being assigned to new translators overseas. While this was undoubtedly an effective move for business, translators, including those recruited overseas, had only to lose from this new state of the affairs. Professionals who used to work close to the large subtitling companies saw their rates fall dramatically, as they had to compete with low-waged outsourced translators, who were frequently inexperienced and untrained.1 This situation continues today, raising complex questions of professional viability and satisfaction for audiovisual translators, but also issues of quality that concern the end user of the localised product, namely, the audience.

The argument of quality

Advocates of spatio-temporal restructuring contend that there are qualitative benefits in it, in addition to cost effectiveness and the streamlining of management. In the case of subtitling, for instance, they stress that through the use of template files, risks of mishearing of the source dialogue are minimised and consistency of research lists and overall technical approach—all of which are important in multilingual projects—is ensured across all target languages (see Georgakopoulou, 2009). These are true advantages, but they are at least partly offset by the risks involved in the twin practice of outsourcing translation work through the use of template files. Some of these risks are highlighted in an article about the British subtitling industry by Nakata Steffensen (2007: 18-19), who describes in alarming terms the profile of a translator that globalised translation companies are looking to work with:

Much of the work previously carried out by British-based freelancers is […] increasingly being sourced in lower-cost countries, carried out by unpaid student interns or by new entrants to the industry willing or able to work for very low pay. […] The simultaneous centralisation and decentralisation of the industry on a global scale also mean that the UK’s position as a centre in this part of the global media industry is under threat. Operations are being centralised in Los Angeles, and subtitles are increasingly sourced in lower-cost countries.

Nakata Steffensen’s point indicates how difficult it is to sustain the quality argument when faced with inflexible market priorities. To be sure, the concept of quality in AVT is itself being redefined so as to refer to the localised product as a whole (in this case, the DVD) rather than the linguistic content only (in this case, the subtitles). Indeed there is no doubt that a DVD which allows choice between multiple subtitle tracks of good production standards is of higher quality than a DVD without them. Minako O’Hagan (2005: 10; 2007: 162) has pointed out that the change of approach from traditional screen translation to homogenised localisation involves a paradigm shift which brings with it new quality criteria, such as the ones described above. In a more conventional vein, Gambier (2006: 6) stresses that quality

[…] is the result of a collective and joint effort, although many translators think that they hold the monopoly on quality. Producers, distributors and viewers are also involved, their expectations and demands not necessarily coinciding with the translators’ since they do not always stem from language considerations, nor are they based on the written language of the subtitlers.

However, none of these arguments can be used to justify any concessions as regards the range and kind of services that subtitlers should provide in order to ensure overall quality. Subtitlers may not prescribe all quality criteria regarding a localised product (although their opinion should still count) but they are best placed tounderstand what good subtitling is about. What is more, in no case should the points made above be used to support the extensive use of non-professional or non-skilled translators. On the contrary, one would think that for the new paradigm to work, reliance on experienced subtitlers would be essential. And yet professional subtitlers are being sidelined and the quality of subtitles suffers. So much is indicated not only by discussions in AVT conferences, but by broadcasters and distributors too.2

Even representatives of major service providers concede that quality is one of the prime casualties of globalisation. In an influential paper entitled “Subtitling: Changing Standards for New Media?,” Mary Carroll of Titelbild states: “The price wars are fierce, the time-to-market short, the fears of piracy rampant. The aim of the subtitling companies is to deliver the best multilingual subtitles possible under the given circumstances” (2004; my emphasis.) To paraphrase this statement, quality subtitling must be enframed within quantitative imperatives, if subtitling companies are to survive at all in this war. In the same paper, Carroll argues that template files were introduced by companies in an attempt to maintain quality while responding to the pressure for “rapid turnaround times and depressed prices.” Nonetheless, she adds the following proviso:

Such a template can make sense if it is thoroughly researched and well-timed, especially if subtitlers are free to use it as an aid but are not compelled to force their translation, regardless of its structure, into its mold. However, the rigidity of such files can result in poor subtitling with little adherence to now common standards of good subtitling practice […]

Given that since 2004, when the above paper was published, many   global subtitling companies have been actively discouraging translators from interfering with the spotting or have made it technically impossible to do so, it is arguable that these companies are knowingly taking a risk as far as subtitling quality is concerned.

