In Sweden, we do it like this

On cultural references and subtitling norms

By Jan Pedersen (Stockholm University, Sweden)


What would the VA say if Al Roker was a Keystone Kop at Panmunjon? Subtitling is not just a matter of linguistic transfer; building bridges between cultures is every bit as important. This article is based on a subtitled translation of the episode of The West Wing which is the basis for this issue of inTRAlinea. The episode has been subtitled using established Swedish subtitling norms for television. These norms are of two kinds, partly technical, dealing with expected reading speed, subtitle density and condensation, and also translation-related. In this article the translation norms under discussion are those that govern the translation of extralinguistic cultural references (ECRs), i.e. references that are expressed verbally, but which refer to cultural items outside of language, such as names of people and places (like Al Roker or Panmunjon). A model for rendering such references in subtitled translations is presented; it consists of two parts: a taxonomy of translation strategies, and a series of parameters that influence the choice of translation strategy. This model is applied to the ECRs in the episode, using Swedish subtitling norms. The results are presented and complex cases are discussed further, as we find out how we can make a target audience understand the connotations of those bungling Keystone Kops.

Keywords: cultural references, subtitling, norms, translations strategies, west wing, audiovisual translation

©inTRAlinea & Jan Pedersen (2016).
"In Sweden, we do it like this On cultural references and subtitling norms"
inTRAlinea Special Issue: A Text of Many Colours – translating The West Wing
Edited by: Christopher Taylor
This article can be freely reproduced under Creative Commons License.
Stable URL:

Scandinavia has for good reasons been described as a bastion of subtitling (Ivarsson & Carroll 1998). In this part of the world, there is a very long and solid tradition of subtitling, with other forms of audiovisual translation (AVT) being marginalized. Dubbing and voice-over are almost exclusively used for pre-literate audiences, and while there is some versioning of off-screen narration, subtitling is so prevalent as to dwarf any other mode of interlingual AVT. For further details on this, and a discussion of AVT for accessibility purposes, the reader is referred to Pedersen (2010a).

Subtitling is a norm-governed activity and the norms that govern Swedish subtitling are laid out in guidelines and other prescriptive documents, and they have also been investigated in a major study (Pedersen 2007a; cf below). This study conclusively showed that Scandinavian subtitling norms are converging, so that Swedish subtitling norms for television are very similar to those of its neighbouring countries. The study also produced a descriptively based model for how cultural references are rendered. This article tests that model by applying it, and other professional subtitling norms, to The West Wing episode ‘In Excelcis Deo’. Rather than investigating the subtitles that were used when the episode was aired and judge to what extent they comply with the norms, new professional subtitles were created by the present author, who apart from being a translation scholar, is also a professional subtitler. The reason for this is partly that such investigations have already been carried out (cf. Pedersen 2011), and partly that this approach gives full access to the subtitling process and tests how the model works as professional tool and it also illuminates the reasoning behind any and all of the complicated translation decisions that are made by the subtitler.

When a proficient translator goes about his or her task, there are stretches of text that are fairly smooth and straightforward, and where the processes involved call for what Lörscher (1991: 81) calls non-strategic behaviour. There are also, however, text features that call for strategic behaviour; features that need the translator’s full attention and where she or he will have to carry out more conscious decision-making. I call these features ‘translation crisis points’ (TCPs; cf. Pedersen 2007a: 89 ff or Remael & Vercauteren 2010: 158 ff). Examples of these in the text that this issue of inTRAlinea is concerned with are puns and other witticisms, high-speed dialogue, the semiotic interplay between dialogue and the visual signs and so on and so forth. One very striking type of TCP is the ubiquitous cultural references that are found in the text. The rendering of cultural items is the main focus of the analysis, and the translation that was created for this article. Hence, after a preliminary section on technical norms in Scandinavia, the rest of this article is devoted to the norm-guided rendering of potential verbal cultural translation problems that I call ‘extralinguistic cultural references’ (ECRs).

Technical considerations for this part of the world

Technical subtitling norms used to differ substantially in Scandinavia. This is due to the fact that for a long time, public service companies were the sole providers – and commissioners – of subtitles, and they developed their own styles, norms and software. The development was not completely dissimilar among the neighbouring countries, but the norms developed in parallel, rather than as a result of much cooperation. It was not until the advent of de-monopolization in the early nineties that the Scandinavian norms started to converge. Multinational TV channels used multinational subtitling agencies that saw the whole of the region as one market. It made sense for these new agents to harmonize the norms for the region, in order to benefit from joint tasks, such as using centrally cued master template files for the subtitled translations. This can be seen to have had an impact on the Scandinavian mediascape from the middle of the nineties, and after the new millennium some of the new norms were used in public service broadcasting as well, as the national broadcasters started to outsource their subtitling. Even though minor differences may still be found, the three important norms described below have converged almost completely (Pedersen 2007a: 65ff).

