RSS feed

Subtitling and Globalisation

Panayota Georgakopoulou, European Captioning Institute, London


In this paper I will discuss recent developments in the subtitling field since the emergence of DVD. The digital format is changing the face of the rapidly expanding subtitling industry as new demands are calling for a new working methodology and the notion of the ‘template’ is born. The description of the way template files are produced and used is an account of the working methodology that has been developed in the ECI Ltd, a London-based subtitling company that specialises in multilingual DVD subtitling.


Subtitling, DVD, template files, multilingual subtitling


Panayota Georgakopoulou is a graduate of the Aristotle University of Thessaloniki, Greece, inEnglish Language and Literature. She completed her education in the UK, acquiring an MA in Translation as well as the degree of Doctor of Philosophy on the subject of subtitling from the University of Surrey. Yota taught translation for four years at the University of Surrey and the University of Westminster, setting up subtitling modules in both universities, while at the same time working at various positions in the subtitling industry. Starting as a subtitler in the mid-1990’s, Yota is now the Managing Director of the European Captioning Institute Ltd, a leading subtitling company in the field of multilingual subtitling for DVD.

Since the birth of interlingual subtitling, the way subtitles have been created and presented on screen has gone through many changes, mainly dictated by numerous developments in technology.

In the beginning, subtitles were created by two types of professionals: technicians in charge of spotting a film, and translators who wrote the subtitles. In some cases, the two activities would be undertaken by the same person. The spotting of a film was done by defining the in and out times of the subtitles based on the start and finish of actors’ utterances or the end of a take. These times were marked on the script and translators were then used to write the subtitles. The amount of time spotted for each subtitle corresponded to a specific number of characters that the translators were allowed to use so that the viewers would have sufficient time to read it. Translators worked from a script and, in the best case scenarios, they would attend a viewing of the film before starting their work and would have the opportunity to check their subtitles against the film later. Despite the fact that detailed timecoded spotting lists are provided for theatrical subtitling today, this is very much the way subtitlers still work when it comes to subtitling for the cinema. With the appearance of television, various means and techniques were developed to cue the subtitles while a film or programme was being broadcast or recorded (Carroll, 2004).

In the mid-1980’s, the use of time codes and PCs was to revolutionise the process of subtitling and the profession of the subtitler achieved proper status in the industry. Subtitlers now worked on workstations that comprised a PC with dedicated subtitling software and a video recorder. They worked off VHS copies of the film and a script in the original language. Later on, the film or audiovisual material was encoded and made accessible to the subtitlers’ digitised workstations via servers. The subtitlers could now 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. Once ready, the subtitle files would be inserted electronically into a dub of the master tape. The process for DVD subtitle creation is similar with the exception that the finished subtitle files are converted into graphic images, i.e. tiffs or bitmaps that are necessary for the DVD authoring process.

The appearance of digital video formats in the market, however, was bound to set yet another milestone in terms of subtitle production, as it brought with it the possibility of centrally controlled services, such as the provision of multilingual subtitling for DVD releases. In the past decade, the subtitling industry has witnessed regional variation being slowly subsumed by global production. This has come as a direct result of the requirements imposed on the market by the large Hollywood studios.

First of all, the quantity of DVD subtitling has boomed to such an extent in recent years, that the ability to produce subtitles in 40 or more languages simultaneously, in very fast turnarounds, has become an imperative for most companies.

Piracy is the other major factor underlying the new state of things. According to recent studies, studios are losing millions of dollars to piracy every year. The Motion Picture Association of America (MPAA) estimates that the U.S. motion picture industry loses in excess of $3 billion annually in potential worldwide revenue due to piracy, excluding any further losses that are caused as a result of Internet piracy. Studios thus need to find ways to control their assets better, and this can be achieved by storing them centrally rather than sending them all over the world to the ‘territories’, a term used in the industry to refer to the countries where the DVDs are to be sold. Furthermore, day and date releases are becoming more widespread, as the studios are trying to shorten the window between theatrical and DVD releases so as to combat piracy.

Cost is also a major market force that shapes the way things are done, and film production tends to be always cash poor. Central production of DVDs as opposed to local production in the territories is obviously less expensive for studios, both in real terms, but also in terms of indirect costs, such as studio administration. It seems only logical that studios would prefer to deal with one central vendor only rather than have to coordinate work in 40 or so territories.

Finally, there is also the copyright issue. Studios generally do not have the infrastructure to deal with and negotiate copyright. They pass on this responsibility to their vendors who are asked to pass on copyright back to the studios. This way it is easier for studios to keep track and archive their assets, such as subtitle files, and hold the copyright to re-use them as necessary for adaptation or reformats to other media, e.g. for VHS releases, airline releases, etc.

As the distribution centres for Hollywood studios are largely based in Los Angeles and London, the answer from the market came in the form of subtitle companies with offices in these two key cities, able to produce subtitles in all the languages the studios needed. These companies largely evolved from companies that originally specialised in the provision of closed-captioning. Faced with the developments of the industry, they took advantage of the availability of professional translators in any imaginable language that lived and worked in these two cities, and diversified into interlingual subtitling to the point that this became their main business. One such company is ECI and there are numerous similar companies appearing in the market every day. The problems such companies were faced with were mainly three:

  • They needed to find a way to produce the subtitle files in all the required languages in increasingly short turnarounds, using cost-effective methods.
  • They also needed to be able to manage and check all files centrally, without necessarily having the in-house capacity in the 40 or so languages they provided subtitle services into.
  • And finally, they needed a vast subtitler resource to produce the work in all these languages.