In the following statement by Georgakopoulou, MD of the European Captioning Institute, the introduction of template files is linked with outsourcing and, perhaps unintentionally on the part of the author, with the risk of a drop in the quality of subtitles:

Translators from all over the world with no subtitling training can now be given a chance, since the issue of timing is taken away from the equation. Any translator can be given basic subtitling training so as to recognise mistakes and good practice in subtitling, and respect its principles. Any timing issues can be dealt with by the project managers, allowing the translators to concentrate on the translation of the dialogue only (2006: 119; my emphasis).

In principle, Georgakopoulou makes a valid point. It stands to reason and experience that skilled translators with an interest in AVT need only basic training in order to develop subtitling competence. When looked more closely, however, this argument is open to challenge. First, it suggests that training would take place from a distance (“Translators from all over the world”). This would presuppose that appropriate training structures are in place within subtitling companies, which is rarely the case. Second, this strategy would only work with experienced translators (not with “any translator”), who would naturally work for higher rates. However, because the localisation approach has been developed in order to cut costs, novices are usually preferred to professionals, and it is difficult to see how they can respond to the complexities of subtitling. For many newcomers, writing subtitles is their first ever attempt at translating. Finally, “basic subtitling training” here means learning how to translate ready-made subtitle lists and adhere to conventions, with little concern for the rest of the subtitling process. Today, template file creation is casually entrusted to people with scant knowledge of basic translation concepts and strategies, while translation is assigned to linguists with no developed sense of synchrony in AVT. The weakening of the status of audiovisual translators and the slashing of subtitling payment rates has to be understood in precisely that context. As Cronin (2003: 134) tersely states: “The less highly a culture values translation, the less it is going to want to pay its translators.” A comparison of the type of subtitler now favoured by large subtitling companies and that outlined in the ESIST Code of Good Subtitling Practice (1998) would be telling. It would suffice here to mention rule number two of that code, which states that: “It is the subtitler’s job to spot the production and translate and write the subtitles in the (foreign) language reputed.”

Quality in focus

While the present discussion has drawn from a limited amount of existing literature on the use of template files, there have not been any published studies providing hard evidence on the effect that these files have on quality. This is a topic that merits dedicated research. In what follows, a first attempt will be made at examining such a project from the point of view of translation.

The material studied is the episode entitled “Two Tonys” of the DVD edition of the television series The Sopranos.3 The reasons for selecting this episode are, firstly, that the template file is generally accurate and well-timed, which ensures that any issues observed across the target languages could be attributable to the presence of a template file and not to its quality; and secondly, that the script of The Sopranos is rich in nuance and idiosyncratic in terms of register, marked language and sociolect  (e.g. use of Italian words), thus confronting the translator with complex situations which necessitate sufficient room for linguistic and technical manoeuvring.

The present study looked at two target languages, French and Greek. The investigation focused on linguistic issues and did not examine dialogue segmentation (how the source dialogue was divided in one or two-line subtitles in the template) or questions of timing. While these are crucial aspects of the way a template subtitling file interferes with the translator’s choices, such an investigation would go beyond the remit of this paper. It must be mentioned here that the French and the Greek subtitles follow precisely the dialogue segmentation and the timing of the English template file—strong evidence that the translators were constrained to abide by the template file creator’s choices.

In terms of methodology, the following steps were followed. Firstly, a full transcription of the dialogue was carried out. Subsequently, all utterances and on-screen text were juxtaposed to the equivalent subtitles in English (template), French and Greek. A contrastive analysis ensued during which each utterance was examined separately in relation to the corresponding subtitle(s) in the template file and the two target languages. Finally, all linguistic differences between the source text (ST) and the three sets of subtitles were highlighted and categorised according to the translation strategy they represented.

As expected, the main strategies followed in all three sets of subtitles were omission, condensation and paraphrase. While these are standard strategies in subtitling (Díaz Cintas and Remael, 2007: 144-181), this study has found that, in the translation of every individual utterance of the ST, the corresponding French and Greek subtitles would follow the same strategy as the one employed at that point in the English template file. Even when the French and the Greek translators were less constrained for time and space (e.g. when a smaller number of words in these languages would be sufficient to translate a comparatively longer utterance in the ST), they would still tend to work within the operational constraints imposed by the English template. This would frequently result in unnecessarily condensed translations and in systematic omissions that could be confusing and occasionally misguiding.