Expected Reading speed

This is the speed with which the reader is expected to read a subtitle, rather than the speed with which a reader actually reads a subtitle, on which there is precious little research. It used to be the case that the Swedish national broadcaster, SVT, worked with a very low reading speed in order to make the subtitles accessible to every single viewer, and to some degree, this is still their vision. However, just like Danish and Norwegian subtitles, Swedish TV subtitles nowadays adhere to the 12 cps rule, which says that a viewer is expected to read 12 characters (including blank spaces and punctuation marks) per second. This leads to exposure times of three seconds for a subtitle of one line (with 36 characters, which is normal), and 5­–6 seconds for a two-liner, as these can be read proportionately faster. The exposure times are getting shorter, however, and in DVD subtitling, it is not uncommon to see reading speeds of 16 cps or higher. The reading speed is supposed to be an absolute norm that should not be affected by the pace of the dialogue, but in practice, it is common to see lower subtitle exposure times, and thus higher expected reading speeds during very rapid dialogue, such as the opening scene in ‘In Excelsis Deo’.

Condensation rate

Condensation (or reduction) rate is a measure of how much shorter a subtitled text is as compared to its source text, using a simple word count. Sweden started to use the electronic time code for cueing fairly late, and that meant that manual cueing was still a common practice in Sweden until the mid- to late 1980s, when Denmark had been using electronic cueing for a decade already. Manual cueing is not as exact as electronic, which meant longer exposure times and more condensation of the verbal message. This difference has now almost completely disappeared, as shown in Pedersen 2007a, where no significant difference was found between Sweden and Denmark. The prevailing norm is now that about 30% of the original dialogue gets ‘lost in translation’. Losing 30% of the dialogue does not equate to losing 30% of the information being transmitted, however, given that much dialogue consists of redundant oral features and much of the information is also communicated in other ways via the audio and visual channels. The condensation rate is of course affected by the pace of the dialogue in the TV programme, and for The West Wing, with its sometimes very rapid dialogue (as in the aforementioned opening scene), some stretches of the text will have a higher condensation rate. On the other hand, there are slower passages, particularly in Toby’s quest for the homeless veteran, where there is hardly any condensation at all.

Subtitle density

The subtitle density is the number of subtitles in a film or TV programme in relation to its length. This is the only area where a significant difference can still be found between the Scandinavian norms, even though the difference is much smaller than it was as late as 1995. Typically, a Danish film would have 850 subtitles per film; whereas there would be fewer than 650 in the Swedish version of the same film. The difference is now much smaller, but a Danish target text (TT) will still typically have about 10 per cent more subtitles than a Swedish one (Pedersen 2007a:78). This difference is mainly due to the fact that Danish subtitlers tend to respect cuts in the source text (ST) more. This leads to a situation where Danish subtitlers tend to have more rapid one-liners, whereas the Swedish subtitling norm calls for more full two-liners. The harmonization of subtitle density, which imposed heavy block subtitles on Danish viewers, is what turned out to be the most unpopular of the changes that took place in the 1990s. The subtitling of The West Wing episode was done according to the Swedish norm, and thus contains quite a few full two-liners.

Extralinguistic Cultural References (ECRs)

When subtitling a text which is as culturally embedded as The West Wing, it quickly becomes apparent that more than linguistic transfer is called for. Cultural mediation is at least as important here, and the focus of this chapter is on describing the process of cultural mediation in subtitling when it comes to extralinguistic cultural references (ECRs). This process can be hard to investigate empirically, as it involves what is often un- or semiconscious decisions in the subtitler’s mind. In the present article, however, the subtitler’s decisions are available to the analyst, as the two are one, and the subtitles where created especially for this article. There is another way of gaining access to these decisions, however, which involves empirical investigation of subtitling products rather than the subtitling process. Granted, this does not give direct access to the process, but if the data is comprehensive enough, the results can be very reliable, as explained below.

I have developed a model for analysing the process of how ECRs are rendered in subtitling and why they are rendered in the ways that they are, and this also works the other way around. The model that helps researchers describe how ECRs are rendered can also be used as a resource for subtitlers to find out how ECRs can be rendered. The model was developed within the paradigm of descriptive translation studies and is thus empirically based. This means that, instead of prescriptively telling subtitlers how to go about their business, the model is based on a description of how actual subtitlers go about their business. This can then be used for instruction as to how the business of subtitling is performed, thus using empirical description as the basis for didactic instruction, and this is the main point of this article.

The model is based on the Scandinavian Subtitling project (Pedersen 2007a), which involved a corpus of 100 Anglophone films and TV programmes and at least one version each of the Danish and Swedish (plus quite a few Norwegian) subtitled translations of said texts. The texts come from many different genres, fiction as well as non-fiction, and from a mix of commercial and public service broadcasters. The findings of this project led to the model applied below. The model is tripartite: it contains a definition and delimitation of ECRs, a set of parameters that influence how ECRs are rendered and a taxonomy of strategies for rendering them. It could be seen as a tool kit for solving culture-based translation problems.

