Subtitles by Machine

A couple of months ago, Swedish TV4 caused a bit of a scandal when they said their subtitles were done by machine. They then backtracked on that, but given some of the mistakes they make, it’s hard to know what to think. Here’s an article about it.

Subbed vs. Dubbed: Where do you stand?

This is a guest post by TeachStreet writer Kenji Crosland. TeachStreet is a website dedicated to providing local and online classes, including foreign language classes in languages like Spanish and French.

There’s a lot lost in translation: humor, depth, and sometimes even the basic meaning of words themselves. Although great translators can manage to capture the spirit of the original, it can take them years to do so. I have even heard of some translators spending days pondering the nuance of a single phrase! Because of strict deadlines, however, film translators don’t have that kind of luxury, and thus the translations aren’t as good as they could be.

A very clever joke in English, for example, might go flat when translated into Japanese. With time, a translator might be able to think of a way to make a joke work, but usually they can’t. I had this experience myself when I watched the movie “Dodgeball” in Tokyo. I must have laughed out loud several times at intervals when the crowd was silent. It was not until I had read the subtitles that I realized why. Oftentimes certain jokes weren’t translated at all, and were replaced with lame Japanese jokes that were similar in nature but failed to hit the punchline.

I was able to forgive the translators their terrible work because I knew just how difficult a job it was. A year or two back I had read a Japanese news article (Sorry, the title escapes me) about how hard it is to translate for movies. You can’t have subtitles crowding half the screen, so you’re limited to a certain amount of characters (just like twitter). When the actors are talking rapid-fire, sometimes you have to cut out part of what they’re saying from your translation just to keep up with the flow. Jokes, which often require cultural and linguistic context, often don’t stand a chance.

If you’re dubbing, however, you have a little more freedom. Although dubbing has a bad reputation in the States because the voice acting for most dubbed movies is horrendous and the words often appear out of sync with mouth movements, we must remember that the budget simply isn’t there to make dubbing better (by hiring better actors, sound technicians, etc). When done right, however, dubbing can be a good alternative to subtitles. First of all, you’re able to add more colloquial language which can be less stilted than subtitles. If you have a comedian voice actor, for example, they might be able to ad lib a joke or even use a certain voice inflections which carry a joke’s meaning much more effectively than a stale sentence would. Furthermore, although you have a time limit, you have no character limit, so you may not have to cut short dialogue in order to fit the screen.

Personally, I haven’t seen many good dubbed movies myself, although some animated movies seem to do a fair job at it. I have heard it’s possible, however. I can’t speak for myself, but a friend of mine told me that the dubbed Italian version of the first Spiderman movie was done so well it was hard to tell that they weren’t the original voices.

If I had a choice, I’d probably choose subtitles over dubbing most of the time. In some movies, however, the subtitles can be so distracting from the action that you’ll spend more time reading than actually watching--especially when you don’t know one word of the original language (I had this experience with “Crouching Tiger, Hidden Dragon”).

My guess is that most people prefer subtitles to dubbed movies. I’m wondering, however, if the vote wouldn’t turn out differently if more time and money was spent improving the quality of dubbed films. So what do you think? Subbed or Dubbed?

Getting a PhD in Translation Studies

I am nearing the end of my time in a doctoral program in translation studies, so I thought I would write a little bit about what it means to get a PhD in the field. A shorter version of this was published as a guest post earlier this month.

In September 2006, I moved from Sweden to Wales in order to study at Swansea University. There are not that many schools yet that offer translation studies; more often, one must study a language or comparative literature. So what does it mean to be in a translation studies program?

Translation programs on the BA or MA level generally focus on training translators. Such programs combine theoretical and practical work. Students improve their language skills, read and discuss translation theory, practice translation, learn about computer programs and terminology, and maybe get information about starting a company or working for agencies, and other such things. In other words, these kinds of programs are aimed at students who are good with languages and want to work in the field of translation.

In a sense, translation studies might as well be totally unrelated. I have met many people who study or work in the field of translation studies and yet have never translated and have no intention of doing so (I tend to find this odd, but that is a different issue). In a PhD program, a student is being prepared to become a researcher, not a translator. As in BA or MA programs, students learn about translation theory, but by the PhD level, they are expected to have (or to quickly obtain) in-depth knowledge about this. Students should already have extensive language skills. One doesn’t really attend courses, although this depends on what country the program is and what individual students require. For example, I chose to sit in on several classes about translation theory and the history of translation, mainly out of interest and a desire to refresh or extend my knowledge. Basically, one spends most of the time researching.