There follows one example of such a misguiding translation. This episode of The Sopranos introduces a new character, “Feech,” a member of the mob. This character is first referred to in absentia as Michele “Feech” La Manna, in a TV news programme announcing his release from prison. In the English template file, the nickname “Feech” was edited out, presumably for reasons of time. This was an unfortunate choice, because the character is thereafter known as “Feech”, and plays an important role in the fifth season of The Sopranos. (The numbers in the column “English Template” in table 1 below indicate subtitle number, and subtitle duration in seconds and frames.)

Source  dialogue

English Template

French Subtitles

Greek subtitles

You've got Michele "Feech" La Manna,

19                02:02

Michele La Manna,

Michele La Manna,

Ο Μικέλε Λα Μάνα...

[Michele La Manna…]

released just this week, a man greatly feared and respected

20                02:13

released just this week,
a man greatly feared

libéré cette semaine,
très craint

[freed this week,
much feared]

μόλις βγήκε έξω. Σκόρπιζε τον τρόμο...

 [just got out.
He used to spread the terror…]

Table 1.

As can be seen, both the French and the Greek translators chose to eliminate “Feech,” even as, with 2 seconds and 2 frames at their disposal, there was enough time to include this very relevant piece of information. As a consequence, the first time that the Greek or the French viewers actually see the character called “Feech,” they do not necessarily identify him with “Michele la Manna,” the recently released mobster.

The same example contains a second point that is worth discussing. As just seen, in the ST, Feech was “feared and respected”; in the Template file, he is only “feared”. The omission of “respected” can be justified in terms of lack of time and of it being a less relevant attribute than “feared.” Nonetheless, “respected” still adds a significant quality to the character of Feech, especially in the context of the Mafia, and should be included if possible.  Both the French and the Greek subtitlers chose to imitate the template file and omit this term, although they both had enough time to include a translation of it.4

It is possible to find many more examples in this episode of the way the template file has interfered with translation choices, by imposing or stressing a set of constraints that are not applicable in the TLs. An example that has a clear impact on the understanding of the story is the omission of the adjective “little,” repeatedly used by one of the characters, Paulie, to derogate and irritate a younger mobster, Christopher, as table 2 shows:

Source  dialogue

English Template

French Subtitles

Greek subtitles

- "Chrissie, please, don't leave me out here.”
-  You little cocksucker!

133              03:17

- "Chrissy, don't leave me out here."
- You cocksucker!

- "Chrissy, ne me laisse pas."
- Enfoiré !

[- “Chrissy, do not leave me.”
- Fuckhead!]

- "Κρίσι, μη μ' αφήνεις μόνο εδώ".
- Παλιομαλάκα!

[- “Chrissie, do not leave me alone here.”
- Stupid jerk.]

-  Always playin' that
blood-relation card. Tony's little favourite.

135              03:13

Always playin' that blood-relation card. Tony's favourite.

Toujours la carte de la famille.
Le préféré de Tony.

[Always the family card. Tony’s favourite.]

Πάντα ρίχνεις το χαρτί της συγγένειας.
Ο αγαπημένος του Τόνι!

[You always play the family card. Tony’s favourite]

Table 2.

The strained relationship between Paulie and Christopher is one of the main narrative threads of the episode under discussion, and the way Paulie emphasises Christopher’s inexperience and immaturity through the use of the word “little” is of some importance. Both the French and the Greek subtitlers comply with the template file and omit this adjective, although it was technically possible and therefore recommended to translate it. The word “little” is included in the template file in only one of Paulie’s derogatory exclamations; it is interesting then that this is also the only time that this adjective finds it way in the TTs, thus providing further evidence of the way templates affect specific translation choices, as in table3:

Source  dialogue

English Template

French Subtitles

Greek subtitles

- You little fuck...
- Excuse me, gentlemen. Was there a problem with the service?

516              03:23 

- You little fuck.
- Was there a problem with the service?

- Petit connard.
- Le service vous a déplu ?

[- Little jerk.
- Did you not like the service?]

- Μαλακισμένο.
- Είχατε πρόβλημα με το σέρβις;

[- Little jerk.
- Did you have a problem with the service?]

Table 3.