The model was used in Pedersen 2007a to reconstruct Scandinavian subtitling norms for translation shifts, using coupled pairs analysis (Toury 1995: 38).These norms say that Scandinavian subtitles tend to be source-oriented, though not extremely so, and they also give fairly clear indications on which translation strategy to use under which circumstances (Pedersen 2007a: 250-265), as will be outlined below. References to Scandinavian subtitling norms below are thus based on that study.

ECRs in general and in The West Wing

Extralinguistic Cultural Reference (ECR) is defined as reference that is attempted by means of any cultural[1] linguistic expression[2], which refers to an extralinguistic entity[3] or process. The referent of the said expression may prototypically be assumed[4] to be identifiable to a relevant audience[5] as this referent is within the encyclopaedic know­ledge of this audience. In other words, ECRs are references to places, people, institutions, customs, food etc. that you may not know even if you know the language in question. (Pedersen 2007a: 91)

The episode of The West Wing analysed here contains an abundance of ECRs, from several domains, and they have been rendered in accordance with Swedish TV norms. The main domains involved are obviously government, but also professional titles, entertainment, weights and measures, and due to the President’s visit to a rare bookstore, literature is also a noticeable domain. Typically, when dealing with ECRs, names make up the lion’s share of the data, and this episode is no different; there is an abundance of personal names, but also many geographical and institutional names. It should perhaps be mentioned that it is fairly common for an ECR name to belong to more than one domain. For instance, José Feliciano (1:07)[6] is a personal name, but he also belongs to the entertainment domain.

For a viewer to make sense of an utterance that contains an ECR, the viewer must be able to access the ECR in some way, i.e. understand to whom or what the ECR refers. By definition, an ECR cannot be accessed through linguistic knowledge alone; instead, a viewer can access an ECR in one of three ways:

  1. Encyclopaedically and intertextually, that is through the viewer’s cultural literacy of the world and other texts.
  2. Deictically, that is through deixis in the context or co-text.
  3. Through intervention from the subtitler, working as a cultural mediator. (Pedersen 2010b)

It could be argued that the greatest concern for the subtitler is to ascertain when i) and ii) do not apply and s/he will have to step in and actively guide the viewer. We will return to this issue presently, but let us first look at the tools that a subtitler has at her or his disposal for rendering ECRs.


The subtitler has basically seven ways of rendering an ECR. Three of these are minimum change strategies, and these are used when the referent of an ECR can be accessed either encyclopaedically or deictically. The minimum change strategies only involve surface-structure changes; no semantic material is added or removed. There are three more, which can be termed interventional, and which are used when a subtitler has decided to assist the viewer in accessing an ECR. Finally, the strategy of omission can be seen as being neither minimum change nor interventional. The strategies can be seen in the simplified taxonomy tree in figure 1, below.

Figure 1: Simplified taxonomy tree of ECR transfer strategies (based on Pedersen 2007a: 154)

Retention is by far the most common way of rendering an ECR in Scandinavia (Pedersen 2007a: 201). It means just transferring the ECR to the subtitles, with no, or only small adjustments to meet target language expectations, e.g. Lowell Lydell (8:30) rendered as ‘Lowell Lydell’.

Direct translation is not nearly as common as retention, but it is frequently used on ECRs that lend themselves to translation in Scandinavia (Pedersen 2007a: 210). In the text, there is the name of the bookstore Rare Books (19:38) which is translated directly into ‘Sällsynta Böcker’ in the Swedish subtitles.

Official Equivalents are not so much true strategies as prefabricated equivalents that subtitlers are required to use, unless circumstances make it impossible. A typical example would be rendering 83 degrees [Fahrenheit] (13:52) as ‘28 grader’ [centigrade], as the Celsius scale is officially adopted in Scandinavia. An official equivalent can be based on any of the other strategies in the taxonomy or on something completely different. The point is that the equivalent has either become entrenched through established usage or brought into existence through some administrative decision. An example of the latter is the title of the series (that is The West Wing (0:05)) which has to be rendered as ‘Vita Huset’ [the White House] in Swedish, as the Swedish distributors have decreed that this is what the series is to be called in Sweden.

Generalization means replacing the ST ECR with something that is more general, either by using a superordinate term (typically a hypernym or a meronym) or by using a paraphrase. The result is always something that is less specific and it often leads to what Leppihalme calls ‘reduction to sense’ (1994: 125). Generalization is mainly used for two reasons. It is either used as an interventional strategy to guide the viewers, or it is used to save subtitling space, whereby long and cumbersome phrases may be rendered by a briefer and more general TT solution (Pedersen 2007a: 212). Sometimes these two reasons are combined, as in example (1) where Toby laments the lack of respect the veteran gets:

TOBY: An hour and twenty minutes for the ambulance to get there. A Lance Corporal, United States Marine Corps, Second of the Seventh. I got better treatment at Panmunjong.

Ambulansen kom efter 80 minuter.

Han var marinkårskorpral. Jag blev
bättre behandlad i själva kriget.

