Abstract

The polysemiotic nature of subtitled films, whereby textual information (the subtitles) is combined with other audiovisual cues (the pictures and the soundtrack of the film) makes them very vulnerable as a form of translation, as well as a peculiar one: a consequence of the process of subtitling is that both the original (or source text) and the translation (target text) are presented simultaneously to viewers. The possibility of clashes between source and target texts is therefore very great, and incoherencies produced by the juxtaposition of visual referents from the source-language cultural sphere alongside textual referents originating from the target language are often commented upon by translation specialists. Whilst it is widely admitted that the use of cultural substitution – that is when a cultural reference in the source-text is replaced by another one in the target language – in subtitles should be resorted to with extreme caution, a coherent and consistent approach is also absolutely paramount in order to maintain viewers’ suspension of disbelief throughout the film. Looking at a corpus of films from the 1990s portraying predominantly African American characters, this paper analyses the ways in which the dialogues have been subtitled into French. While in most films subtitles display a tendency to neutralise the non-standard features of the original, some also reveal a great level of inventiveness and creativity: the juxtaposition in the subtitled films of features that are culturally-bound (whether to the source or to the target culture) produces a hybrid. Whilst the use of non-standard features from the target language in the subtitles can be justified by connotations they have in common with African American Vernacular English, the use of such features is not unproblematic and raises important issues.

Keywords: domesticating, foreignizing, African American Vernacular English, subtitling, verlan

1. Introduction

As the notions of domestication and foreignisation have become one of the dominant shibboleths of an increasing number of translation specialists, it is perhaps surprising that audiovisual translation, and specifically subtitled films, has seldom been discussed in the light of these two concepts brought to the fore of Translation Studies by Lawrence Venuti (1995). Subtitled films are semiotically very rich objects, and viewers are permanently reminded of their foreignness, both visually and auditorily: visually because of the text they have to read at the bottom of the screen when they are watching a subtitled film, and auditorily because of the foreign dialogue. The polysemiotic nature of subtitled films, whereby textual information (the subtitles) combines with other audiovisual cues (the film’s images and soundtrack) makes them a very vulnerable form of translation, as well as a peculiar one, to say the least. A peculiarity of subtitling is that both the original (or source text, henceforth ST) and the translation (or target text, henceforth TT) are presented simultaneously to viewers. The possibility of clashes between source and target texts is therefore very great, as is often commented upon by translation specialists who point out the incoherence resulting from the juxtaposition of visual referents from the source language (henceforth SL) cultural sphere with textual referents from that of the target language (henceforth TL). For instance, the use of features of African American Vernacular English (henceforth AAVE) to translate banlieue French in Mathieu Kassovitz’s La Haine (1995) and the systematic transposition of cultural references from the source culture to the target culture (which saw ‘Kronenbourg’ beer become ‘Bud’, ‘cinq francs’ become ‘two bits’, and characters’ names ‘Astérix’ and ‘Darty’ translated respectively as ‘Snoopy’ and ‘Walmart’) were commented on by a number of film critics who attributed the commercial failure of the film in the United States in part to poor subtitling: ‘a sloppy pastiche of black American slang hinders rather than helps an understanding […] of the film’ (an anonymous film critic, cited in Jäckel 2001: 227). The juxtaposition of a depiction of the daily lives of a trio of black-blanc-beur youths with American cultural references is indeed odd.[1] In other words, because of the use of features of AAVE in the subtitles and of the systematic transposition of cultural references for the target (American) audience, American viewers were presumably led to believe that African American youths shared important similarities with youths from the French banlieues, only to be brought back to reality when particular themes in the films made it all the more conspicuous that the action and setting were indeed foreign. There are specific elements of the narrative that make American viewers aware of the foreign nature of the film: as a critic points out, ‘the media hysteria about a stray revolver in a housing project might appear touching to a society in which prepubescents tote Uzis, but not to American youths for whom the young men’s inability to drive would be “quaint”’ (Jäckel 2001: 233). Besides making an argument in favour of an extremely cautious use of cultural substitution in subtitles – that is when a cultural reference in the ST is replaced by another one in the TL – this also suggests that adopting an approach that takes into account the links between the source and target cultures is absolutely paramount in maintaining suspension of disbelief. Therefore, the context of culture-specific elements and the relationship between the source and target cultures are of ultimate importance when choosing a translation strategy.

We are now going to examine the challenges of subtitling from a theoretical perspective in order to crystallise the dynamics that govern the subtitling into French of non-standard features or varieties, such as AAVE.

