Metonymy in subtitling

This piece is based on a presentation given to fellow PhDs at the University of Agder on 29 November 2023. The brief for the presentation was to present my own research to a lay audience attending UiAs popular dissemination series, Lørdagsuniversitetet.

When I started as a PhD at the University of Agder, and my friends and family would ask what I would be researching and when I answered “subtitling”, it turned out to be quite a good conversation starter. Turns out that since Norwegians have a lot of experience with subtitles (some estimate that the amount of subtitles we consume per year estimates 1400 pages of literature per year[1]) they also have quite strong opinions on the subject. But not just any subtitles, people rarely talk about a particularly well subtitled programme, but we seem to remember and love to talk about the bad subtitles. So, safe to say, some of my friends and famlily seemed almost disappointed when I told them that I would not be spending three years identifying mistakes in subtitles and pointing out to the translators what went wrong in the process and that they should get their act together.

Instead, I spend my time looking into getting more insight into how subtitlers deal with certain features of our language and if the finished subtitles can teach us something about how subtitlers process the information given to them through spoken dialogue, the images on screen and the overall narrative. In order to do this, I have chosen to focus on a fun little feature of language and thought that we call metonymy. Now, many of you have probably already heard of metaphor, and acquainted yourself with metaphor analysis in fiction and poetry, whether it was by your own free will or being forced to do so in school.

Although metaphor and metonymy are in many ways similar, the main difference is that whereas we use metaphor to create novel images comparing for example red cheeks to rose petals, in metonymy we use one part of for example an object to describe or make reference to that same object.

Such as the title of the movie Jaws in which we use part of the shark, namely it’s mouth, to describe the main antagonist of the movie. The jaws of a shark is particularly striking image that is bound to invoke a sense of foreboding or dread on audiences. One could also argue that «Jaws» is far more striking than if the movie had just simply been called «Shark», and this also shows us how metonymy can be effective as a rhetorical device.

Metonymy is not only restricted to written or spoken language, but also found in visual images. If we look at the poster for Jaws, we only see the head where the teeth are the most prominent feature, instead of the whole shark. And whenever you see an image of a shark fin gliding across the water surface and you think of the shark lurking underneath the surface, that is also an example metonymy. Commercials are also a good source of visual metonymies, e.g. when you see a commercial for fizzy drinks, we often see the image of the bottle or can, but the point isn’t for you to want to buy the bottle, but what’s inside of it. In other words the image of the can or bottle, invokes the contents of the container.

Returning to Jaws one last time: metonymy can also occur in sound, and John Williams’ chilling theme from the movie is a particular good example of this. The dissonant two chord theme has become so closely associated with the image of the shark, that instead of saying «shark» you could just sing that theme instead[2].

Where does metonymy come from and what allows us to use it in communication? Back in the late 70s and early 80s a group of linguists began to look deeper into how we use metaphor and metonymy in our daily language[3], and what they discovered is that we use it so frequently that we don’t even realise it most of the time. This tells us that both metaphor and metonymy is a central feature of our cognition and way of thinking.

Imagine that your brain is made up of lots of little boxes, each box representing some part of your knowledge about people, things, events, places and so on. Our ability to link this information with each other is what allows us to use things such as places and names as metonymies, think for example of how we say «Westminster» when referring to the British government or «Wall Street» as a name for several financial institutions. That’s because our knowledge tells us that these things are connected to each other, Westminster is where the government of Britain sits, and Wall Street is where the biggest banks are located.

Metonymy is grounded in our experience, when we meet and interact with people, objects, places, and events, our interaction with them allows us to use properties and parts of these to speak of the whole or even parts to describe parts. It helps people with shared interests and knowledge to communicate effectively. Think for example of people working in a specialised profession such as medicine, where doctors will use jargon and medical terms between themselves as a group, but they will most likely use a different type of language when talking to their patients. Think about your own work, how do you talk with your colleagues, or how do you talk with close friends? In social settings, metonymy can help us build relationships with people whom we share experiences with.