Back translation:
The ambulance arrived after 80 minutes.
He was a Marine Corps Corporal.
I was treated better in the war itself.

The rather lengthy and complicated military ECR a Lance Corporal, United States Marines, Second of the Seventh was generalized as ‘marinkorpral’ [Marine Corporal]. This solution leaves out the details of the military references that would be hard for the TT viewer to process, and it also saves space. It also, of course, removes much of the local flavour of the ST, but then again, to render the connotations of the heroics carried out by the Second Battalion of the Seventh Marines within the brief space of the subtitle is a virtually impossible task. We will return to the other ECR in (1), i.e. Panmunjon, in example (6) below.

Specification is the complete opposite of generalization. Instead of ‘chunking up’ (Katan 2004: 199) the subtitler ‘chunks down’ and makes the TT message more specific than the ST message. This could be done for reasons of idiomaticity, as when holiday cheer (6:17) becomes ‘julglädje’ [Christmas cheer], but the strategy is more importantly used to aid the viewer in understanding the relevant function or connotations of the ST ECR. An example of this would be when Stephen J. Gould (1:10) is rendered as ‘forskaren Stephen J Gould’ [the scientist (or researcher) Stephen J Gould]. This is a very felicitous strategy for adding the information that might be lacking in the encyclopaedic knowledge of most TT viewers, without subtracting any ST information. Unfortunately, it is also very space-consuming, which means that it is a rarely used ‘luxury’ strategy, particularly in comparison with generalization, which aids the viewer and saves space, but at the cost of information loss. Scandinavian subtitling norms thus favour generalization rather than specification (Pedersen 2007a: 209).

Substitution means that the subtitler does not transfer the ECR at all, but substitutes it with something else. The replacing item can be a different ECR, either a similar one from the target culture (TC) or a more well-known ECR from the source culture (SC) or one from a third culture; this is called cultural substitution. It can also be something completely different that just works in context; this is called situational substitution. The latter option is basically a last resort when all other strategies would fail, and the subtitler just tries to come up with coherent subtitles. Fortunately, this is an extremely rare strategy in Scandinavia (Pedersen 2007a: 216). Cultural substitution, on the other hand, has a long history in Scandinavia and used to be a common way of rendering ECRs, particularly in Denmark (cf. Pedersen 2007b). These days in Sweden, the use of cultural substitution is limited to a few domains, one of which is government, and it is thus possible to replace the Secretary of Labor (14:45) with ‘Arbetsmarknadsminister’ [literally ‘labour market minister’]. This is the minister responsible for the work of the Labour Department in Sweden and thus a minister with similar tasks to that of the ST ECR.

Omission can, according to Leppihalme (1994: 93), be used responsibly, after testing and rejecting all other options, or irresponsibly, for lack of trying or caring. It should be pointed out that using omission responsibly is the only ethical way of translating. The media-specific constraints of subtitling make this a not uncommon strategy, particularly when the dialogue is very high-paced. For example, in the very rapid-fire dialogue in the opening scene, when the staff are discussing the preparations for the White House Christmas celebrations:

MANDY: Now, we have José Feliciano, we have Sammy Sosa and his wife...
SAM: Did you know that recordings of Feliz Navidad outsold recordings of White Christmas?

-Vi har José Feliciano och Sosa.
-Folk älskar Féliz Navidad.

Back translation:
-We have José Feliciano and Sosa.
-People love Feliz Navidad.

The only way of creating coherence in this fast-paced dialogue, is by omitting the ECR White Christmas and focussing on the more important (and apparently more popular) ECR Feliz Navidad, as can be seen in example (2) above.

These seven strategies are basically the subtitler’s tool-kit for dealing with ECRs. Just as hammers are not used to insert screws, however, certain strategies are only used with ECRs from certain domains under certain circumstances if the result is to be felicitous. On the other hand, just as a hammer actually can be used to insert a screw, if you do not care too much about the result, the ECR transfer tool kit can be used just as carelessly. However, if quality is to be a priority, certain circumstances will have to be taken into account. I call these circumstances influencing parameters, as they influence the choice of strategies used.

Influencing parameters

Just as the strategies above explain how an ECR can be rendered, the influencing parameters explain why an ECR is rendered in a certain way. The influencing parameters all deal with various aspects of ECRs, the medium and/or other aspects of the translation situation. They are intertwined and each can work for or against the subtitler in a given situation. However, knowing about them helps the subtitler make an informed decision when it comes to the rendering of ECRs. Just as for the strategies above, the influencing parameters are based on empirical description, and they can also be used for didactic instruction, which is the case here.