2. Domestication and foreignisation

Venuti’s original idea of foreignisation involves the use of the remainder which is defined as ‘the collective force of linguistic forms that outstrips any individual’s control and complicates intended meanings’ (Venuti 1998: 108). For Venuti, the remainder is not necessarily motivated by elements of the ST: the foreignness invoked by the remainder is not qualified, and although the remainder points towards the foreignness of that translation, it does not assert anything specific about the nature of this foreignness. This particular point can be addressed in the context of subtitled films: translators certainly have an impact on the content of subtitles, and their use of the remainder can express foreignness, precisely because this foreignness can in turn be supported by the other channels of the film (in other words, the other channels of a film can be used to orient the understanding of viewers in the TL towards a foreignness that is more specific),[2] or it can clash with the subtitles. Venuti’s paradigm would seem to be relevant to the examination of the content of subtitles. There are, however, two important provisos: first, translators have to be cautious in their use of the remainder, for if they use archaisms or forms that have been under-used or that are not codified in writing, they might affect the ability of viewers to read the subtitles with sufficient speed. Secondly, and perhaps obviously, the use of words in the TL that have strong connections to a particular region or to a particular group of people might of course be problematic and trigger cognitive-narrative dissonances. In other words, the remainder used in subtitles should be compatible to some extent, with the images of the film, and this compatibility should pre-date its use. This is in fact an issue that is not specific to subtitled films, but to the foreignising approach in general. The main issue here is that the use of the remainder (archaisms or dialectal features) can be very problematic, precisely because dialectal forms of the TL are often bound to the target culture, and as such can clash with visual elements of a film that are bound to the source culture.

One of the main concerns with subtitled films stems precisely from the fact that foreignness is built from domestic material. Non-dominant discourses are used to create the impression of the foreign, and the foreign is thus built using local blocks of meaning, blocks that might consist in deliberate divergence from the standard but are still, ultimately, part of the target culture. When these blocks from the target culture are superimposed onto images in the form of subtitles, there is always the possibility that the two – images and subtitles – may clash, generating a failed bi-cultural object that culminates in a break of viewers’ suspension of disbelief. In the case of the French film La Haine, it is certainly odd for the American audience (those that frequent art houses and consume foreign films, and therefore presumably experience the foreign) to be faced with three protagonists who wander between their housing estate and Paris where hotdogs cost ‘two bits’, and who speak (in the subtitles) using certain forms of AAVE. The remainder, here (features of AAVE), does not fulfill its foreignising purpose, but rather the opposite. It literally dis-locates the film, it changes its locus – partially anyway – and renders it a cultural aberration that does not belong anywhere. In this particular context, the use of the remainder was also accompanied by a systematic transposition of cultural references to the target culture, as observed in the introduction of this paper. This dislocation of the film meant that the dissonance between the subtitles and the images was too great, it was too deliberate an attempt to make the unfamiliar become familiar, in the face of many indications of foreignness (images, soundtrack, subtitles).

Judging from this film, it would seem that the use of the remainder in film subtitles is doomed to failure, necessarily creating a schizophrenic object that inhabits two spaces at once. In what precise circumstances, then, is it possible to use the remainder in subtitles, without leading to two cultures clashing on screen? Cultures, far from being discrete entities, sometimes share a fair amount of overlap, can take inspiration from one another, and fuse. This contribution seeks to argue that in the event of such an overlap, it is possible to hint at the foreignness of the original by using a remainder that shares, to a more or less direct extent, associations with the culture of the ST. By playing on the overlap between two cultures, it is possible, through the use of linguistic devices from the TL, to evoke and even invoke, to awaken and bring to the fore meanings of the ST that might have been thought ‘lost’ in translation, precisely by taking advantage of the polysemiotic nature of films. We will see below that although the translators’ cultural transposition of La Haine proved unsuccessful (commercially and critically at least), such strategies are not necessarily condemned to failure. Whilst the use of African American slang to subtitle La Haine was generally deemed unsuccessful, I will argue that verlan can serve a valid purpose when used to translate AAVE.

3. The use of verlan in subtitling

A particularly noticeable feature used in the subtitles of some of the films portraying speakers of AAVE is verlan, which is often referred to by French linguists as a feature of banlieue French. The term banlieue French is used here to refer to the variety of French spoken mostly by teenagers and that developed initially in poor suburban areas of Paris before spreading to other urban areas. This variety of French has been described and discussed by several French sociolinguists, who have generally emphasised the link between its spatial and social dimensions (Armstrong & Jamin 2002; Gadet 1998; Jamin et al 2006; Lepoutre 1997; Liogier 2002; Trimaille 2004; Trimaille & Billiez 2006). Sociolinguists have given different names to this variety, each emphasising a particular characteristic: it has been referred to as ‘parler véhiculaire interethnique’ (Offord 1996: 109), ‘langage des jeunes des cités de banlieue’ (Lepoutre 1997: 153), ‘argot contemporain des cités’, ‘ parlers jeunes urbains’, ‘sociolecte urbain générationnel’, ‘français contemporain des cités’ (all found in Trimaille & Billiez 2007), ‘langage des jeunes’ or ‘français des cités’ (both in Liogier 2002). This diversity echoes the various labels used to designate AAVE, and accounts for a certain heterogeneity of practices as well. Variables such as age, ethnic origins and social networks have all been shown to be important criteria in the definition of banlieue French. Although there is an abundant literature on the subject, the terminological fuzziness also illustrates that banlieue French, as a variety, is particularly difficult to circumscribe. Gadet, for example, suggests that this variety is difficult to link primordially to a social environment (‘populaire’), to age (‘jeunes’) or to geographical areas (‘cités’ or ‘banlieues’): ‘Y a-t-il lieu d’opposer au français populaire traditionnel une “langue des jeunes”, objet difficile à nommer (langue des cités, des banlieues)?’ (Gadet 2003: 85)[3] However, David Lepoutre, an ethnographer and sociolinguist, describes what he refers to as ‘le langage des jeunes des cités de banlieue’ (Lepoutre 1997: 153) in his book Cœur de banlieue, and explains that some of its features are particularly distinctive.