Research what? Well, there are many different possible areas. One can research and analyze the translation of specific kinds of non-fiction or fiction works or specific types of language, the translation of a particular author, what it means to translate between two or more different languages, how translators feel about their jobs, what translators actually do as they work, how translators are or ought to be trained, how translators use (or don’t use) computer tools, how ideas of translation have changed over time, critiquing translation, how translation can be used to control certain populations, how translation can develop a target language, what conditions translators work in, differences in how translators and those studying to be translators work or think about their work, and much, much more. Remember that much of this can apply to interpreting too, which is generally subsumed under the field of translation studies, though interpreting studies as a separate field is growing, and also to subtitling.

As an example, my own research has been focused on children’s literature and I have been particularly interested in how figurative language is used and translated in books for children from English to Swedish. I also know people who research the translation of medical texts between English and Chinese, and the translation of idiomatic phrases in non-fiction from Spanish to English, and the subtitling of talk shows. Some researchers use computer programs to help with their research (particularly if they need a large corpus of texts), while others interview translators or sit with them while they work, and still others focus on close analysis of texts.

Those who are starting out in the field often spend a lot of time learning about translation theory in general and their particular field specifically. For example, in my first term or two in the PhD program, I read everything I could find on the translation of children’s literature, on translation in the colonial and postcolonial contexts (this was related to my need to learn more about translation and power), on functionalist theories and skopos, on translating dialects and wordplay, and related topics. Others might want to read about gender theories or issues of in/visibility or financial translation or interpreting in a legal setting or think-aloud protocols.

The next step is picking one’s texts and starting the research and, of course, trying to find something new and important to contribute to the field. I use primarily textual analysis and statistical analysis, which means I study texts and their translations, and then compute how common certain translatorial strategies are. In the first term, students often begin writing literature reviews and chapters of their dissertation. Here, one’s supervisors should give detailed criticism on one’s writing style and ideas.

In the first year, many students start attending academic conferences and sometimes even presenting at them. Conferences are an excellent way to learn about what research is taking place in the field and also to get feedback on one’s budding research. Next, one ought to try to get articles published. Attending and speaking at conferences and having work published are both essential when one is finished and looking for a job. Research trips may also prove beneficial; I spent two wonderful weeks at the National Library in Stockholm, studying various translations of work by Lewis Carroll and Mark Twain.

Students must be independent and good at working hard and making their own schedule. Many people don’t understand that being a PhD student is very different from studying at the BA or MA level. No one will give you deadlines or tell you what to do (generally, that is; some supervisors might be a bit more hands-on). You have to recognize that everything is up to you and that you have to prepared to be very active.

I have really loved my time being a PhD student in translation studies. I have continued to translate, edit, write, and teach throughout my years in Wales, and that has been really stimulating for me, although many PhD students prefer to focus solely on their research. It’s a lot of fun to research translation and to try to contribute to the field and in the future, I hope to continue combining research with being a practicing translator.

Conference on Subtitling

I saw this announcement for a conference on subtitling and thought some of you would be interested:

Languages & The Media
7th International Conference on Languages in the Audio Visual Media

www.languages-media.com

**************************************************
Quality Audiovisual Media for All

// Quality standards of subtitling and translation are high on the agenda of this year's Languages and The Media conference //

Berlin, Germany. The conference programme of Languages and The Media has been finalised. The 7th International Conference and Exhibition on Language Transfer in the Audiovisual Media will take place from October 29th - 31st at Berlin's Hotel InterContinental.

Bringing together delegates from more than 20 countries, the conference will contribute to the international debate on inclusion and universal access to mass media on a global scale.

The event focuses on the translation and transfer of language in films and on television, as well as in interactive media such as computer games and the Internet. Experts from the fields of media, translation and academia from all over the world discuss current developments in the media industry and exchange their expertise.

The conference programme offers insight into quality standards of translation, synchronisation and subtitling. Further topics are the localisation of content, as well as the effect of new tools and future technologies on the transfer of language, like machine translation and speech-recognition captioning.

Localisation refers to the process of adapting digital content to culture, locale and linguistic environments at a high quality. Carmen Mangiron, who is one of the localisers of the Final Fantasy series into Spanish, will show how language barriers in video games can be overcome through audiovisual translations and editing techniques.

Subtitling and audio description enable the deaf and hard-of-hearing as well as the blind or partially sighted audiences to access media. Bernd Benecke from Bayerischer Rundfunk - Germany's only full-time editor for audio description - will offer insight into this rare discipline in a pre-conference workshop.

The conference will be accompanied by an exhibition, showcasing vendors and manufacturers of language technology products and providers of language services.