Transculturality refers to the familiarity of the ECR. In other words, it is a way of gauging how well-known an ECR is to the ST and TT audiences. Transculturality works on a cline, from ECRs that are virtually unknown, to ECRs that are virtually universal. However, for practical reasons, it is helpful to divide the cline into three parts: transcultural, monocultural and infracultural. Transcultural ECRs are ECRs that most people know about in the SC and the TC (it is irrelevant whether the ECRs are known in other cultures as well) e.g. Christmas (4:57), to use a rather obvious example. Infracultural ECRs are generally known by neither of the two audiences involved; examples of these could be any of the book-binding terminology ECRs that the President shows off at the rare bookstore. Infracultural and transcultural ECRs do not normally cause translation problems, as they would either be accessible through the audience’s encyclopaedic knowledge (transcultural ECRs) or would have to be made accessible deictically in the ST, as the ST writers cannot expect their primary audience to be aware of them either (infracultural ECRs). Monocultural ECRs, on the other hand, may cause translation problems as these are ECRs that are known to the ST audience, but not to the TT audience, and these are the ones that should be made accessible ‘translatorically’, i.e. by making use of interventional strategies. Gauging transculturality can be difficult at times (for a longer discussion on this, please see Pedersen 2010b), but it is important to do it properly, as you either patronize your audience, if you explain too much, or leave them in the dark, if you explain too little. The ST can sometimes help when it comes to this, as it can have clues to the original author’s transculturality appraisal, in other words how well-known the author has thought a particular ECR to be to his primary audience. A good example of this is (3) below, which represents the conversation that young Charlie has with Mrs Landingham about her twin boys, who were drafted and subsequently killed in Vietnam:

MRS LANDINGHAM: They went off to medical school together, and then they finished their second year, and of course their lottery number came up at the same time.
CHARLIE: For the draft?

The function of Charlie’s remark in example (3) is to elicit some elucidation for him and probably those members of the ST audience who are too young to remember the system of drawing a lottery number based on birthdays that was used for induction into active duty in Vietnam. This indicates that the ECR lottery numbers is on the border between monocultural and infracultural from a Swedish perspective, which means that it is certainly not transcultural, and should thus be treated accordingly. In this case, the ECR is made accessible through the co-text so there is no real problem here.

Extratextuality deals with the question of whether an ECR exists outside the text (or series of texts) at hand. If an ECR is text internal, it refers to something that has been specially created for this TV series, in our case, e.g. President Jed Bartlet. If an ECR is text external it has a life of its own outside the series, even though it may still be fictional. This distinction is important in that text internal ECRs do not normally cause translation problems as they have no real connection to reality, whereas text external ones have, and that limits the ways in which such ECRs can be rendered.

Centrality is a way of expressing how important an ECR is to the text at hand, and this works both on the micro level and the macro level. Since the main storyline in our ST is about Toby finding the dead tramp and his subsequent quest to have him buried with dignity, the ECR Korean War Vet (8:56) is central on the macro level and it is very important to ensure that this and other connected ECRs are accessible to the TT audience. If an ECR is peripheral on the macro level, a subtitler has more freedom of choice in the rendering of it, unless it happens to be central on the micro level, i.e. important for local level discourse. Furthermore, if it is peripheral on the micro level as well, how it is treated is no longer very important. This is illustrated in example (4), where the President asks Josh to join him in his shopping excursion to the rare bookstore and Josh replies:

JOSH: An hour with you in a rare bookstore? Couldn't you just drop me off the top of the Washington monument instead?

En timme med er i bokhandeln?
Kan ni inte döda mig istället?

Back translation:
An hour with you in the book store?
Couldn’t you kill me instead?

The ECR in (4), the Washington Monument, is peripheral even on the micro level, serving only as a colourful ingredient in Josh’s sarcastic reluctance to join his boss. Since the pace of the dialogue is rapid, I have deleted the ECR here and rendered the reply as the Swedish equivalent of ‘Couldn’t you kill me instead?’, which is admittedly less colourful but much shorter and has the same function in the conversation.

Polysemiotics (cf. Gottlieb 1997) is a term for the interplay between the verbal audio (that is the dialogue), the non-verbal audio (that is sound effects and music), the verbal visual (captions and other relevant on-screen text), and the non-verbal visual discourse channels that make up the polysemiotic texts that subtitlers translate. The polysemiotics can help or hinder the subtitler. If an ECR is shown on screen, for instance, it would not do to omit it or replace it with something else. In our ST, there is a great deal of interplay between the dialogue and the image, which influences the rendering of almost every ECR.

The co-text is actually part of the polysemiotics, but it is extra important as it has to be coherent, and can also be useful in making an ECR accessible, as we saw in example (3), lottery number. Also, if an ECR has been rendered accessible once, it can be rendered through retention on subsequent appearances in the text.

Media-specific constraints come in two kinds for subtitling. First, there is what Gottlieb calls semiotic jaywalking (2001: 16) in that the translation normally goes from source language spoken form to target language written form in subtitling. Second, there are the time and space constraints. These were explained above under technical considerations, so there is no need to reiterate them here. Suffice it to say that the constraints are often severe in the highly-paced dialogue of our ST.