One of these features is called verlan, a type of slang (argot) that consists in inverting the sounds or syllables of a word or short phrase when speaking in order to create a new one. Verlan has been discussed in a vast number of studies.[4] As the lack of space does not allow us to go into great detail on all that is known about verlan, we will concern ourselves here with listing its more general principles. The word verlan usually describes both the process – the inversion of sounds or syllables – and also the end product. The new word created is ‘un mot de verlan’ or ‘un mot en verlan’. In the words or sometimes phrases that undergo this process, two sounds or two syllables are usually inverted, although it has to be said that the rules vary greatly depending on the original word used (which may have one, two, or more syllables – verlan is not restricted to two-syllable words). The word verlan itself comes from ‘l’envers’: the two syllables are inverted and form a new word, verlan. Plénat (1995) provides a large number of examples of words that have undergone this process, among which:

Fou [fu] becomes [uf] in verlan

Froid [fʀwa] becomes [wafʀ]

Pourri [puʀi] becomes [ʀipu]

Mystique [mistik] becomes [stikmi]

Verlan, according to Lepoutre (1997: 155-6), has a cryptic function: ‘Le verlan, comme tous les argots, est également un langage de fermeture, une langue du secret. Cette fonction cryptique du langage des rues s’exerce dans le cadre de l’école et plus largement dans les rapports avec les adultes’ [Like all forms of slang, verlan is also a closed language, a way of expressing whatever is secret. The cryptic function of street language thrives in schools and more generally in the relationship with adults]. Gadet (2003: 88) points out that verlan has ‘un renouvellement rapide’ [a fast-paced renewal], and that some words or expressions can be re-encrypted again and again, for instance ‘comme ça’ [kɔmsa] can become [sakɔm] and then [kɔmas] and even [askɔm] or [asmɔk]. Certain words in verlan have spread to other layers of the population or have become widely understood in France because of its widespread representation in the media (in particular for the purpose of caricaturing the young people who live in the banlieues) as early as the 1990s.[5] Therefore, some words such as ‘meuf’ (femme [woman]), ‘teuf’ (fête [party]), or ‘keuf’ (flic [cop]) have lost their cryptic values, and can all be found in the Larousse dictionary.

Lepoutre argues that verlan also has an ‘identity’ function:

La fonction identitaire prend une nouvelle dimension dans le contexte social et culturel des grands ensembles de banlieue: la juxtaposition des migrations, la communauté de situation entre Français et étrangers, dans l’exclusion comme dans la révolte, tout cela concourt à une recherche d’identité que marque le langage’. (Lepoutre 1997: 157)

[The identity function takes a more significant dimension in the social and cultural context of the suburban housing estates: the juxtaposition of migrations, the common experience of exclusion and rebellion shared both by the French and the immigrant population – all these factors contribute in fostering the search for an identity which is embodied in language.]

What this means is that verlan is still very stigmatised today and yet prized by those who use it. Although some words have made it into mainstream French dictionaries, it still bears strong links with banlieue youths.

This feature is of particular interest to us here because some words of verlan are used, to a lesser or greater extent, in the subtitles of an important number of films portraying speakers of AAVE. ‘Keuf’ for instance is used only once in the subtitles of In Too Deep (Michael Rymer, 1999), but ‘meuf’ is used a number of times in White Men Can’t Jump (Ron Shelton, 1992). In other films, however, such as Boyz n the Hood (John Singleton, 1991), Menace II Society (Albert and Allen Hugues, 1993), and New Jack City (Mario Van Peebles, 1991), verlan in the French subtitles is often used. In Boyz n the Hood, ‘meuf’ is used ten times, ‘keum’ (mec [guy]) and ‘keuf’ are both used once. More interestingly from the perspective of a film, the word ‘keubla’ (verlan for ‘black’) is used in the opening line of the film. In Menace II Society, ‘meuf’ is used eight times, ‘tassepé’ (pétasse [floozy]) is used three times, and ‘teuf’ and ‘keuf’ are used twice each. ‘Beuh’ (herbe [grass]), ‘chelou’ (louche [weird]), ‘renoi’ (noir [black]), ‘pécho’ (chopper [catch]) are all used once. Again, ‘meuf’ and ‘teuf’ appear in the subtitles of the very opening exchanges of the film. Finally, in New Jack City, , ‘refré’ (frère [brother]) is used five times, while ‘feuj’ (juif [Jew]), ‘kepa’ (paquet [bag]) and ‘keuf’ are used once.