The above example suggests that, in addition to the technical constraints they impose (in-times, out-times, reading speeds, dialogue segmentation), template files also operate as authoritative and detailed sets of guidelines which translators are reluctant or unable to breach. What is more, these guidelines do not always reflect objective and observable constraints in the source language, as in the above examples. Sometimes they can be based on subjective or stylistic strategic choices that need not concern the target languages. Still, these guidelines are being implemented in the TTs, as it will be shown in the final example to be discussed here.

One such strategic decision in the episode under discussion is to eliminate most interjections in Italian—which remind the audience of the origin of this fraction of the Mafia (see table 4):

Source  dialogue

English Template

French Subtitles

Greek subtitles

What a beating we had to give this prick, oh Madonn’!

123              01:22

What a beating we had to give him.

On a dû lui passer un sacré savon.

[We had to give him some hell.]

Του ρίξαμε καλό μπερτάκι.

[We gave him a good beating.]

You don't need another one of those. Last month, it was the upstairs toilet. Basta.

217              03:15

You don't need another one. Last month,
it was the upstairs toilet.

Tu n’as pas besoin d’une neuve.
Le mois dernier, c’étaient les W.C.

[You don’t need another one. Last month, it was the toilet.]

Δεν χρειάζεστε άλλη.
Τον προηγούμενο μήνα, χάλασε η τουαλέτα.

[You don’t need another one. Last month, the toilet broke down.]

20 years, Madonn’!

253              02:03 

20 years.

20 ans.

[20 ans.]

Πάνε είκοσι χρόνια.

[It’s now 20 years. ]

Table 4.

These interjections belong to a characteristic and colourful sociolect that connects the Italian mobsters between them. They also hint to an intertextual relationship between The Sopranos and previous cinematic representations of the Italian-American Mafia, e.g. Coppola’s Godfather films. They have an important function and should arguably be included in the translation. Their transfer into Greek or French is not unproblematic, but it would be possible, at least in some occasions, to use them verbatim. However, they have been systematically omitted in both TTs.

This analysis does not provide conclusive evidence on the effect of the use of template subtitling file on the quality of the translation. A more thorough research that also looks at dialogue segmentation and time-cueing is needed, as mentioned above. Still, the present findings positively support the hypothesis that template files operate as much more than simple aids to the translation (compare with Carroll’s longer statement quoted above). Whether their mediating role is meant to be restrictive or is perceived as such by the translators (or both), they have a substantial impact on the subtitling process. On the basis of the evidence provided here, it is arguable that template files effectively indicate what to translate and how to translate it. To a significant extent, they dictate specific or strategic choices that are often debatable as far as the TL is concerned, and ultimately they tend to replace the audiovisual material as the source text of the translation.  Directly (e.g. through guidelines issued by the translation agency on how to use the template file) or indirectly (pressing deadlines and low-pay contracts), audiovisual translators are encouraged to rely for matters of quality on decisions taken for them by the template file creator.

In that sense it is possible to speak of the degradation of the role of the audiovisual translator as a result of the fragmentation of the translation process.  As this process was broken down into distinct hierarchically structured stages, allowing for a one-way flow of information from the centre, audiovisual translators found themselves constricted to the periphery. As suggested above, “periphery” signifies not only the margins of this system of production of subtitles, but also the geographically remote places, in relation to the centres of production, where the translation is carried out. In both senses of the word, the peripheral role of audiovisual translators is connected with the new globalised character of the business and marks an unequal relationship between translators and translation companies. While this state of the affairs hampers the quality of the localised product, it also has an impact on the professional identity of translators, who are asked to provide less and less of their expertise.

The transfer of trust away from audiovisual translators can be better understood in the context of the troubled position of specialised translators in general, in the age of globalisation. The discourse used to describe the state of AVT in the current market conjuncture is the same as for any other type of professional translation:

Today’s business proposition in translation and localization […] is better, faster, cheaper. […] Translation and localization are seen only as a cost driver and not as an asset; consequently, they are becoming ‘offshorable,’ particularly since they are seen as commodities that anyone can produce because anyone can translate. (Barabé 2007: 424)

Barabé’s remark on the market imperatives for “better, faster, cheaper” translation echoes Bassnett and Bielsa’s point, mentioned earlier, that the pressure of globalisation on translation is for instantaneity, transparency and invisibility. Quite logically, the more invisible translators become the less recognition they enjoy and the poorer their social and financial rewards will be. If “any translator” can become a subtitler after only basic training, and if, as Barabé suspects, the market now tends to assume that “anyone can translate,” then what place is there for specialised translation today and why make the great effort required to study it?