The subtitling situation (Pedersen 2011: 115) gives rise to aspects that are not in the text, but rather about the text. These can be found through asking clusters of questions about various aspects, such as translation norms (both national and local), broadcasting (high prestige broadcaster? primetime TV?), TT audience (age group? level of education? expert or general?), the ST (genre? skopos? style? register?) and pragmatic aspects (deadlines? salary?). Even though the last one tends to be very important in reality, it is of little importance here, as the subtitles were produced as part of the preparations for the present article. Instead, questions about broadcasting and TT audience get priority here. The West Wing was aired in Sweden by the national public service broadcaster at prime time, which means that the subtitles should be of the highest quality, which they undoubtedly were, even though they are not used as the base for the present article. Thus, cutting corners by not researching ECRs should not be an option here. The TT audience probably consists of people of relatively high education and reasonably high age, who have an interest in American politics. They could be assumed to be familiar with many of the ECRs in the series. This means that when the President says that his daughter ‘Zoey is starting Georgetown [sic!] in two weeks’ the ECR Georgetown (22:41) can be retained in the subtitles, as the Swedish audience may have heard of this ECR, and there is also some guidance from the co-text, since the structure of the utterance makes it clear that it refers to a university.

What would the VA say if Al Roker was a Keystone Kop at Panmunjon?

The model above will help to solve any ECR-related translation crisis point, and most of them will be solved very easily. Under the circumstances laid down by the parameters, and given the domain of the ECR, the national norms will in most cases give the subtitler a sense of which strategy to use, or at least help to narrow the choice down to a couple of strategies. However, there are a few cases where the answer is less obvious, and these are the most interesting ones. In our ST, I found four such ECRs, and they are: the VA (4:39), Panmunjom (38:55), Al Roker (0:59), and The Keystone Kops (33:42). I carried out a reception study/brainstorming session with 21 English linguistics and/or translation scholars at Stockholm university, who had a crash course in using the model and then came up with solutions to these ECRs, and their (and my) solutions form the basis for the discussion that follows.

The VA

Toby Ziegler is called away to identify the dead body of a homeless man found on a park bench. Toby does not know the man, but is dismayed by the delay in removing the body, and by the apparent indifference of the authorities, represented by an officer from the DC police.  Part of their conversation goes like this:

TOBY: And then you’re gonna call the VA, right?
TOBY: Tattoo on his forearm is Marine Battalion Second of the Seventh. This guy was in Korea.

Sen kontaktar ni väl veteranbyrån?

Hans tatuering är från marinkårens
andra bataljon. Han slogs i Korea.

Back translation:
Then you will contact the veterans’ agency, I presume?
His tattoo is from the second battalion
of the marine corps. He fought in Korea.

It does not take a great deal of research to find out that the monocultural VA (which has no direct explanation in the co-text) is an abbreviation of the Department of Veterans Affairs, thus fully identifying the ECR. The problem for the subtitler is how this should be rendered in the Swedish subtitles. According to Swedish norms, government domain ECRs are normally rendered through cultural substitution (Pedersen 2007a: 217), using a similar TC ECR. However, since Sweden has not been at war for some 200 years, there is no similar organization in Sweden. It was suggested that generalization be used, so that the subtitle simply said ‘the authorities’, which is also an option according to Swedish subtitling norms (cf. Pedersen 2007a : 263-264). That might prove a confusing solution, however, given that ‘the authorities’ have already been notified and there is a police officer and a man from the White House there. Another solution could be to use the Swedish Department of Defence as a substitute. The solution that was finally decided on was to construct an ECR through paraphrased translation, i.e. ‘veteranbyrån’ in example (5) above.


Toby finally manages to arrange a military funeral for the homeless man, and has to explain why to the President. He does this by explaining how disrespectfully the homeless man was treated, as we saw in example (1), above. This is reproduced here as example (6) for ease of reference:

TOBY: An hour and twenty minutes for the ambulance to get there. A Lance Corporal, United States Marine Corps, Second of the Seventh. I got better treatment at Panmunjong.

Ambulansen kom efter 80 minuter.

Han var marinkårskorpral. Jag blev
bättre behandlad i själva kriget.

Back translation:
The ambulance arrived after 80 minutes.
He was a Marine CorpsCorporal. I was
treated better in the war itself.

The military ECR in this example has already been explained earlier in example (1). The problematic ECR here is instead Panmunjom, which admittedly is a Korean ECR, but it is also a monocultural SC ECR, because of the American involvement in the Korean War. It is used here to imply that American soldiers were treated better in a war zone than back home, which is basically the point of this whole storyline. In the transcript, the ECR co-text erroneously says ‘the guy got better treatment at Panmunjom’, but that is clearly not what comes through the verbal audio channel. There are two problems here: i) Toby is too young to have fought in Korea (and he also tells the homeless man’s brother that he has not been there) ii) Panmunjom is the site of the truce talks in Korea’s demilitarized zone, so the connotations of active fighting are not very strong. The ST is thus contradictory. Did the actor (Richard Schiff) bungle his line, or did the scriptwriters (Aaron Sorkin and Rick Cleveland) not care too much about veracity here, and wanted to give the impression that Toby had in fact been to Korea, and then chose Panmunjom as the most recognizable Korean War ECR? That issue is impossible to resolve. The problem can, however, be solved from the subtitler’s view, by using generalization, and letting the subtitle read ‘Jag blev bättre behandlad i själva kriget’ [I got better treatment in the actual war].