The use of verlan, naturally, is far from being unproblematic, because it is so deeply connected to banlieue French and to images of French cités, that its use in subtitles runs the risk of making a strong and perhaps unjustified association between speakers of AAVE and speakers of banlieue French. However, authors like Hervé Vieillard-Baron or Lepoutre have shown that connections exist between the American and the French street cultures, and between the young people in particular who live in the cités (and who might be considered the primary speakers of banlieue French, and the main users and innovators of verlan) and their African American counterparts. Of course, the conditions of life are different in the American inner cities and in the French cités, as Lepoutre highlights:

Si l’on ne peut parler, comme dans les quartiers noirs des métropoles étatsuniennes, généralement composées de plus de 90 % d’Afro-Américains, d’‘hyperségrégation’ raciale, ni même de véritable homogénéité culturelle, puisque des populations d’origines très différentes cohabitent dans le grand ensemble. Du moins la forte proportion de population d’origine étrangère d’une part, la nette domination arabe et plus largement musulmane d’autre part, donnent au quartier une indéniable dimension ethnique et religieuse et, partant, un caractère marqué de ghetto contemporain.’ (Lepoutre 1997: 84)

[Unlike the black neighbourhoods of American cities which generally comprise more than 90% of African Americans, it would be unfair to describe French housing estates in terms of racial ‘hypersegregation’ or of true cultural homogeneity, since population groups from very different origins live together in the housing estate. Both the high proportion of immigrants and the clear Arabic and more generally Muslim domination give the neighbourhood an undeniable ethnic and religious dimension, and thereby a strong sense of a contemporary ghetto.]

Vieillard-Baron (1996: 46) also states that ‘depuis trente ans, la ville de Chicago occupe une place mythique dans l’imaginaire des jeunes de banlieue. Elle exprime à la fois le rêve américain, l’exotisme, la relégation des ghettos et le grand banditisme avec Al Capone’ [For 30 years, the city of Chicago has enjoyed a legendary status in the imagination of suburban youths. It simultaneously embodies the American dream, exoticism, ghettoization, and organized crime through Al Capone.]. French youths draw their inspiration from their American counterparts for music, fashion, and also a certain idea of ghettoisation as highlighted by the two quotations above.

There is therefore a fairly strong overlap between the two cultures, and the translators are clearly trying to take advantage of it. By using certain salient features of banlieue French in the subtitles, they can trigger certain associations with a particular socio-economic background, as well as possibly give a sense of counter-culture through linguistic rebellion. Verlan words might thus also be considered to be cultural metonymies, devices that are commonly used in fiction and consist in using a particular element associated with a cultural group in order to evoke the cultural group itself and other features that may usually be associated with it. In the words of Maria Tymoczko:

A piece of literature customarily evokes its culture through consequential and telling signals or details, typically parts or aspects of the culture that are saturated with semiotic significance and emblematic of the culture as a whole, both in terms of objective structure and subjective experience […] In this regard, such cultural elements within a literary work are metonymic evocations of the culture as a whole, including its material culture, history, economy, law, customs, values, and so on (Tymoczko 1999: 45).

Although Tymoczko here describes the relationship between text and culture as a metonymic one, the same is arguably true of films and culture, as Monaco illustrates:

Because metonymical devices yield themselves so well to cinematic exploitation, cinema can be more efficient in this regard than literature can. Associated details can be compressed within the limits of the frame to present a statement of extraordinary richness. Metonymy is a kind of cinematic shorthand. (Monaco 2000: 167-8)

Metonymy is widely used in films, and is a particularly useful rhetorical device, as viewers are given ‘comparatively limited time to identify and understand what is represented on screen’ (Di Giovanni 2007: 96), and culturally salient features (whether visual or verbal) are used for their intrinsic metonymic value. If we apply Monaco’s statement, then visual or verbal on-screen representations can be treated as metonymic evocations of a culture. In the films mentioned above, there are recurring elements, features and themes which appear to be used to represent a certain idea of African America in the films: loose-fitting clothes, baseball caps, sports, crime, violence and AAVE are all elements that contribute to creating or recreating stereotypical representations easily associated with African Americans. The subtitles, also presented ‘within the limits of the frame’ can yield the same power, as the words used in the translation can serve a metonymic purpose as well and can evoke a particular group in the target culture. And if a translation takes place between two cultures that share some contact or familiarity – as is the case in the films with American and French street cultures – then two assumptions can be made: first, that the cultural metonymies of the original may be accessible to viewers of the translated film, and secondly, that the cultural metonymies used in the subtitles may well exploit this familiarity and ‘bridge the gap’ between the two cultures for the viewers in the TL. The use of verlan in the subtitles suggests that translators perceive possible links between the source and the target culture and that because of the overlap between the two, which is also complemented by the other elements of the film (whether visual or acoustic), the use of a salient feature of the target culture such as verlan can help complement viewers’ understanding of a film.