The solution to all these problems came in the form of universal ‘templates’ (also referred to as ‘transfiles’ and ‘genesis files’ by some companies), subtitle files in English that are meant to be used as the basis for translation into all languages (Georgakopoulou, 2003: 210-221). In other words, the subtitling process was once again split in two phases: timing and translating. The timing of a film or audiovisual programme is carried out by English native speakers, who produce a timed subtitle file in English. This file is then used as the basis for translation into all languages required in a project, a task carried out by native speakers of these languages.

As the use of template files became more widespread, it was obvious that a lot of ancillary information needed to be kept so that all subsequently created language files could be properly checked before they were rendered into tiffs or bitmaps. Such information was kept in a number of files that are produced together with the templates:

  • First there are the translation notes files, which offer explanations in terms of slang, culture-specific expressions, or anything with which non-native speakers of English might be unfamiliar. These notes are there to help the translators overcome any problems of understanding the source language and produce a more accurate target text. Translation notes are not dissimilar in nature to the notes one finds in combined dialogue and continuity lists, but include further instructions to translators, as cases of irony, play on words, and the like are also pointed out.
  • There are also lists of names and titles, especially in the case of series, that are created to help the various translators working simultaneously on the same project ensure that their translations are consistent in this respect. This is particularly essential in cases of translators working on different episodes of the same series, as well as of translators working on the feature and bonus material of the same title.
  • Lists with information on burnt-in text are also created to cater for subtitles that need to be treated differently in different languages. This information might require translation in some languages but not in others, whereas all subtitles appearing on the screen at the same time as burnt-in text need special care in terms of positioning. An extreme example would be Japanese subtitles that require vertical positioning in such cases instead of horizontal. Such lists are also used for the creation of forced subtitle files that are meant to be used together with the dubbed versions of films that are included on DVDs.
  • More lists are created for subtitles that appear in both the feature and bonus material subtitle streams, such as Making Of’s, audio commentaries, etc. The reason again is the need for providing consistently translated subtitle files.
  • Instructions on client-specific stylistic issues, such as the treatment of songs, are also given, and glossaries are provided to ensure consistent translation of bonus material titles that need to correspond with the translations used on the DVD menus.
  • Finally, there are instructions on technical issues relating to the use of fonts, italics, alignment of subtitles, etc. as requested by the clients, in order to achieve the consistent and specific look studios want their DVDs to have so that they are easily recognisable in the market.

With all these demands placed on subtitling companies in terms of DVD production, the use of base English files (i.e. templates) with fixed timings and a fixed number of subtitles in all subsequently produced subtitles files has proved to be the necessary means to make such work possible. The main advantages of using the template working method are the following:

  • Mishearings of the original audio (usually English) are kept to a minimum, as native English speakers are used to produce the English template files. Non-native speakers, on the other hand, would probably be faced with serious difficulties, especially in the case of material such as audio commentaries, Making Of’s, or Deleted Scenes. Such bonus material is largely unscripted, often poorly enunciated, or of poor acoustic quality (often the reason why a particular scene was in fact deleted from the final cut of the feature), and invariably hard to understand. One need only watch a locally subtitled production to encounter various examples of such mishearings.
  • Costs are minimised, as the film or audiovisual programme need only be timed once, in English, and lower rates of payment can be used for its translation in other languages, as translators are only asked to concentrate on the translation of the subtitles.
  • Faster turnarounds are also achieved this way.
  • Furthermore, an English reference is provided for every single subtitle in every language, so that subsequent changes that need to be made to files when adapted or remastered for other media can easily be applied. The same English reference is necessary in the management of the multiple language files produced, as well as in the process of the technical checks these files have to undergo before they are ready for DVD simulation.
  • Finally, the recruitment needs of subtitling companies catering for multilingual subtitling are solved, as the potential resource of freelancers is instantly expanded. It is no longer necessary to recruit subtitlers in all the languages involved in a project, which could prove to be quite a challenge. 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.

This working methodology has in the last few years attracted much criticism in relation to the quality of the subtitle files produced this way. Of course nobody can claim that there are no mistakes made in the process, especially since much depends on the care and effort put into the creation of the template files and all the subsequent quality procedures language files undergo. The quality of the end result is always down to the experience and professionalism of the staff involved in the process, and this is as much the case in the DVD subtitling industry as it is in the translation industry in general. However, it should be pointed out that the DVD subtitle production process with the use of templates bears some affinity with the way theatrical subtitles were originally spotted and translated, only with the disadvantages of this method minimised to a large extent. Furthermore, the three ‘rhythms’ (Carroll, 2004) that a subtitler adheres to when subtitling the ‘traditional’ way, i.e.

  • the visual rhythm of the film as defined by the cuts,
  • the rhythm of the actors’ speech, and
  • the audience’s reading rhythm

are still adhered to, since the source material and therefore the points at which actors open and close their mouths to utter their dialogues still remain the same. Also, regional variation is still possible within these strictly timed subtitle files. For example, different subtitle conventions such as the use of single as opposed to double dialogue dashes is still possible, as traditional subtitling required in Dutch and Bulgarian files for instance. Translators also have the choice to leave entire subtitles blank (or untranslated) if they feel that a translation would be unnecessary in their language. This is also often the case in Dutch files, due to the fact that the Dutch audience would easily understand simple phrases such as ‘Help!’ or burnt-in text that is spelled in a nearly identical way in the their language as in English. Even different reading speeds can be catered for in different languages, with the provision of bespoke software to the translators working on a multilingual project, so they can set the reading speed of their files to what is conventionally used in their countries.

So, what does the future hold? Every time the word ‘digital’ is used, studios see this as an opportunity for centrally produced work. Studios are the driving force behind this working methodology and they are continuously exploring ways of using this working method further, as it fulfils their main requirements: it is cheap, fast, and more secure. Right or wrong, it is a working methodology that is here to stay as studios would only change their minds if the work produced affected their sales – and to date there is no evidence that this is happening.