Screening the foreign: directionality strategies used in the dubbing of Eat, Pray, Love and Under the Tuscan Sun

By Dominic Stewart (Università di Trento, Italy)


In the literature on dubbing there has been very little focus on the issue of directionality within the context of multilingual films. This paper hopes to redress the balance by focusing on the Italian dubbing of two English-language films – Eat, Pray, Love and Under the Tuscan Sun – the aim being to exemplify how the inherent alterity of foreign films can be foregrounded or suppressed by directionality strategies in the dubbing process.

Keywords: audiovisual translation, dubbing, directionality, accents, multilingual films

©inTRAlinea & Dominic Stewart (2016).
"Screening the foreign: directionality strategies used in the dubbing of Eat, Pray, Love and Under the Tuscan Sun", inTRAlinea Vol. 18.

This article can be freely reproduced under Creative Commons License.
Stable URL:


In the burgeoning research on film and television dubbing some important themes have received recurrent attention, while others have received very little. Much discussion has been devoted, for example, to topics such as dubbing vs subtitling, dubbing as duplicity and as verbal acrobatics, dubbese and viewers’ perceptions of it, but there has been relatively little focus on how dubbing deals (i) with directionality, that is, the rapport between native and non-native usage in the transfer from source text to target text, and (ii) with multilingual films offering metalinguistic scenarios where the language of either the source text or the target text is the object of discussion or attention. Bollettieri-Bosinelli (2002: 84) recalls ‘l'incongruità di certi western in cui l'ufficiale yankee chiedeva al capo indiano "Capisce l'inglese?", del tutto improbabile in un film parlato interamente in italiano’ [‘the incongruousness of westerns in which the Yankee officer asked the Indian chief “Do you understand English”, totally implausible in a film entirely in Italian’]. Along the same lines Bollettieri-Bosinelli (1994: 25) discusses ‘…certi doppiaggi degli anni cinquanta, dove poteva capitare di sentire in un western un ufficiale bianco […] chiedere con inconfondibile accento romanesco al capo indiano se parla italiano’ [‘…some examples of dubbing from the 1950s, in which you could hear a white officer in a western […] asking the Indian chief in an unmistakable Roman accent if he speaks Italian’]. In these examples it is the multilingual scenario which is salient, but the issue of directionality is also relevant, because a decision has to be taken regarding the presumably non-native English of the Indian chief and how his voice is to be dubbed into Italian.

In this paper I intend to focus on directionality within the framework of multilingual films dubbed from English into Italian, with a view to illustrating how directionality strategies can foreground or suppress the cultural and linguistic otherness of foreign films. Particular attention will be devoted to Eat, Pray, Love (Ryan Murphy, 2010) and Under the Tuscan Sun (Audrey Wells, 2003).

Directionality in dubbing

It should be underlined from the outset that the interpretation of directionality in this paper is unorthodox. The concept usually refers to whether translators work into a native language or into a non-native language, and therefore the mother tongue of the translator is central to the issue (see for example Pokorn 2005, Beeby Lonsdale 2009). In this sense directionality is to be distinguished from direction of translation, which is concerned simply with which languages are part of the translation process and in which direction one moves between them (for example from Swedish to French or from French to Swedish), irrespective of native/non-native language contrasts. For present purposes, however, ‘directionality’ or ‘dubbing directionality’ refers to the interplay between native and non-native language production across original films and dubbed films, in other words the way in which native and non-native usage in the source text are reproduced in the target text.

Four modes of directionality may be posited. In the explanations, as in the entire paper, I shall refer to the direction English to Italian:

Native → native

It goes without saying that native language production in the original is almost always translated with native language production in the dubbed version. Native English, whether it be Scottish, Canadian, New Zealand etc. is almost always dubbed into native Italian. Of course the distinction between mother tongue and foreign language is not always clear-cut, since there are many varieties of a language which hover between native and non-native status (see Davies 2003), but  the native → native directionality accounts for the great majority of dubbing activities, whatever the languages involved.