Verlan is the only feature of banlieue French that is used in the subtitles, but it is a very salient one, and most importantly, one that is possible to put into writing. It would therefore be wrong to say that the subtitles are written in banlieue French, and more accurate to say that translators are relying on a particular feature, and on the associations that are stereotypically made with it, to trigger from viewers a certain indexing of characters.

This indexing is social, related to age, geography, and to a certain extent, ethnicity. Armstrong and Jamin (2002: 130) point out that banlieue French ‘is essentially a young working-class phenomenon’. By definition, banlieue French is also a (sub)urban variety. In the films mentioned above, verlan is exclusively used to subtitle the lines of characters who are young and from poor social backgrounds. In the films, only urban contexts are represented, and the level of social and ethnic isolation is always very important, both visually (the characters live in all-African American neighbourhoods) but also thematically (this relegation is often referred to by characters). Armstrong and Jamin (2002: 129), in their study of La Courneuve, highlight that ‘the majority of [banlieue French] features could not be allocated to any specific ethnic group, contrasting with the case of African American Vernacular English’, but although banlieue French is not specifically associated with black people, the social and ethnic relegation of the population who live in the French banlieues causes the variety to have strong associations with ethnic minorities.

The link between banlieue French and AAVE, because of these similarities and of the cultural overlap discussed above, is therefore very pertinent. But while the parallel between AAVE and banlieue French can be attractive, it is not unproblematic since one always runs the risk of making undesirable associations between the two and turning the subtitled films into hybrid objects.

Some verlan words like ‘teuf’, ‘keuf’ and ‘meuf’ are now in the Larousse dictionary. Words like ‘keubla’, ‘tassepé’, ‘feuj’ and ‘kepa’ (or any alternate spellings thereof) are not, which reflects their lack of integration in the French lexis. In other words, the former have all been codified, and to some extent, can be legitimately written, whereas codification has been ‘forced’ onto the latter words by the subtitlers, using certain tacit rules that appear to be used when it comes to verlanisation (for instance, the use of the letter ‘k’ rather than ‘qu’ with verlan words starting with the sound /k/). I would contend that words like ‘teuf’, ‘keuf’, and ‘meuf’ are recognised and understood by a larger part of the French public than ‘keubla’, ‘tassepé’, ‘feuj’, or ‘kepa’. If the codification of these words into writing is problematic (and not unanimous – ‘keubla’ can for instance be found spelt ‘kebla’, or ‘tassepé’ without the middle ‘e’), what is even more problematic is whether they will be understood by viewers or whether they will prevent them from engaging with the film.

The ‘Code of Good Subtitling Practice’, available on the webpage of the European Association for Studies in Screen Translation, states that in subtitles, ‘the language should be grammatically correct since subtitles serve as a model for literacy’ (Carroll & Ivarsson 1998).[6] Whilst verlan is not grammatically incorrect, its use certainly challenges accepted models for literacy as well as a certain idea of the norm. Díaz Cintas and Remael (2007: 9) also suggest that subtitles have to be unambiguous and ‘semantically adequate’.[7] Whilst it is difficult to establish precisely what ambiguity and semantic adequacy might be, and what particular elements in subtitles may be ambiguous for viewers, words in verlan – more particularly those that are less widely used – certainly run the risk of not being understood and of generating ambiguity. Naturally though, viewers can rely on other contextual clues to speculate on the meaning of those words, which overall, represent a very small portion of the dialogue. In addition, the opacity of verlan might even be desirable, in the same way that AAVE can potentially be opaque for non-speakers of AAVE.

In the subtitles of the films examined, the use of verlan is particularly salient because it is associated with a very specific culture, one that is local. The case of verlan, then, engages us in a particularly interesting analysis a posteriori: because it is so strongly associated with a particular age group (youths) from particular areas (the euphemistic banlieues) that are socially deprived,[8] verlan runs the risk of triggering cognitive-narrative dissonances when used in film subtitles.

However, we have shown that French street culture has strong ties with its African American counterpart, which is absolutely crucial to our study. The link between the two cultures is a strong and significant one, and is constantly re-asserted in the media by artists, through clothing, parallels made between American inner cities and French cités. They are so deeply connected that a layperson might think that they are, from a French perspective, very similar. Whilst representations of African America are ubiquitous in France, the opposite is far from true. I argue here that it is possible, through the use of verlan in the subtitles, to trigger associations with French street culture and also, by proxy to its overarching ‘mother’ culture. In the end, the use of verlan to translate AAVE can benefit from networks of representations and stereotypes that associate French street culture and African Americans to trigger connotations for viewers in the TL, connotations which go beyond mere domestication. The use of verlan may actually serve a purpose that can be called foreignising (I qualify this statement below).