Further information: www.languages-media.com

LANGUAGES & THE MEDIA
7th International Conference & Exhibition on Language Transfer in Audiovisual Media
October 29 - 31, 2008, Hotel InterContinental Berlin

Participation fees: 400 Euro, students 190 Euros
Organiser: ICWE GmbH, Leibnizstr. 32, 10625 Berlin, Germany
Contact: Ms Astrid Mendoza, Tel: +49 (0)30 310 18 18-0
info@languages-media.com, www.languages-media.com

A Bad President Under a Crowd

Not long ago, I was somewhere that had several flat-screen televisions lining the walls. The volume on the tvs was off, but programs were playing anyway, and closed-captioning was used so those watching could know what was being said.

I know that closed-captioning, unlike subtitling, is generally in real-time, but I was still surprised by the number of mistakes -- there were errors in nearly every sentence. Some were really odd, though many were clearly based on phonetic confusions. Sometimes a caption was corrected, but usually the viewer was left to puzzle it out (and to giggle, as in my case).

Here are a few of the wrong captions I recall:

“This sets a bad president” instead of “This sets a bad precedent”
“Now things are under a crowd” instead of “Now things are under a cloud”
“This is about award” instead of “This is about a war”

Bad closed-captioning and bad subtitling can definitely set a bad president.

Subtitling

Subtitling is a topic that has not been mentioned on this blog, though it is of course a kind of translation. So I was interested by an article on subtitling movies.

In this article, the director Timur Bekmambetov is quoted as saying, “We thought of the subtitles as another character in the film, another way to tell the story.” This is certainly a different view of subtitling than the usual idea that it is a somewhat unfortunate, distracting necessity. Subtitles were often expected to be as unobtrusive as possible, whereas this director wants to highlight them, and make them a real part of the movie, and he does that by using colors and effects.

The article goes on to explain: “The subtitles that will allow non-native viewers to follow the stories are crucial because no matter how flashy or impressive a movie may be, it’s the subtitles that can stifle or showcase its quality. Although many audiences around the world, most of whom see foreign films dubbed, consider them the cinematic equivalent of Brussels sprouts, subtitles remain an unsung yet essential tool of moviegoing. And with technology improvements, more people speaking foreign languages and the modern habit of multi-tasking, the traditional aversion to watching a film while reading it just might be on the wane.”

One small quibble I had with the piece was the idea that while “literature [which] has the safety net of footnotes, film subtitlers have to make it work in the moment, all while trying to adapt wordplay and cultural references.” I think many translators of literature would be surprised to hear this, since they obviously also attempt to wordplay and cultural references and make the story or poem work as it is, without resorting to footnotes. It’s true that translators can use footnotes or endnotes if necessary, but many would prefer to avoid it, so as not to take attention away from the piece.

It was interesting to learn how much time (1.5 seconds per subtitle) and space (45 characters per line max) there is for subtitles and also to be reminded that there “are logical rules as well, such as finishing a subtitle when a character stops speaking and not extending it over a cut, which can be disorienting. Good subtitles work with the rhythm of the scene, based on accurate spotting that captures that timing.”

Translatorial Censorship

In the last post, I quoted from Dr. Marjorie Garber’s book Vice Versa: Bisexuality and the Eroticism of Everyday Life, in which she mentioned some translatorial/editorial changing/censorship (or “gender bowdlerization”) of work featuring homosexuals and bisexuals. Unfortunately, this is not a unique or altogether rare phenomenon.

In the special Translating Humour issue of The Translator magazine from 2002, in an article entitled “Francoist Translation Censorship of Two Billy Wilder Films,” Jeroen Vandaele writes about how, during Franco’s regime, translatorial censorship took place of work that was considered inappropriate or immoral. For example, Some Like it Hot might be considered amoral because of issues relating to its portrayal of cross-dressing, the potential gay implications of the movie, and other sexual topics (especially, it might be noted, sexual issues that are outside the realm of what is accepted as “normal”). Dr. Vandaele says that some of the sexual humor in Some Like it Hot and The Apartment was “changed or deleted because of immorality” or “replaced by morality,” and one can assume that if this sort of censorship happened to films, it was part of a general view of culture and society that also affected literature.

Whether for political or other reasons, translators and editors (and other people with power, such as teachers) sometimes censor or change material that they consider improper or otherwise unsuitable. I remember someone who grew up in Iran telling me about seeing foreign films in Iran and then, once she had moved to Europe and later to America, seeing the same movies and being surprised at how much longer they were; in other words, “inappropriate” material had been deleted or changed before the films were deemed acceptable in Iran.

Personally, I’m a strong believer in having as little as possible come between the audience and the text (or film or whatever) as the author (or director or whoever) envisioned it, and I hope that we translators will be cautious about (ab)using the power the wield.