This directionality naturally includes those cases where a regional English accent is rendered with a regional Italian accent (Pavesi 2005: 38, Bruti 2009, Chiaro 2009: 158-9), though in practice this happens very rarely, the most obvious example being The Simpsons, in which for example a Scottish accent is rendered by a Sardinian accent. In passing it is worth noting that regional Italian accents are certainly used in dubbing, but as a rule they do not correspond to marked regional English accents in the original – perhaps the most striking example of this is the Italian version of Monty Python and the Holy Grail (Terry Gilliam and Terry Jones, 1975), which contains a generous sprinkling of Tuscan, Bolognese, Milanese and Sicilian, but for the most part these do not translate regional accents in the English version. A further example of this is the Roman accent adopted in the Night at the Museum films (Shawn Levy) for the Roman centurion Octavius, assigned standard British English in the original film.

Non-native → non-native

The modality whereby a non-native English accent is dubbed with a non-native Italian accent is common, and examples readily spring to mind: the voices of the French Inspector Clouseau, of Agatha Christie’s Inspector Poirot, of the character Zorro, and, to cite an instance from one of the films that will be analysed later in this essay, the Brazilian businessman played by Javier Bardem in Eat, Pray, Love. Examples abound in animated pictures too: Puss-in-Boots (Antonio Banderas) in the Shrek films and Lumière the French candelabra in Beauty and the Beast (Gary Trousdale and Kirk Wise, 1991). Here the decision to dub these characters with non-native production is an obvious one because the non-native accent is a defining characteristic of the personalities in question (see Heiss 2004: 212-3 for further discussion), but it should also be pointed out that this directionality is frequently associated with comic, caricatural figures whose non-native accent contributes significantly to the humour. In this connection it is interesting to compare this modality with those described below.

Native → non-native

The native → non-native modality is predictably less common than those outlined above, given that this study is confined to native / non-native production from English to Italian, and therefore does not embrace instances where, for example (in an English-language film), native French – perhaps subtitled in the original – is rendered with non-native Italian (Italian pronounced with a French accent and/or characterised by imperfect lexis and grammar). Within the parameters set by this paper the most obvious example of native to non-native is Stanlio e Olio / Laurel and Hardy, where Stan’s native English and Oliver’s native American are both rendered with comically non-native Italian (see Bollettieri-Bosinelli 1994: 20). Here again it is the comic nature of the protagonists which is a crucial factor in the decision to dub them as non-native, and again this modality is more widespread when comic or cartoon characters are involved. Further examples are the English geese in The Aristocats (Wolfgang Reitherman, 1970; see Bruti 2009) and the Scottish hen in the animated picture Chicken Run (Peter Lord and Nick Park, 2000), which is assigned a strong German accent in the Italian version (Chiaro 2009: 159).

One might also include occurrences where pidgin English is rendered with non-native Italian, but on the whole this directionality is comparatively rare.

Non-native → native

As a rule this modality applies when the non-nativeness of a character’s accent is barely noticeable or is irrelevant. For instance in the film Love Actually (Richard Curtis, 2003) there are a few non-native English accents, two of which are scarcely perceptible either because their English is near-native (the boss’s secretary Mia played by Heike Makatsch), or because they say very little (the office-worker Karl, played by Rodrigo Santoro), in contrast to Aurelia the Portuguese maid/waitress (Lúcia Moniz), whose difficulties in English are central to one of the film’s storylines. Thus Mia and Karl are dubbed with a native Italian accent, while Aurelia is assigned non-native Italian. In some films of Arnold Schwarzenegger, the actor’s non-native English accent is not germane to the American character he portrays and in any case remains unemphasised, so his non-nativeness is eliminated in the Italian dubbing. The same goes for certain films of Antonio Banderas, for instance You Will Meet A Tall Dark Stranger (Woody Allen, 2010). This happens in animated pictures too, for example in The Aristocats the character Duchess, voiced in the source text with a French accent by the Hungarian actress Eva Gabor, is assigned native Italian in the target text, notwithstanding the occasional French word (in particular Monsieur). A truly exceptional case is once again to be found in the dubbing of Monty Python and the Holy Grail, in which a French soldier speaking English with a French accent is transformed into an Austrian speaking native Sicilian.