It is worth noting that whilst the English subtitles of La Haine were singled out as problematic, French film critics rarely discuss the quality of the translation (whether dubbing or subtitling) of films in their reviews. In the vast majority of cases, it is in fact impossible to determine whether the critics have seen a film in a dubbed or subtitled version. Whilst the use of features of AAVE to translate banlieue French was deemed a hindrance, the use of verlan to translate AAVE has not been commented upon. The use of verlan is relevant precisely because French street culture is in many ways subordinate (for lack of a better word) to its American counterpart, because it draws so much from it – in terms of streetwear, sports, and arts – and because it draws its inspiration and momentum from it. Whilst French teenagers in the cités (those we have described as the primary users of verlan) certainly hold their American counterparts in high regard, the opposite is not true, or certainly not to the same extent, as most Americans teenagers would be ignorant of the French reality, being considerably less exposed to it. American youths know little about French rap music or sports culture, and have little interest in French cités. A careful assessment of the cultural dynamics at play is therefore necessary, should translators want to use features that are connotated geographically or socially in the TL. The important difference with Venuti’s theory is that the choice of verlan in the TL is clearly motivated by the ST. Verlan metonymically invokes French street culture as well as African America, and the images and soundtrack, rather than clashing with verlan, provide a context for its understanding. The meaning of verlan is channeled by the other semiotic systems of the film: they make verlan make sense. This is only possible in a situation where the overlap between the source and target cultures is appropriate, and, crucially, shared by viewers.

The use of verlan in subtitles certainly has to be controlled and contained. The subtitles are trying to make sense of the film, and have to be in cognitive-narrative harmony (we talked about dissonance above, this is its virtuous opposite) with the other channels. Verlan has resonance potential: that is, it can summon SL connotations in the TL, it can metonymically invoke a network of associations that is directly in relation with the source culture. Although not the case in any of the films discussed here, it can be conjectured here that such features (particularly when they have a cryptic component, the way verlan does) could potentially be overused to the point of obscuring meaning irremediably. In the films discussed here, however, verlan words do not feature frequently enough to obscure meaning significantly (should viewers fail to identify these words as cases of verlan). The opacity of verlan could in fact be desirable, in that it deliberately confronts viewers with a departure from the standard that establishes characters as speakers of a non-dominant discourse. Verlan in the films is used significantly (and usually in the opening lines of the films), but in overall quantitative terms not very frequently or systematically by characters portrayed as young speakers of AAVE. It should also be noted that unlike in the English subtitles of La Haine, where every cultural reference was transposed to the target culture, cultural references in the films below usually retain their foreignness: ‘dollars’ remain ‘dollars’ in the subtitles, ‘Miller High Life’ beer is also retained in Do the Right Thing (Spike Lee, 1989), as is Old E (Menace II Society), and as are references to prominent American people (such as David Dinkins, Michael Jordan or Mike Tyson). In some cases, when references would be unknown to a French audience or unclear from context, they are neutralised, as in the examples provided in the following table:

Table 1 – Examples of the neutralisation of cultural references in the subtitles

The examples in this table illustrate that when cultural references are not retained in the French subtitles, they are not transposed to the target culture either. In all these examples, a cultural reference in the original has been made more generic in the subtitles – Keith Sweat, an African American singer and his lyrics become ‘vieux refrain’, a Sizzlers restaurant becomes ‘resto’, and ice hockey superstar Wayne Gretzsky becomes a somewhat surprising ‘hockey man’. In White Men Can’t Jump, Billy is compared first to American gymnast Cathy Rigby (‘M le gymnaste’) and then to the fictional character Opie Taylor from a small community in North Carolina (‘péquenaud’). In The Wood (Rick Famuyiwa, 1999), references are made to Willy Nelson, a country music singer (‘péquenot’) and to Alice, the housekeeper (‘bonniche’) of the Brady family in the American show The Brady Bunch. In all these examples, translators have clearly picked up on one particular characteristic of each reference. In none of these cases have references been transposed to the target culture. Verlan, then, represents the only significant effort to adapt culturally-bound target culture elements.

This analysis indicates that Venuti’s paradigm, if used in the context of films, needs elaboration. Since the use of the remainder in the TT is not necessarily linked with any singular properties that may be found in the ST, in The Translator’s Invisibility Venuti does not carry out the type of cultural assessment that we are suggesting in this paper. According to his analysis, unlike ‘fluent’ translations, the translation that ‘releases the remainder’ (Venuti 1998: 10) opens itself to the incursion of the foreign, ‘the substandard, and the marginal’ (Venuti 1998: 11), and this in spite of the fact that the idea of the foreign is built with domestic material. Venuti (1998: 11) elaborates: ‘Cultivating a heterogeneous discourse […] does not so much prevent the assimilation of the foreign text as aim to signify the autonomous existence of the text behind (yet by means of) the assimilation process of translation’. This is perhaps never more correct than in the case of subtitled films: the autonomous existence of the foreign text (film) is not in doubt, as outlined above. What is more crucial here is the expression of the marginal and of the substandard (in the words of Venuti), specifically through the use of verlan. The use of this heterogeneous discourse (verlan), although eminently assimilating, does not and cannot domesticate or foreignise the source film on its own. Rather, it signifies the non-standard qualities of the dialogue in the source film. The difference with Venuti’s theory is that the use of verlan in the TL is motivated precisely by the relationship between source and target cultures, and relates to the variation and to the non-standard qualities found in the original.