This situation gives rise to a series of further questions in addition to that of quality. In the context of globalisation, questions of ethical and socio-political nature acquire a specific relevance. Ethical issues include the need not to compromise translation service providers’ commitment for high quality service and, equally importantly, the responsibility of both TSPs and translators to maintain proper working relationships between them. Socio-political issues revolve around the positive role of translators in safeguarding cultural and linguistic difference and in halting the trend towards acculturation. This paper, however, will tackle briefly a more specific issue, namely the challenges facing specialised translator training at University level as a result of globalisation in industry.

Prospects for translator training

Higher education translator training programmes have to respond to rampant changes in the professional profile of translators. In an effort to keep up with market trends and anticipate student preferences, specialised translation programmes have multiplied over the past twenty to thirty years. The proliferation of AVT courses in the 2000’s is a case in point. If it is true that recent developments in industry affect the very identity of the specialised translator, then it is legitimate to ask to what extent narrow specialisation continues to be the way ahead for these programmes.

Although it would be useful to know exactly what translation specialisations are currently catered for, and at what level (UG or PG) across Europe, it is beyond the scope of this paper to offer an overview of current translation courses. The general but important question as to the social role of university in helping to shape, rather than simply following, perceptions and trends in the market, will also not be discussed here. This paper will instead propose that a shift of focus should now be considered away from specialisation and towards a holistic approach to translator training that privileges core translation skills, transferable skills and market awareness. The given situation is the changing landscape of the global translation market, the ongoing fragmentation of the role of the translator and the advances in technology with which Universities can hardly hope to keep up. The question is to identify those competences that will equip students with better survival and adaptability skills without compromising quality.

In his report on “Competences for professional translators,” Yves Gambier (2009: 4-7) on behalf of the European Master's in Translation expert group identified six competences to be acquired by a translator. These are translation service provision competence, language competence, intercultural competence, information mining competence, thematic competence and technological competence. Tellingly, in its schematic representation (Gambier, 2009: 4) the EMT report places translation and service provision as a focal point around which other competences revolve. The representation directs attention to the non-subject-specific, non-language-specific and non-culture-specific dimensions of translator training.

In their analysis of this representation, Gambier and the EMT group refer more extensively to the translation service competence than to any other, and explain its two dimensions—the interpersonal and the production dimensions. Broadly speaking, the interpersonal sub-competence refers to knowledge about the translation profession, while the production sub-competence refers to procedural and declarative knowledge relating to the production of the target text. By emphasising these two sub-competences, Gambier’s model places market awareness and understanding of quality at the heart of translator training, a position that is currently occupied by subject and technical specialisation in many translation courses. Significantly, references to adaptability and awareness abound within the specific skills proposed by the EMT group. Thus within the interpersonal sub-competence, specific skills include:

  • Knowing how to follow market requirements and job profiles (knowing how to remain aware of developments in demand.)
  • Knowing how to self-evaluate (questioning one's habits; being open to innovations; being concerned with quality; being ready to adapt to new situations/conditions) and take responsibility (Gambier 2009: 4).

Within the production sub-competence, specific skills include:

  • Knowing how to create and offer a translation appropriate to the client’s request, i.e. to the aim/ skopos and to the translation situation.
  • Knowing how to establish and monitor quality standards (ibid.: 5).

What these specific skills have in common is the acknowledgement of the ongoing diversification trend within the profession and a strong emphasis on the translator’s ability to respond to new requirements. Importantly, the thematic competence, that is the competence most closely related to the drive towards specialisation in University training programmes, also refers to adaptability and research skills rather than to domain-specific knowledge: “Learning to develop one’s knowledge in specialist fields and applications […] (learning to learn)” (Gambier 2009: 7). Equally, as far as technological competence is concerned, along with specialised computer skills, the authors specify the following, as a separate skill: “Knowing how to adapt and to familiarise oneself with new tools, particularly for the translation of multimedia and audiovisual material” (Gambier 2009: 7). Emphasis is again placed on resilience and flexibility.