A case apart within this category is that of Italian actors who speak English in English-language films. In the majority of cases the actors in question are playing Italian characters and so naturally they dub themselves into Italian, thus their non-native English in the source text is converted to native Italian in the target text. (Of course there are also cases of Italian actors playing non-Italian characters in English-language films, for example Gian Maria Volonté as a Colombian in Chronicle of a Death Foretold (Francesco Rosi, 1987) and Giancarlo Giannini as a Mexican patriarch in A Walk in the Clouds (Alfonso Arau, 1995).) Examples of this are not hard to locate: a fairly recent example is the film The Tourist (2010, Florian Henckel von Donnersmarck), set in Venice, in which several Italian actors, including Christian De Sica, Nino Frassica and Neri Marcoré dub themselves in the Italian version. The transformation is usually non-traumatic, since the characters are Italian anyway, but of course with this solution the linguistic differentiation between the Italian and non-Italian characters of the source version disappears in the dubbing, in which for example the characters played by Christian De Sica and Johnny Depp address each other in native Italian. This is just one more case among countless others where ‘il doppiaggio non riesce […] a riproporre le spesso numerose varianti linguistiche e sociolinguistiche presenti nella versione di partenza’ [‘the dubbing cannot […] preserve the often numerous linguistic and sociolinguistic variants present in the source text’] (Perego and Taylor 2012: 124). To offset the lack of differentiation the Italian characters’ voices are sometimes given a more regional or colloquial emphasis, in contrast to the fairly flat Italian accent used for the non-Italian characters; for example Raoul Bova speaks non-native English in Under the Tuscan Sun but has traces of Neapolitan in the dubbing.

Multilingual films

Multilingualism is another term which needs to be qualified, at least within the context of film production. As the epitome of a multilingual film one instinctively thinks of works such as L’auberge espagnole (Cédric Klapisch, 2002) in which an array of languages is spoken, but in reality the range can be restricted to just one more language beyond the basic language of the film – see Chaume (2012: 131) and Chiaro (2009: 159), who alludes to the ‘translational quandary’ deriving from films of this type.  See also Heiss (2004).

In this regard Chaume (2012: 132) distinguishes between the main language of a film (L1), the dubbing language (L2) and any other languages (L3) used during the course of the film. He goes on to discuss the various ways of dealing with L3 in dubbing, but points out that the issue becomes more complicated when the L3 corresponds to the L2, that is, to the dubbing language, and here too different strategies may be adopted: zero translation, translation into a different accent, translation into a different language altogether. This latter strategy is exemplified by Chaume with reference to the Spanish version of Pulp Fiction (Quentin Tarantino, 1994), in which an impromptu Spanish lesson is transformed into a lesson of Portuguese, but in the literature one can find references to analogous occurrences between English and Italian, notably in A Fish Called Wanda (Charles Crichton, 1988), where Wanda (Jamie Lee Curtis) becomes aroused whenever she hears Italian words (Bollettieri-Bosinelli 2002: 84-5, Chiaro 2009: 259). In the Italian version the words spoken by her lover Otto (Kevin Klein) are transformed into Spanish, for example spaghetti, Mussolini and San Pietro are converted to paella, Franco and El Prado. In this regard one is reminded of the Spanish dubbing of the celebrated British sitcom Fawlty Towers, in which Manuel, the waiter from Barcelona, is transformed into Paolo from Naples.

Comparable problems arise when it is the L1 which is the object of attention, for example in Love Actually when the Englishman Colin is in an American bar getting to know some local girls who are fascinated and amused by his southern English accent:

Stacey: That is so funny! What do you call that?
Colin: Er, bottle.
Girls: Bottle!!
Carol-Anne: What about this?
Colin: Er, straw.
Girls: Straw!!
Jeannie: What about this?
Colin: Table.
Jeannie: Table. The same.
Stacey: Oh, it's the same.

Here the humour of the scene hinges on regional accents rather than different languages: in the dubbing the phonetic differences between English and American are dealt with on the lexical axis, albeit rather improbably: bottle is converted to boccia (which usually corresponds to ‘carafe’ or ‘jug’) and straw becomes succhialiquido (an unusual term literally corresponding to ‘liquid sucker’, i.e., an object used for sucking liquid).

When L2 and L3 coincide: Eat, Pray, Love

The fascination that Italian life and culture continue to hold above all for the British and the Americans is reflected in the fact that so many films have been dedicated to the experiences of British and Americans in Italy, whether the protagonists take up residence there or are just passing through. Such films can be divided, albeit simplistically, into two main types. The first is where the protagonist merely scratches the surface of the Italian language and culture, for example Letters to Juliet (Gary Winick, 2010), in which the American and British protagonists make no earnest attempt to speak Italian or learn about Italian culture (though the character played by Vanessa Redgrave is in search of her Italian love of long before). The second type features characters who take active steps to familiarise themselves with the Italian language and/or Italian culture, for example Eat, Pray, Love and Under the Tuscan Sun. For both these types the main characters are dubbed according to the modality native → native described above, that is, native English is transformed into native Italian.