How is the knowledge of metonymy, how we use it and situations where it might occur, relevant to translation? Since metonymy is often closely tied to specific discourse communities, they do not necessarily travel well across linguistic and/or cultural borders, and they therefore become problems that a translator needs to solve. The translator is in one way the outsider that is trying to get access into the discourse community by figuring out what is said. To do this, they need to be able to access the same mental box that the metonymy originated from and make the proper connections in their own language

The point of subtitling is not to translate just what is being said, but what is the overall meaning of the story being told on screen, and this creates a contract of illusion where the viewer is able to follow the narrative on screen with as little awareness of the subtitles as possible.

In addition to paying attention to the dialogue, the subtitler must always keep in mind what is happening on screen, such as who is speaking; where are they; what tone of voice are they using; what kind of situation do they find themselves in? Since metonymy is such an integral part of our way of communicating with each other, it can often be difficult to make the distinction between literal or figurative meaning, and therefore the context becomes important to figure the intended meaning.

Because if there is a mistake (and this doesn’t just happen with metonymy and metaphor) then there is a disconnect between what the viewers hears and sees and what it says in the subtitle, and the contract of illusion between viewer and subtitler is broken. This is probably  why we notice the mistakes more than we notice the good subtitles. However, if the subtitler is able to piece the puzzle together correctly, then the job is done and hopefully, the viewer will be able to enjoy the programme without noticing the subtitles more than necessary.

To finish, I will illustrate how metonymy can be dealt with in subtitling with an example from the comedy series “Veep” (2012-2019).

Here is a screenshot showing the English dialogue:

Veep, S3 E8 ‘The Debate’ Copyright: HBO

In this example we have two instances of metonymy, first “park” which according to the dictionary (e.g. Collins Dictionary[4]) is used to refer to a baseball stadium. The second metonymy is the name “Shoeless Joe”, which is the nickname for the baseball player Joe Jackson who was central in a game fixing scandal in the 1920s. The whole exchange is a reference to the character being addressed being a former baseball player, and not a particularly good one. Since baseball is not a major sport in Norway, the subtitler therefore has to make the decision whether to keep the terminology and the nickname or choose a different approach, such as finding an equivalent in Norwegian or leaving the name out altogether.

Veep, S3 E8 ‘The Debate’. Copyright: HBO

As we see from the Norwegian subtitles, the subtitler chose to replace “park” with “tribunen” (lit. the stands) and decided to leave the nickname out altogether. My guess would be that most Norwegian viewers would have no problem with this translation, and would probably have wondered more if the name was kept in its original form or replaced with a Norwegian equivalent.


References

[1] NRK (2023) Maskinene har begynt å oversette film og TV. (https://www.nrk.no/kultur/gode-oversettelser-blir-oversett-pa-strommetjenestene-1.16199448. Accessed 26 Januar 2023)

[2] https://www.classicfm.com/composers/williams/music/jaws-theme/

[3] George Lakoff and Mark Johnson’s Metaphors We Live By (1980) is one of the most seminal works in this line of research.

[4] https://www.collinsdictionary.com/dictionary/english/park

A few thoughts on swearing in dubbing and subtitling

Over the Christmas break in December 2022, Netflix released their movie version of Matilda – The Musical, which has run for a number of years in London’s West End and on Broadway in New York. Based on Roald Dahl’s book by the same name, both the musical and the movie are aimed at families, with the movie version receiving a rating of PG (Parental Guidance) in the UK and the US and 7+ in Norway, meaning that children down to the age of five can watch the film when accompanied by an adult.

So, taking the main audience into account, one might expect not to find taboo language in either the dubbed or subtitled version. However, in one of the major musical numbers “The Hammer” the evil and narcissistic Ms Trunchbull sings about her determination to become a hammer throwing champion in her younger years and the English lyrics goes as follows:

“Do you think in that moment, when my big moment came
That I treated the rules with casual disdain?
Well? Like hell!”

In the Norwegian version, both the dubbed and the subtitled, the lyrics are as follows:

“Tror du i det sekundet da jeg fikk sjansen min
At jeg lett som en fjert kastet reglene hen?
Ja? Ikke faen!”

Which loosely backtranslates as: Do you think in that second when I got my chance/that I light as a fart threw the rules away?/Yes? No devil (or ‘the devil I didn’t)!