It is interesting to note that in the competences list proposed by Christina Schäffner in 2000, the need for adaptability was implicit but not clearly spelled out. Schäffner had back then defined six types of translation competences: Linguistic, cultural, textual, domain/subject specific, research, and finally transfer competence (Schäffner, 2000: 146). There is a perceivable shift of emphasis from specific knowledge acquisition to the development of versatility skills (“learning how to learn”), while the quality requirement remains the same. This shift reflects well the growing realisation that it is difficult to reconcile the ever-changing landscape of professional translation in a globalised setting with the drive for specialisation.

In a different but still relevant way, the European Standard for translation services, EN 15038: 2006, avoids any reference to subject-specific knowledge for professional translators. It concentrates instead on fundamental competences, such as language, research, culture and technology, all of which follow a first and basic “translation competence” (see Olohan, 2007: 50-51). Translation competence is defined here both in absolute terms, i.e. with regard to a required quality level specified in the standard, and in relational terms, i.e. as “the ability to render the target text in accordance with the client-TSP agreement.” Again, the focus is on quality and market awareness. The standard is not concerned with translator training, but it gives a good idea of what is expected of a professional translator, which is of course a key factor in the designing of University translator training programmes.

It is thus worth thinking whether Universities should focus more on educating translators whose competitive advantage would be the ability to adjust to emerging roles. Anderman and Rodgers (2000: 68) have expressed this very well:

Narrow specialisation would more appropriately be the remit of professional organisations than academic institutions, as translators already working in the professional market are clearly more aware of its specific requirements, leaving the responsibility to the universities to introduce students to research, evaluation and translational practice, i.e. to introduce them to transferable skills.

While Anderman and Rogers make a very valid point, the present paper’s suggestion is for a change of focus and not for the exclusion of specialised translation modules from curricula. Without doubt, domain-specific competence should continue to form part of translation programmes, especially at postgraduate level. After all, specialised translator training is not merely a matter of acquiring knowledge on a subject, but also, and perhaps chiefly, of learning how to select, generate and critically support a specific text type (e.g. legal, technical, audiovisual) for a specific purpose. Training in specialised translation should therefore in itself aim at fostering adaptability skills, or, as Gambier puts it in the extract quoted above, at “learning how to learn.” Thus translators will not only be able to continually update their knowledge in the subjects in which they specialise, but they will also be ready to tackle new subjects in a market where few specialisations remain profitable for more than a few years.

A useful approach to non-subject and non-cultural–specific competence is Pym’s “minimalist concept of translation competence” (Pym 2003). Pym responds to the same urgencies as this paper, namely rapid technological and professional change that Universities are not always in a position to influence or anticipate. He identifies the “particular kind of competence that translators bring to other professional communities” as the ability to produce viable text variants for a given source text, and then to select one of these variants “with justified confidence” (2003: 489). Pym’s proposition places this fundamental translation competence at the heart of translator training. It has the additional benefit of distinguishing the role of the translator from the other roles that he or she has to play (those or terminologist, reviser, localiser, subtitler and so on, as per Gouadec’s list, quoted above).

It is not my purpose here to debate the details of Pym’s proposition or to compare it with other competences lists. For one thing, his minimalist model is less concerned with market awareness and transferable skills than those mentioned above and other vocation oriented propositions (e.g. PACTE 2003; Olohan 2007). The reason I am mentioning it here is because it provides a single point of reference for translation programmes. At a time when translation as a materially, socially and personally rewarding profession is imploding, Pym’s model determines an identity for the translator—albeit a minimalist one. It is also a non-exclusive model, in the sense that it can easily co-exist with skills that are not part of translation competence as Pym understands it, such as knowledge about the profession and technological skills.

To sum up, this section suggests that a training programme concentrating on a core translation competence with parallel emphasis on market awareness and transferable skills would present significant advantages. Firstly, there would be advantages for the university, as the pressure to keep pace with market trends would ease, leaving more room and resources for academically informed curricular renewal and research. Secondly, there would be advantages for students: a broader range of analytical and transferable skills would protect them from (or at least prepare them for) the shock of entering a market that may not place enough value on any one translation specialisation. By the same token, the translation market would benefit, as translators would be readier to adapt to emerging roles and more capable of sustaining quality and the argument for quality in the business. Finally, translation as a professional activity will also benefit, as its role in the age of market and cultural globalisation will be better defined. Rather than semi-anonymous helpers at the margins of the translation service business, translators would be specialists equipped with a unique recognisable competence and an ability to apply it within diverse formative and professional contexts.