Since in part these films are set in Italy the modality native → native can produce dubbing complications, particularly as regards the second type above, and it often takes a real sleight of hand to make it work. In the film Eat, Pray, Love the character Liz (Julia Roberts), who initially speaks no Italian, has a sojourn in Rome – during which she has Italian lessons and makes a number of friends – before continuing on her travels. As stated, the character is dubbed into native Italian, so the fact that she needs to pull out an Italian-English dictionary every now and then is incongruous, but this is played down in the dubbing by making the character recite names of celebrated Italian dishes whenever she does so, a logical move in that she subsequently enthuses about Italian cuisine. This is the first example in the film of the coinciding of L2 (the dubbing language) and L3 (a foreign language in the original film). Then when Liz first meets the Swedish woman Sofi in a bar, her remark ‘your Italian’s fantastic’ is transformed into ti comporti proprio da vera romana [‘you really behave like a true Roman’] and Sofi’s reply ‘I have a wonderful tutor’ becomes ho un cicerone personale, conosce Roma a memoria [‘I have a personal guide, he knows Rome by heart’].

Here the L2-L3 conundrum, as in the sequence described above from A Fish Called Wanda, is easily solved, but this is a different scenario from Otto’s passionate articulations in Italian because (i) the relevant section of Eat, Pray, Love is set in Italy whereas A Fish Called Wanda is set in London, and (ii) Italian is much more explicitly the object of metalinguistic discussion. Indeed the coinciding of L2 and L3 becomes more convoluted when Liz subsequently begins Italian lessons. Here transforming the language being discussed into Spanish (as in A Fish Called Wanda), French etc. is clearly not an option – the idea that an American woman with no knowledge of foreign languages settles in Rome for a few months in order to learn Spanish or French would be decidedly far-fetched, whatever the degree of tolerance of Italian audiences when confronted with foreign influences in film dialogues (Ulrych 1996: 157). In the case in point the strategy adopted is to convert Liz’s meetings with her teacher into lessons on Latin and on the city of Rome. In the source text the Italian lesson includes an explanation of the past tense conjugation of the verb attraversare [‘to cross’], which in the target text becomes:

Original version

Dubbed version

Back translation

Liz: (out of shot) - - -

Liz: I monumenti di Roma sono pieni di frasi latine. Mi aiuti a capire qualcosa?

The monuments of Rome are full of Latin sentences. Would you help me understand?

Giovanni: You can say: Egli attraversò [=‘He crossed’].

Giovanni: Come no! Partiamo dalla prima in classifica: SPQR.

Of course. Let’s start from the most popular: SPQR.

Liz: Egli attraversò [=‘He crossed’].

Liz: Non significa senatus qualcosa?

Doesn’t that mean senatus something?

Giovanni: It’s a past. Noi attraversammo [=‘We crossed’].

Giovanni: Senatus populusque ecc. Sai come lo traduciamo qui a Roma?

Senatus populusque etc. Do you know how we translate this in Rome?

Liz: Noi attraversammo [=‘We crossed’].

Liz: No, come lo traduciamo a Roma?

No, how do we translate this in Rome?

Giovanni: Voi attraversaste, essi attraversarono [=‘You crossed, they crossed’].

Giovanni: Sono porci questi romani. Quale preferisci dei due?

These Romans are pigs. Which do you prefer of the two?

Liz: Too fast. OK.

Liz: La versione originale. Vai avanti.

The original version. Go on.

Aside from the fact that the solution adopted challenges the much-hyped notion that translations tend to be more conservative than originals (see Pavesi 2005: 56-7), it seems to work, except where Liz subsequently expatiates upon the beautiful sounds of Latin. More generally in the film the protagonist’s progress/difficulties in Italian are converted into progress/difficulties concerning the Roman way of doing things, particularly eating. When some time later Liz shows off to her Roman friends by summoning the waiter and ordering a meal in a Roman restaurant, once again she clearly cannot do so in Spanish, French etc. in the target text – the only option is Italian (or perhaps Roman dialect, though that really would stretch the boundaries of credibility), the emphasis falling upon her conversance with local culinary specialities rather than with the Italian language.