As we can see, the Norwegian translation can be said to be fairly close in meaning to the original text. In the second line casual disdain has been replaced with a more colloquial saying in Norwegian lett som en fjert. Such choices are most likely due to the translator having a focus on essential features of the dubbing process, where the need for lip, or phonetic, synchrony as well as temporal synchrony being of equal importance to the meaning (Chaume 2012)[1]. Also, since this is a song translation, the number of syllables are identical, in order not to disturb the rhythm of the musical soundtrack.

In this post I want to focus on the translation of the phrase “like hell”, which in both the Norwegian dubbed dialogue and the subtitles is translated with “ikke faen”. As can be seen from the loose back-translation, “ikke faen” is a close equivalent to “like hell” if we just look at the literal meaning. However, when we look at the pragmatic function of the English idiom “like hell” and the Norwegian “ikke faen”, then a more marked difference between the two appears.

Emma Thompson as the terrifying Ms Trunchbull. Copyright: Dan Smith/Netflix

According to Cambridge dictionary (dictionary.cambridge.org)[2], most swearwords and taboo expressions in English involve reference to religion or body and bodily processes, often of a sexual nature. In general, English taboo expressions such as “like hell”, which involves religion, are thought of as being weaker and less impactful than those which involve bodily action or functions, such as fuck or shit. Therefore, it might not be unreasonable to assume that to an English native speaker, an expression such as “like hell”, might be more casual and less offensive than other swear words, perhaps even to the extent that people will hardly notice it. Such taboo expressions are perhaps more commonly used for emotional emphasis, rather than to cause offense.  

In Norwegian, on the other hand, the strongest taboo expressions are associated with religion (Hasund 2006)[3], meaning that expressions that invoke references to hell and Satan are emotionally stronger. Even though Norway is a secular country, where a decreasing proportion of the population belong to a religious community, language norms governing the use of profanities still stand strong and using expressions such as “ikke faen” is still considered taboo in Norwegian culture (Fjeld 2014)[4].

Some might argue that a better equivalent to “like hell” would be “ikke søren” in Norwegian, where the male name Søren is used as a euphemistic derivative for Satan (naob.no)[5]. This would then fall under what Hasund (ibid.) refers to as nestenbanning (lit. almost swearing), and would have a similar pragmatic function to the source language idiom. However, this is where the intricacies of dubbing comes into play, because as already mentioned, dubbing not only has to take into account translation, but also lip synchronization and speed, the dubbed voice has to stop when the lips of the actor stops moving, in order to maintain the illusion that the actor is speaking in the dubbed language. Adding to that that the character is singing, it is also not desirable to add extra syllables to lyrics, as this will interfere with the rhythm of the song.

Another interesting point is that not only is the target language much stronger than the source language, it also makes the character, who is otherwise extremely concerned with discipline and self-control, seem uncharacteristically vulgar. Later in the movie there is a scene where Ms Trunchbull goes into a long tirade where she uses several profanities and strong language, but in that scene the fact that she loses control of her emotions is part of the plot.

This does not only happen in dubbed films, but also in other audiovisual settings the concept of hell and other taboo expressions and their pragmatic function in English is sometimes lost. In a subtitled version of Love Actually (2003), the line “Hello! What the hell are you doing here?” was translated as “Hei. Hva faen gjør du her?” in one Norwegian version (different broadcasters and streaming companies use different subtitle vendors, so depending on which streaming). While this is a literally correct translation of the expression “what the hell”, it cannot be said to be semantically accurate. Whereas “what the hell” in English is a pretty casual way of expressing surprise, “hva faen” is far more aggressive in Norwegian. Add to the fact that the scene shows a smiling and surprised man in a light and airy room, and the stark contrast between what is being said and what is written in the subtitles, becomes perhaps even more evident.


Pleasantly surprised man or angry man? While the picture indicates the former, the subtitles suggest the latter. Colin Firth in Love Actually (2003). Copyright: Universal Pictures/Viaplay

This is just one example, and it must be taken into consideration that this could be a result of machine translation, where the software is unable to pick up context and visual clues. In other Norwegian versions, the subtitles are toned down, with one version (NRK) using a non-offensive expression of surprise “Hva i alle dager gjør du her?” (lit. what in all days do you here?) and another (Amazon Prime) neutralizing the expression with the help of reduction of the source text: “Hva gjør du her?” (lit. what do you here?)