Al Roker

In the fast-paced opening scene alluded to above, the White House staff discusses the Christmas arrangements and the following conversation comes up just before the one in example (2) above:

SAM: Who’s playing Santa?
MANDY: Al Roker.
SAM: (raising eyebrows) Playing Santa?
MANDY: What’s wrong with that?
SAM: (slight pause) Went on a diet.
TOBY: How do you know these things?
SAM: I read.
MANDY: We’ll pad him if we have to.

-Vem ska vara tomte?
-Al Roker. Varför inte?

-Han har bantat.
-Då får han ha lösmage.

Back translation:
-Who will be Santa?
-Al Roker. Why not?
-He has dieted.
-Then he will wear a false belly.

The monocultural ECR Al Roker (famous African-American TV personality and NBC weatherman) is central on the micro-level, as it is used for the joke, and also as the centre of this rather lengthy, if rapid, conversation. The joke is based on Sam’s non-verbalized objection that it would be odd to have an African-American man playing Santa. Mandy’s question is then either truly naïve or is daring him to verbalize his bigotry, and Sam finally saves face by hinting that Al Roker is now too thin to be Santa. The most obvious Swedish solution in a case like this would be to use specification to add the information that the TC viewers do not have. However, choosing to specify ‘den svarte Al Roker’ [the black Al Roker] would clearly have been unacceptable. Generalization through the use of a superordinate term would not work either, for the same reasons. A paraphrase in Sam’s third line with something like ‘inte för att han är svart, utan för at than är för smal’ [not because he’s black, but because he’s too thin] was suggested, and that has some merits. However, that also verbalizes the ‘black’ objection, which is something that would spoil the joke, albeit less bluntly, and also, the media-specific constraints would not allow it. It was also suggested that cultural substitution be used, and that Al Roker be replaced by some other portly African American media personality who would be more transcultural, e.g. Forest Whittaker, James Earl Jones or Oprah Winfrey. The last one of these would probably be the most felicitous, as the joke would also be enhanced by adding sexism to the racism that is its unstated basis, plus the fact that Oprah Winfrey is known for her yo-yo dieting. There is of course the polysemiotic problem of the feedback effect from the original (Gottlieb 1997:93); the viewers do not hear Oprah Winfrey, they hear Al Roker. On the other hand, the pace of the dialogue is very rapid, and even native English speakers have problems keeping trace of everything that is said. There is also no credibility gap, since Winfrey is also a SC ECR, whereas it would have been contrary to Swedish norms to use a TC ECR here. Had I  been subtitling into Danish, I would not have hesitated to use Oprah Winfrey, as the norms are traditionally more open to this sort of substitution there (for a further discussion this, cf. Pedersen 2007b). But it is more in line with Swedish subtitling norms to just use retention (Pedersen 2007a: 259), and let the viewers fend for themselves. The ECR is, after all, not exclusively monocultural and it is also peripheral on the macro-level.

The Keystone Kops

After failing to carry out an elaborate blackmail plan to save the reputation of White House Chief of Staff, Leo McGarry, Sam and Josh receive a severe reprimand from McGarry, who is not happy with their behaviour:

LEO: Like I don’t have enough problems without the Keystone Kops.

Jag har problem nog,
utan era klumpiga upptåg.

Back translation:
I have enough problems,
without your clumsy stunts.

The monocultural ECR the Keystone Kops refers to characters in a series of silent films from the 1910s. The Keystone Kops were notorious bunglers, and the phrase is often used in the US to refer to people who is incompetent. The ECR is clearly monocultural, as the Keystone Kops films were never aired in Scandinavia, and unlike e.g. Laurel & Hardy or Charlie Chaplin, the Kops are completely unknown to the vast majority of Swedes. Retention would thus be a misleading strategy here, since the ECR is used metaphorically. Substitution is an option, like using ‘Laurel and Hardy’, who are transcultural also and also express the idea of someone who is less than competent. It was also suggested at the brainstorming session mentioned above that ‘Kling & Klang’, the two incompetent policemen from Pippi Longstocking be used for the substitution (which has the added bonus of keeping the police reference and the alliteration). Unfortunately, Swedish norms do not allow that sort of domestication these days (Pedersen 2007a: 238). Instead, generalization through the use of a paraphrase was used, as this is more in line with the current Swedish norms. Current guidelines tend to recommend generalization or praphrase in cases such as this, and that is also in line with my own results (Pedersen 2007: 212).