As regards directionality, in the dubbing the voice of the character played by Roberts undergoes for the most part the transition native → native, though there is a very restricted smattering of non-native → native, because in the original Liz utters a few phrases in (non-native) Italian which are duly converted to native Italian in the target text. The voices of the Roman characters undergo the transformation non-native → native, that is, they are converted from non-native English to native Italian, with the actors dubbing themselves (the Swedish character Sofi is also converted to native Italian, though she has near-native English anyway). There are thus two different directionalities at work here, and yet, intriguingly, both are characterised by a single underlying precept: the suppression of the foreign. As a result of the modality native → native, Liz’s difficulties as a foreigner in Italy learning Italian are eliminated, and with the conversion non-native → native all the foreign (English) accents have disappeared. These factors, particularly the prophylactic solution of blocking out Liz’s Italian language-learning activities altogether, mean that, at least from a linguistic point of view, the foreignness of the Italian part of this film has been diluted considerably. Before commenting on this methodology let us now compare the strategies adopted in another film recounting the experiences of an American woman in Italy: Under the Tuscan Sun.

When L2 and L3 coincide: Under the Tuscan Sun

The film Under the Tuscan Sun also concerns an American woman with no Italian who goes on holiday to Italy, but in this case the protagonist decides on the spur of the moment to buy a house in Cortona, Tuscany. Most of the dialogue in the source text takes place in English, whether the native English of the protagonist Frances (played by Diane Lane) and of her friends Katherine (Lindsay Duncan), who is English, and Patti (Sandra Oh), who is American, or the non-native English of Frances’ Italian friends, including Marcello (Raoul Bova) and her estate agent (Vincent Riotta), and of Pawel (Pawel Szajda), one of the Polish labourers who are restoring her newly-acquired house. There are several sequences in Italian, produced for example by the elderly owner of the property that Frances then purchases, by an elderly grandmother who complains that her South American e-mail buddy no longer writes to her, by Chiara, a young Italian woman who falls in love with Pawel, and very occasionally by Frances herself. Thus again we have moments when, in Chaume’s terms, L2 (the dubbing language) coincides with L3 (a foreign language in the original film) within a multilingual film. Yet as far as directionality is concerned, it is particularly striking that although Italian is rarely the explicit object of discussion or attention in the dialogues, directionality issues in the Italian version of Under the Tuscan Sun are frequent. Indeed all four modalities of directionality are recurrent:

  1. native → native: as a rule the American and English characters’ native English is dubbed into native Italian, though some exceptions are made, particularly for the protagonist (see below).
  2. non-native → non-native: the young Polish labourer Pawel speaks non-native English and is occasionally dubbed into non-native Italian, although as discussed in (iii) below his exchanges with Frances in English are almost always subtitled into Italian.
  3. native → non-native:  earlier it was underlined that this modality is normally reserved for comic situations but here it is not. Firstly, the histrionic Katherine is basically dubbed into native Italian but with the occasional non-native sound, presumably to emphasise her quaint Englishness. Secondly, there are several moments when Katherine and above all Frances are dubbed in English, with Italian subtitles, by the Italian dubbers themselves. Sometimes it is a question of just a couple of words inserted here and there – which usually results in some very abrupt code-switching – and sometimes several lines in succession. This strategy occurs most often when Frances addresses the Polish Pawel. An example is the following, where Frances discovers Pawel and his girlfriend Chiara in bed together in her bedroom. The dubbed exchange has Italian subtitles:

Original version

Dubbed version

Pawel: I’m sorry Frances.

Pawel: I’m sorry Frances.

Frances: Sorry?!

Frances: [Italian dubber’s voice] Sorry?!

Pawel: We have nowhere else to go to be together!

Pawel: We have nowhere else to go to be together!

Frances: Well what does that make me? Saint Francesca: patron saint of horny teenagers? Pawel, you were doing it in my bed!  I don’t even do it in my bed!

Frances: [Italian dubber’s voice] Well what does that make me? Santa Francesca: patron saint of horny teenagers? Pawel, nel mio letto, in my bed!  I don’t even do it in my bed!

In this and in many other cases Frances’ lines are reproduced in English (sometimes with interjections in Italian, such as nel mio letto above) by the Italian dubbers and with a non-native accent, a move which constitutes a curious instance of the modality native → non-native.