This paper set out to discuss the impact of globalisation on translation and translators and the relevance of recent developments for translator training at University level. The paper began with the remark that translation is nowadays characterised by increasing invisibility—a situation which has direct effects on the professional identity and self-perception of translators. In practical terms, the pressure for instantaneous, transparent and invisible translation is being expressed through the relentless fragmentation of the profession into multiple roles. Focusing on subtitling as an example, it was suggested that fragmentation operates on a temporal and a spatial level, through such developments as the introduction of template subtitling files and outsourcing. According to subtitlers and at least some theorists and representatives of the industry, these structural changes are detrimental to the quality of the translations produced. A first attempt at providing hard evidence of the impact of template files on the linguistic quality of subtitles has confirmed these concerns, although more research is needed. It is possible to argue, then, that subtitlers, just like the rest of specialised translators, have fewer opportunities today to practice their specialisation or to be heard in matters of quality. In addition to various ethical and political questions, it is important to ask exactly where this state of the affairs leaves translator training programmes.

Looking at a small number of recent translation competences lists, this article found that there is relatively more focus today on adaptability, market awareness and transferable skills than there is on specialised translator training. Theoretically at least, translation educators and trainers begin to worry about narrow specialisation and its adverse effects on translators’ professional prospects. Despite the trend for specialised translation courses over the past twenty years, there have also been voices calling for a holistic approach based on a core translation competence that is not language, culture or subject-specific. As an example of such an approach, Pym’s minimalist concept of translation competence was highlighted. This concept could serve as a single point of reference for the development of translation courses that place relatively less emphasis on specialisation and more emphasis on market awareness and transferable skills than is the case today. Perhaps more importantly, a holistic approach based on a single core translation competence would help translators redevelop a sense of unique professional identity. At a time when fragmentation is set to increase, a solid identity that is adaptable to various professional roles rather than being incessantly redefined by them would constitute an indispensable asset for translators.