Do I think that Norwegian parents watching Matilda with their children will react to one of the characters saying “ikke faen” as part of a musical number? Not necessary, swearing in film and tv has become increasingly more common, even in programmes aimed at younger audiences, so it might not cause much offense among audiences. Incidentally, the actor voicing Ms Trunchbull also has a Northern Norwegian accent, and in Norway, people from the North are known to have a much stronger and colourful language, so this might also lead to people not noticing. A number of cases from the Norwegian broadcast regulator, Kringkastingsrådet[6],[7],[8], however, does show that swearing and taboo language, particularly in programmes aimed at younger audiences are one of the issues that garners most complaints from viewers.

Fjeld’s article from 2014 shows that there was a considerable increase in the number of swearwords and taboo expressions on tv in the period between 1998-2010, both in Norwegian language programming and in foreign language content. With the enormous growth of streaming platforms and online video content in the last decade, there’s no reason to believe that this hasn’t increased even more since Fjeld conducted her study. Also, as it becomes more common to use machine translations in subtitling[9], cases like the example from Love Actually, might become even more common. Recent studies point to the quality of subtitles involving machine translation being poorer than those done ‘manually’ by a human translator (Hagström & Pedersen 2022)[10].

If nothing else, as long as machines are not able to interpret context, emotions and other intertextual clues, the need for human agency in subtitling seems to be very much called for. For translators it seems to come back to the question of cultural awareness, meaning that in order for the translation to be correct, it is important to be aware of the cultural differences in language and that the pragmatic function of expressions in the source language might be quite different than similar expressions in the target language. Not least does it perhaps also show that for translators working with dubbing and subtitles, the context, such as scene, character and plot, is key for getting the tone of the translation just right, or else they risk breaking with viewers expectations and impressions of both characters and the narrative taking place on screen.


References

[1] Chaume, Frederic (2012) Audiovisual Translation: Dubbing. Manchester: St Jerome Publishing

[2] Cambridge Dictionary online. https://dictionary.cambridge.org/grammar/british-grammar/swearing-and-taboo-expressions. Accessed 20 January 2023

[3] Hasund, Ingrid Kristine (2006) Religiøs banning og nestenbanning. https://www.sprakradet.no/Vi-og-vart/Publikasjoner/Spraaknytt/Arkivet/2006/Spraaknytt_1_2006/Religioes_banning/. Accessed 23 January 2023

[4] Fjeld, Ruth Vatvedt (2014) Banning på tv: mer vanlig, mindre tabu. https://www.sprakradet.no/Vi-og-vart/Publikasjoner/Spraaknytt/spraknytt-2014/Spraknytt-22014/Banning-pa-tv-mer-vanlig-mindre-tabu/. Accessed 23 January 2023.

[5] NAOB Det Norske Akademis Ordbok online. https://naob.no/ordbok/s%C3%B8ren. Accessed 23 January 2023.

[6] «Kringkastingsrådet advarer NRK om banning» (2014) https://www.aftenposten.no/norge/i/gPM41/kringkastingsraadet-advarer-nrk-om-banning

[7] «TV-seere klager på banning i MGP Jr: -Udannet og grotesk» (2019) https://kampanje.com/medier/2019/06/tv-seere-klager-pa-banning-i-mgpjr/

[8] «Uenighet om banning i NRK» (2020) https://m24.no/banning-karoline-riise-kristiansen-kathrine-kleveland/uenighet-om-banning-i-nrk–vi-ma-tale-et-visst-niva-av-banning/271619

[9] «Maskiner har begynt å oversette film og TV» (2022) https://www.nrk.no/kultur/gode-oversettelser-blir-oversett-pa-strommetjenestene-1.16199448

[10] Hagström, H., & Pedersen, J. (2022). Subtitles in the 2020s: The Influence of Machine Translation. Journal of Audiovisual Translation5(1), 207–225.