  1. non-native → native: all the English-speaking Italian characters in the original are converted to native Italian in the dubbing, again with the actors dubbing themselves. The sporadic instances of Frances expressing herself in non-native Italian are converted to native Italian.

Code-switching is a salient feature of the dubbing of Under the Tuscan Sun. The exchange between Frances and Pawel above provides a glimpse of this, but a more evident  example is when Frances visits Rome and is importuned by three workmen who chase her through the streets. Desperate to get away from them, she grabs the arm of a passer-by, Marcello (Raoul Bova), and kisses him on the cheek. Once again in the dubbed version there are Italian subtitles for the parts in English:

Original version

Dubbed version

Back translation

Frances: I’ve been looking for you everywhere! You said you were going to meet me, I’ve been waiting for 20 minutes. What am I going to do with you? …

Frances: [Italian dubber’s voice] There you are, I’ve been looking for you everywhere. Hai detto che saresti venuto a prendermi. Ho aspettato venti minuti, ma dove sei finito?


There you are, I’ve been looking for you everywhere. You said you’d come and get me. I waited 20 minutes, where did you get to?

Marcello: - - -

Marcello: Ma lei chi è?

But who are you?

Frances: I’m sorry. Mi scusi, eh? Thank you.

Frances: [Italian dubber’s voice] I’m sorry. Mi scusi eh? Grazie, scusi.


I’m sorry. I’m really sorry.

Thanks, sorry.

Marcello: You just kiss me and now you’re going?

Marcello: Wait, un momento, non si fa così. Mi ha appena baciato!

Wait a minute, you can’t do that! You’ve just kissed me!

Frances: Yes, I’m sorry.

Frances: [Italian dubber’s voice] Yes, I’m sorry.

Yes, I’m sorry.

The brusque code-switching, the recurrent subtitling and the liberal use of all four directionalities in the dubbing of Under the Tuscan Sun are in marked contrast with the strategies used for the dubbing of the Roman part of Eat, Pray, Love, where there is very little code-switching and subtitling, and where just two directionalities predominate: native → native and non-native → native. Overall, the approach adopted in Under the Tuscan Sun informs the dubbed product with a degree of linguistic complexity which is perhaps to be attributed to the wish to retain the foreignness and alterity of the film, by contrast with the Roman part of Eat, Pray, Love, where the foreignness is toned down considerably by the dubbing.

An obvious question at this point is whether such linguistic complexity can be taken on board by the average viewer, and an equally obvious question regards the pros and cons of these two very different methodologies, but in order to suggest some constructive answers we need to consider the issues raised within the broader topic of the tolerance of filmic language and situations in general.

Tolerance thresholds

In the literature on dubbing there is no shortage of references to filmic language – both original and dubbed – as an illusion. According to Perego and Taylor (2012: 71-2):

La doppiezza che caratterizza il dialogo filmico si riscontra anche in altre dicotomie interessanti, tra le quali ricordiamo il fatto di non essere spontaneo ma di dover apparire tale; il fatto di essere permanente ma di dover apparire effimero; il fatto di essere proferito da attori che sembrano parlare naturalmente ma in realtà recitano; e infine il fatto che il pubblico accetta l’illusione linguistica cui è esposto pur nella consapevolezza che si tratti di finzione.

[the dual nature of film language is also to be found within other interesting dichotomies, such as not being spontaneous but having to seem spontaneous, such as being permanent but having to seem temporary, such as being produced by actors who appear to be talking naturally but in reality are acting, and such as the fact that the audience accepts the linguistic illusion despite being aware that it is all fiction].

Lionello (1994:  46) takes the view that ‘il primo falso del doppiaggio è l’originale stesso’ [‘the primary falsity of dubbing is the original film’], and Fink (1994: 34) asserts that ‘il doppiaggio non fa che aggiungere artificio ad artificio, un livello ulteriore di finzione a un testo [‘dubbing simply adds artifice to artifice, constituting a further level of fiction’].’ See also Chiaro (2008: 243).