  • Alberge, Dalya (2007). “The greatest stories ever filmed (mangled by outsourced subtitles).” The Times, March, 19. (consulted 03.04.2011).
  • Anderman, Gunilla and Margaret Rodgers (2000). “Translator Training between Academia and Profession: A European Perspective”. Christina Schäffner and Beverly Adab (Eds) (2000). Developing Translation Competence. Amsterdam/Philadelphia: J. Benjamins, 63-73.
  • Barabé, Donald (2007). “Towards a New World Order in Translation.” Marcel Thelen and Barbara Lewandowska-Tomaszczyk (2007) (Eds). Translation and Meaning, Part 7. Maastricht: Zuyd University, 421-428.
  • Bielsa, Esperança and Susan Bassnett (2009). Translation in Global News. London and New York:Routledge.
  • Carroll, Mary (2004). “Subtitling: Changing Standards for New Media?” (consulted 03.04.2011).
  • Cronin, Michael (2003). Translation and Globalization. London and New York: Routledge.
  • Díaz Cintas, Jorge and Aline Remael (2007). Audiovisual Translation: Subtitling. Manchester: St. Jerome.
  • Gambier, Yves (2006). “Multimodality and Audiovisual Translation”. in MuTra 2006 – Audiovisual Translation Scenarios: Conference Proceedings. (consulted 03.042011).
  • — (2009). “Competences for Professional Translators, Experts in Multilingual and Multimedia Communication,” Brussels. (consulted 03.04.2011).
  • Georgakopoulou, Panagiota (2006). “Subtitling and Globalisation.” JoSTrans, The Journal of Specialised Translation 6, 115-120.
  • —  (2009). “Subtitling for the DVD Industry.” Jorge Díaz Cintas and Gunilla Anderman (Eds) (2009). Audiovisual Translation: Language Transfer on Screen. Basingstoke/New York: Palgrave Macmillan, 21-35.
  • Gottlieb, Henrik (1996). “Theory into Practice: Designing a Symbiotic Course in Subtitling.” Christine Heiss and Bosinelli R. M. Bollettieri (Eds) (1996). Traduzione multimediale per il cinema, la televisione, la scena. Bologna: Clueb, 281-295.
  • Gouadec, Daniel (2007). Translation as a Profession, Amsterdam: John Benjamins.
  • Ho, George (2008). Globalisation and Translation: Towards a Paradigm Shift in Translation Studies. Saarbrücken: Berlag DFr Müller.
  • Media Consulting Group (2007). “Study on Dubbing and Subtitling Needs and Practices in the European Audiovisual Industry: Executive Summary”, Paris/Peacefulfish, London.  (consulted 03.042011).
  • Nakata Steffensen, Kenn (2007). “Freelancers and the Crisis in British Subtitling”. ITI Bulletin, May-June 2007, 18-19.
  • O’Hagan, Minako (2005). “Multidimensional Translation: A Game Plan for Audiovisual Translation in the Age of GILT”. MuTra 2005: Challenges of Multidimensional Translation: Conference Proceedings.'Hagan_Minako.pdf (consulted 03.04.2011).
  • —  (2007). “Impact of DVD on Translation: Language Options as an Essential Add-On Feature.” Convergence 13(2), 157–168.
  • Olohan, Maeve (2007). “Economic Trends and Developments in the Translation Industry: What Relevance for Translator Training?”. The Interpreter and Translator Trainer 1(1), 37-63.
  • PACTE (2003). “Building a Translation Competence Model”. Fabio Alves (Ed.) (2003). Triangulating Translation: Perspectives in Process Oriented Research. Amsterdam/Philadelphia: J. Benjamins, 43-66.
  • Pym, Anthony (2003). “Redefining Translation Competence in an Electronic Age. In Defence of a Minimalist Approach.” Meta 48(4), 481-497.
  • — (2010). Exploring Translation Theories, London/New York: Routledge.
  • Schäffner, Christina (2000). “Running before Walking? Designing a Translation Programme at Undergraduate Level”. Christina Schäffner and Beverly Adab (Eds) (2000). Developing Translation Competence. Amsterdam/Philadelphia: J. Benjamins, 143-156.
  • Venuti, Lawrence (1995). The Translator’s Invisibility: A History of Translation. London: Routledge.

The Sopranos (1999-2007). Tim Van Patten et al. USA


Dionysios Kapsaskis photoDr Dionysios Kapsaskis is Senior Lecturer in Translation at Roehampton University, London. He has published on Audiovisual Translation and Comparative Literature and has worked for twenty years as a specialised translator, subtitler and proofreader.

Note 1:
On the recent fall in subtitling rates, see SUBTLE (2007). See also the relevant discussion on the webpage of the equivalent French Association, Atta (2011), especially section 3, “Assurer une veille sur l’état de la profession .” On the issue of outsourced subtitles, there is an interesting article by Dalya Alberge (2007).
Return to this point in the text

Note 2:
The Media Consulting Group’s study on dubbing and subtitling in Europe (2007) contains the following assessment:  “As regards DVD distribution, the interviews reveal that the quality of subtitling is quite low. For example, the German dubbing/subtitling industries complain of the quality of DVD subtitling and explain that the German subtitles for this format were not done in Germany but in Austria, Switzerland or the Netherlands without any guarantee of quality. Also, the translations for multilingual versions of American DVDs are sometimes done in non-English speaking countries like India or the Philippines, which affects the linguistic quality of the final product.” (Media Consulting Group 2007: 72). See also the report’s conclusion on the issue of quality (Ibid.: 128).
Return to this point in the text

Note 3:
The Sopranos is an American television drama series which ran for six seasons, from 1999 to 2007. The episode under discussion, “Two Tonys,” is the first of the fifth season and was originally aired in March 2004. The subtitles are from the DVD edition of the complete fifth season (2004).
Return to this point in the text

Note 4:
For example, in French, “libéré cette semaine, très craint et respecté” (freed this week, much feared and respected) would give a reading speed of 166 words per minute; in Greek, “μόλις αποφυλακίστηκε. Eνέπνεε εξίσου τρόμο και σεβασμό” (was just released. He inspired fear and respect in equal measure) would also give a reading speed of 166 words per minute. This is a perfectly acceptable value.
Return to this point in the text