Díaz Cintas and Remael note that good translators somehow manage to ‘suggest’ variation in subtitles. Whilst their assumption is quite vague and the authors do not explain exactly how translators can manage to ‘suggest’ particular information to viewers, we would like to offer the explanation that the use of verlan in French subtitles, when juxtaposed with images of African American youths, can aptly ‘suggest’ the variation of the original.

The assessment that translators make to decide whether some forms of the remainder can be used in the TT involves looking at a film as a whole – and not merely at the original dialogue in an atomised way, translating item-for-item, word-for-word. Translators analyse the culture(s) portrayed in the film and decide whether features of the TL can trigger the right associations from viewers and give them access to the meanings of the ST. Because of the constraint of reduction and the presence of the other channels in subtitles, pieces of information are discarded (because they are deemed superfluous, irrelevant or semiotically redundant), while others are selected (because they are deemed meaningful or salient) and undergo some level of reorganisation. Subtitling therefore involves a de-atomisation of the ST: translators must look at the film in its entirety and in its complexity before they are able to make a judgment and decide whether a particular use of the remainder makes sense from a cultural perspective.

‘Suggesting’ the right associations for viewers in the TL requires knowledge of the networks of representations that are intertwined in subtitled films in order to work out the meanings that viewers will understand. While this may sound commonplace in translation and translation theory, it has deeper implications in audiovisual translation. The semiotic hybridity of films implies that meanings are constantly the subject of a negotiation between the different channels: images portraying African American youths, over which subtitles are presented, can ‘channel’ meanings and have the power to privilege one or more interpretations over others. The foreignness inherent to subtitles, the other channels of the films, and the fact that cultural references are not transposed to the target culture invite viewers to consider the film as foreign. Because of this we would like to argue that Venuti’s domestication/foreignisation paradigm does have some operational relevance in the context of subtitled film: the subtitling mode is a constant reminder of foreignness, but the content of subtitles can use features like verlan which are in a way domesticating because they are so deeply rooted in the target culture. However, when juxtaposed with the image of the foreign (and with the subtitles as a further reminder of foreignness), viewers can match the socio-cultural features associated with the remainder with the socio-cultural properties of the original, and in those cases only can translators make use of the domestic to express the foreign. In the case of subtitled films, the use of domesticating forms can serve a foreignising purpose and create a specific Other, and say something of the nature of the Other’s foreignness.

In this context, the task of translators is most particular: it obviously involves a willingness not to standardise the dialogue, but also and more crucially, a great deal of empathy to determine what associations viewers will be making. Translators have to draw on a pool of widely acknowledged (and therefore hopefully more recognisable) stereotypes to construct and express difference. There is a tension – almost a discomfort – in the practice of audiovisual translation: whilst we have acknowledged that translation is always-already ideological, translation seems to require from translators a combination of specialist – and to a certain extent elitist – skills, and the use of lay persons' assumptions. In relying on recognised stereotypes, translators run the risk of perpetuating those very stereotypes. For example, the use of verlan may well bear social and geographical connotations that are appropriate in the representation of African America, but the themes tackled in the films discussed here (such as racism, substance abuse, and violence) may well be associated with speakers of verlan (i.e. people who use verlan are violent and so on) and perpetuate negative perceptions of certain social groups. The use of verlan, whilst very interesting from a translational perspective because of the parallel it exploits with African America, also means that negative stereotypes associated with speakers of verlan can be attributed to the characters portrayed. In other words, alongside the socio-geographical information, negative stereotypes are also ‘carried over’ in the translation process.

It can be argued here that this is also the case with such media of representation as films: films that claim to realistically portray the experiences of African Americans in the United States somehow pigeonhole them to a limited set of stereotypes. Ghettoisation, drugs, and violence are the only leitmotivs, and it is no wonder, then, that speakers of AAVE are often associated with negative stereotypes, such as poor education and aggressive behaviour. AAVE is not sub-standard, but rather is a rule-governed system that is learnt and known by its speakers. AAVE is often thought of as somehow inferior to English, and the characters portrayed in the films discussed in this paper who use features of AAVE are systematically perpetuating negative stereotypes. Paradoxically, the films that portray speakers of AAVE were said to be presenting a more realistic representation of African America but appear to be themselves perpetuating to some extent the stereotypes that plague the perception of AAVE and its speakers. Whilst translators may well be aware of issues connected to prestige and of the social status of languages, it is very difficult – perhaps impossible – to ignore these issues and find a middle ground in the form of translations that do not reinforce these poor perceptions. If the films could be accused of perpetuating negative stereotypes about African Americans, then some might argue that the subtitles ought to do the same. What the subtitles do, however, is perpetuate stereotypes about the French banlieues (more specifically about speakers of verlan) and reassert the link between American inner cities and French banlieues.