It is clear that as film viewers we are invited to take on board a series of linguistic illusions already present in the original version. I was interested to note that in two recent English-language films, Night Train to Lisbon (Billie August, 2013) and The Family (Luc Besson, 2013) set respectively in Lisbon and in a village in France, the local people, almost without exception, boast a most impressive command of English. This already stretches belief, as does the fact that when the Portuguese characters of Night Train to Lisbon address each other with no-one else present they do so in English with a Portuguese accent. These are common scenarios – one need only recall war films where for example German officers regularly communicate with each other in accented English – now so tried, tested and accepted that even when situations of this type are far-fetched they may still pass unnoticed. The Family describes the experience of an American mafia boss who becomes an informer and then leaves America with his family for a village in France under the FBI witness protection scheme. A part of the film describes the first day in the nearby school of the two children in the family, one of whom, the fourteen-year old Warren (John D’Leo),  is confronted during the break by two of his new French classmates. Bear in mind that this is not by any stretch of the imagination an international school – it is a small, local school chosen by the FBI for reasons of anonymity:

French classmate: You gonna act the smartarse with us?
Warren: OK. Can we just cut to the chase here? What game are you in? Bullying? Protection? Shakedowns? You got a monopoly or do you divvy up the market? What do you reinvest your dough in?

I am a native speaker of English but on first hearing this sequence I had to replay it. The fast delivery, the New York accent and the generous sprinkling of colloquialisms, plus Warren’s almost expressionless and gesture-free communication, made it difficult to follow. Things improved with the second listening but only by the third did I manage to grasp all of it. Fearing the age factor (the boy is 14, I am considerably older), I played the clip to my final-year Italian university students (target C2 English language level), who were barely able to understand even the drift of the exchange, let alone the individual words. The two local 14/15-year old French boys, however, not only assimilate Warren’s message in a flash but are so completely enraged by it that they beat him up on the spot. Yet despite the total implausibility of all this it is hard to imagine that anyone but a language expert would object to it. It seems legitimate to assume here that the viewer’s tolerance threshold is fairly high.

Similarly, in Under the Tuscan Sun the farmers of Cortona speak fluent English and there are even a couple of elderly country women who can follow rapid English dialogue rather too effortlessly. Thus it is often the case that, in the name of smooth, comprehensible running of a story, we are required as viewers to suspend linguistic disbelief, and within this framework dubbing is just one more element contributing to that suspension of disbelief across the board (Bucaria 2008, Chaume 2012: 16). In consideration of this one wonders why the dubbers of Eat, Pray, Love went to such extraordinary lengths to cover up the fact that an American woman in Rome is learning Italian. Since her Italian lesson (which occupies a good 5-6 minutes of the film) regards the passato remoto / simple past of the verb attraversare, it would not have been an insurmountable to task to convey that the protagonist is used to adopting only the passato prossimo / perfect tense and that therefore she is experiencing a degree of difficulty with the passato remoto, so widespread in Rome and in the south of Italy but much less so elsewhere, and in the dubbing this could have been mixed in with the already present element of learning aspects of Roman life and culture. In another part of the film Liz is given an informal lesson on the Italian habit of gesturing to support the spoken word, but for present purposes this seems to me on more or less the same level as her taking lessons of Italian. If viewers can suspend enough belief to take on board that a character producing native Italian is unaware of the fact that Italians frequently accompany their speech with gestures, then they should be able to handle that same character having difficulty with the passato remoto.

The dubbing strategies characterizing Under the Tuscan Sun, on the other hand, are very different in that the foreignness of especially Frances but also of other characters is constantly foregrounded, is very ‘in-your-face’. As noted above in the analysis of directionality, in the more methodologically complex dubbing of Under the Tuscan Sun non-nativeness is much more salient, combined as it is with code-switching and the recurrent use of subtitles, and it makes sense to ask ourselves whether viewers are able to absorb and accept such strategies. My hypothesis – based on the suspension of disbelief argument above and on the fact we do not need to understand absolutely everything in order to appreciate a film – is that they are, notwithstanding the fact that there are undoubtedly moments when the complexity of the translation issues in Under the Tuscan Sun risks compromising a smooth understanding of the dialogues. During a dinner hosted by a family from Cortona, a man without much English sitting next to Frances is becoming very friendly. The source and target dialogues are as follows:

Original version

Dubbed version

Back translation

Man: Nubile? [=‘Are you single?’].

Man: È sposata?

Are you married?

Frances: Eh?

Frances: No, non più.

No, not any more.

Man: Ahm… Celibate?

Man: Ah no? Meeting? Un appuntamento?

Ah no? Meeting? An appointment?

Frances: Celibate… Celibate? Celibate! No, I mean… Ah, well, no … Well, actually I have to admit it has been a while!