This article has shown Swedish subtitling norms in action when used on the present episode of The West Wing. The subtitling was carried out in accordance with the technical norms and the solutions were all based on subtitling norms that are active in Sweden today. Thus, retention has been used most frequently on the monocultural ECRs, and then generalization, substitution, omission, direct translation, substitution, and specification in that order in accordance with Swedish subtitling norms (cf. Pedersen 2007a: 201). This illustrates how the model can be used not only as a tool for analysis, but also as a working tool for subtitlers. The strategies give the subtitler the subtitler the means by which to solve ECR-related problems and the parameters, in conjunction with the knowledge of subtitling norms, help the subtitler produce translation solutions that are felicitous and which will be accepted by her or his audience.

The article has shown that there are always ways of solving translation problems caused by cultural references, even though not all connotations can be transferred on all occasions. If the subtitler is aware of the whole range of strategies available, and also considers the circumstances under which the problem appears, a well-informed decision can be made. The fact that some references are harder than others to render in a felicitous way should not be seen as a problem, but instead as something that makes subtitling the interesting task that it is.


Gottlieb, Henrik (1997) Subtitles, Translation & Idioms, Copenhagen, Center for Translation Studies, University of Copenhagen.

Gottlieb, Henrik (2001) Screen Translation: Six Studies in Subtitling, Dubbing and Voice-Over, Copenhagen, Center for Translation Studies, University of Copenhagen.

Ivarsson, Jan and Mary Carroll (1998) Subtitling, Simrishamn, TransEdit.

Katan, David (2004) Translating Cultures: An Introduction for Translators, Interpreters and Mediators (2nd edition), Manchester and Northampton, MA, St Jerome.

Leppihalme, Ritva (1994) Culture Bumps: On the Translation of Allusions, English Department Studies 2, Helsinki, University of Helsinki.

Lörscher, Wolfgang (1991) Translation Performance, Translation Process, and Translation Strategies: A Psycholinguistic Investigation, Tübingen, Gunter Narr.

Pedersen, Jan (2007a) Scandinavian Subtitles: A Comparative Study of Subtitling Norms in Sweden and Denmark with a Focus on Extralinguistic Cultural References, Doctoral Thesis, Stockholm University, Department of English.

Pedersen, Jan (2007b) “Cultural interchangeability: The effects of substituting cultural references in subtitling”, Perspectives. Studies in Translatology 2007:1, 30 – 48.

Pedersen, Jan (2010a) “Audiovisual Translation – In General and in Scandinavia”,  Perspectives: Studies in Translatology, 2010:1, 1 – 22.

Pedersen, Jan. (2010b) “When do you go for benevolent intervention? How subtitlers determine the need for cultural mediation”, Díaz Cintas, Jorge, Anna Matamala & Josélia Neves (eds.) New Insights into Audiovisual Translation and Media Accessibility, Amsterdam, Rodopi, 67 – 80.

Pedersen, Jan (2011) Subtitling norms for television: an exploration focusing on extralinguistic cultural references, Amsterdam/Philadelphia, John Benjamins.

Remael, Aline & Gert Vercauteren (2010) “The translation of recorded audio description from English into Dutch”, Perspectives: Studies in Translatology, 18: 3, 155 — 171.

Toury, Gideon (1995) Descriptive Translation Studies – and Beyond, Amsterdam/Philadelphia, John Benjamins.


[1] In a very wide sense of the word, including e.g. geographical names.

[2] Regardless of word class, syntactic function or size.

[3] Including fictional ones.

[4] As implied in the speech situation.

[5] E.g. a TV programme’s primary target audience.

[6] In this chapter, examples from the ST will be given in italics with a time reference in minutes and seconds.

About the author(s)

Jan Pedersen was educated at the universities of Stockholm, Copenhagen and Uppsala. He received his Ph.D. from Stockholm University in 2007 with a dissertation entitled Scandinavian Subtitles, which is a comparative study of TV subtitling norms in the Scandinavian countries. Jan's research interests include translation studies, translation theory, audiovisual translation, pragmatics and comparative linguistics. He is the president of the European Association for Studies in Screen Translation (ESIST), member of the European Society for Translation Studies (EST), founding member of the Nordic Network for Translation Studies (TraNor) and co-editor of the journal Perspectives – Studies in Translatology. He is a frequent presenter at international conferences and his publications include the 2011 monograph Subtitling Norms for Television, as well as several articles on subtitling, translation and linguistics. He has also worked as a television subtitler for many years, subtitling shows like Late Show with David Letterman, the Simpsons and Nikolaj og Julie. Jan is currently Director of the Institute for Interpretation and Translation Studies at Stockholm University, where he also researches and teaches audiovisual translation as Associate Professor.

Email: [please login or register to view author's email address]

©inTRAlinea & Jan Pedersen (2016).
"In Sweden, we do it like this On cultural references and subtitling norms"
inTRAlinea Special Issue: A Text of Many Colours – translating The West Wing
Edited by: Christopher Taylor
This article can be freely reproduced under Creative Commons License.
Stable URL:

Go to top of page