4. Conclusion

What does this study reveal about domestication and foreignisation as general concepts for translation? According to Venuti, the use of the remainder is not motivated by specific elements of the ST, and as a result the foreignness is one that disturbs dominant discourses. It is a foreignness that is not revelatory of the nature of the original. This study addresses this particular point and suggests that foreignisation can be qualified. The use of the remainder (such as verlan) can, in some cases, go beyond hinting at the foreign nature of the original and can actually explore this foreignness. Foreignisation as a concept needs to see its scope broadened: beyond a disturbance of the dominant discourses, it can also serve more specific purposes. Naturally, this is done through the use of domestic material (which, as Venuti rightly pointed out, is always the case with translation), and foreignisation remains essentially reliant on the use of this material. One of the main issues that translators face when they are translating non-standard varieties is the illustration of specific connotations of the original in the translation. The use of verlan to translate AAVE reveals that this can be achieved to some extent when two cultures share strong ties that are widely recognised. However, such solutions are far from universal and are only possible in very limited cases. It is therefore important to explore the cross-cultural potential of non-standard features used in the TL. Relations between the source and target cultures need to be explored case by case in order to find out how much overlap there is between the two, and how familiar viewers in the TL are with the source culture. While French viewers can be expected to be quite familiar with African American culture (either through media exposure or even by proxy through exposition to French street culture), American viewers do not benefit from the same level of exposure to French street culture. Consequently, the use of verlan to subtitle AAVE can trigger associations with the United States, but the use of AAVE to subtitle banlieue French cannot, and instead brings the film back ‘home’ to American viewers in a way that is deeply problematic.

Subtitlers work under strict time constraints, and their research into how much is shared by the source and target cultures is bound to be limited. Their assessment relies on systems of representations and existing stereotypes, and their translation decisions are informed by these assessments. This might in itself be considered problematic insofar as it means that translation can confirm existing stereotypes, in a negative way, and there seems to be an irreconcilable tension between relying on stereotypes to establish characters quickly on the one hand, and perpetuating those stereotypes on the other.

Subtitled films offer the possibility of familiarising oneself with foreign languages and cultures: first the soundtrack retains intonation and pronunciation patterns that would be replaced through dubbing, and in parallel, the images bring viewers into contact with ‘mannerisms and behaviours of other cultures (gesticulation, way of dressing, interpersonal relationships, geographical spaces)’ (Díaz Cintas & Remael 2007: 15). And whilst this possibility of having direct access to the original is part of the reason why subtitling is such a vulnerable form of translation, it has also been described by specialists of subtitling as one of the most positive aspects of subtitling (D’Ydewalle & Pavakum 1992; Koolstra & Beentjes 1999). It is because of the foreignness already present that the idea of foreignisation is so much more relevant in the context of subtitled films: the subtitles can make use of the domestic tools and values described by Venuti specifically for the purpose of the foreign, because the foreign is already there and is complemented by the subtitles.

Whilst linguists are well aware that linguistic varieties are not hierarchically organised, it is possible that because of the mechanics described above, the use of the remainder in translation reinforces the status of the standard as the perceived superior variety. In films, as in literature, linguistic forms have high metonymic values, and stand for whole networks of representations. As we have seen, Venuti’s concepts – domestication and foreignisation – prove very useful when considering the cultural dynamics at play in interlingual subtitling. Because of the nature of films, however, the level of cultural proximity between features in the dialogue of the original and the subtitles should be looked at carefully before translation is attempted. While Venuti argues that most translations are transparent and translators invisible, subtitles are anything but transparent. Consequently, their foreignising form can potentially be benefited from, and subtitles (both in terms of their form and their content), can work together with images to constitute a system, which despite having domesticating elements, serves a purpose that is ultimately foreignising.

About the author(s)

Dr Pierre-Alexis Mével is currently a lecturer in Translation Studies at the University of Nottingham, and is course director of the MA in Translation Studies. He teaches Translation Theory and Audiovisual Translation at both undergraduate and postgraduate level. His experience as a freelance translator working with both subtitles and websites has given him the practical background he needed to complete his doctoral thesis at the University of Nottingham in 2011. He has a particular interest in the representations of non-standard varieties on screen, and on the theoretical underpinnings of juxtaposing two linguistic varieties on screen. This raises particular issues relating to the formation of identities, about their cultural porosity, and the transferability of culturally bound features and the nature of their adaptation in another culture.

Professor Jo Robinson is Professor of Theatre and Performance and Head of the School of English Literature, Language and Linguistics at Newcastle University (United Kingdom). Her research cycles between theatre history and historiography and work on twentieth/twenty-first century theatre and performance – always with a particular focus on the relationships between place, community and audience. She is interested in the capacity of digital humanities as broadly conceived, both in terms of creating innovative digital histories and working with augmented reality to enhance the accessibility of live performance.

Dr Paul Tennent is an Assistant Professor in the Mixed Reality Laboratory at The University of Nottingham (United Kingdom) with a long track record of developing internationally recognised immersive experiences with artists and cultural heritage institutions. He leads or participates in a number of projects around hybrid physical-digital interactions, across several interdisciplinary collaborations from art, culture and history, to engineering, to education and accessibility.

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

•  Previous article
• Next article