Frances: Meeting… Un appuntamento?

[Man: Sì!]  Un appuntamento! No, insomma … Lei vuole prendere un appuntamento?!

Meeting? An appointment?

[Man: Yes!]  An appointment! No, that is … You’d like a meeting with me?!

Katharine: Celibe in Italian means single. He’s not asking when you last had sex, he’s asking whether or not you’re married.

Katharine: Credo che Alberto voglia dire ‘a date’. Vuole sapere se hai una relazione, ‘date’, ecco, non se vuoi un appuntamento con lui.

I think Alberto means ‘a date’. He wants to know if you have a relationship, a ‘date’, not if you want a meeting with him.

Frances: Thank you. No, I’m not.

Frances: Grazie. No, sono single.

Thankyou. No, I’m single.

In my view the original scene is played out well and makes us laugh, perhaps due especially to Frances’ rather crestfallen mien and to the condescending explanation provided by the very posh Katharine, but the dubbed exchange is bewildering because two characters who are after all speaking to each other in native Italian suddenly hit a lexical blind alley for no good reason, especially as it is the Italian man (rather than the non-Italian woman) who inexplicably finds himself unable to call to mind such commonplace words as relazione (relationship) or fidanzato (partner/boyfriend). Then in the following scene the dubbed Katharine, who had confidently asserted in the above exchange that the man was in reality not asking her out, then comforts Frances – who has confessed that she feels she’s made a fool of herself – by saying not to worry about it, that all Italian men try it on, be they married or unmarried.

Of course there are technical restraints such as timing and lip-synch, but there is no escaping that the dubbed exchange reported above doesn’t make much sense. Further, there is some curious subtitled code-switching immediately before and after the exchange that doesn’t aid comprehension at all, which is a pity because as stated above the original scene is clear and amusing. Nevertheless, the images save the day, enabling the viewer to recognise that some sort of attempt to ‘approach’ Frances has been made.

Transparency vs duplicity

Galassi (1994: 65) takes the view that dubbing ‘spesso è un’acrobazia, talvolta un funambolismo’ [‘is often a matter of acrobatics, sometimes of tightrope walking’]. Proof of this is provided by the directionality strategies adopted in Under the Tuscan Sun and Eat, Pray, Love, though the specific strategies adopted in each film are in stark contrast with each other. The dubbing of Under the Tuscan Sun is characterised by complex directionality (all four directionalities are used), and some of the procedures adopted – in particular where the Italian dubbers code-switch, producing English subtitled into Italian – are elaborate and in any case not entirely consistent. Further, the translation of language mistakes and misunderstandings is clumsy, though on occasions the mistake is judiciously ignored, for example when Marcello points to a little girl eating an ice-cream and proudly announces ‘My nephew!’, which is rendered with ‘Mia nipote!’ [= ‘my niece’]. Overall, the foreignising thrust of the target text is conspicuous: the dubbing is in Italian, but the foreignness of the film is all-pervasive. The directionality strategies adopted in the Roman part of Eat, Pray, Love, on the other hand, tend to block out the foreign: certainly the scenarios of the foreigner struggling with Italian and of Italians coping with English have been removed. It is worth noting in passing that not one of the ten or so Italian students of mine (English level C2) who had seen this film only in Italian had actually realised that the real purpose of the meeting between Liz and Giovanni was an Italian lesson.

If, like Paolinelli (2000: 54), we take the view that ‘l’unico doppiaggio degno di nota è quello che non si nota’ [‘the only dubbing worthy of note is that which cannot be noted’] (see also La Trecchia 1998: 116), then clearly the dubbing strategies privileged in Eat, Pray, Love win the day. The dubbing of this film domesticates otherness (see Danan 1991 and Denton 2000) and runs very smoothly as a result. If on the other hand we take the view that ideally the dubbed version should try as far as possible to reproduce the conceptual and situational content of the original, then the more foreignising dynamics of the dubbing of Under the Tuscan Sun, notwithstanding the occasional hiccup, would be prioritised.


Generally it is true, as Perego and Taylor (2012: 124) affirm, that ‘il doppiaggio non è una forma traduttiva trasparente ma opaca. Consente di manipolare – e addirittura di censurare, come succedeva soprattutto in passato – il messaggio di partenza senza mettere lo spettatore nelle condizioni di verificare seduta stante il contenuto del messaggio originale’ [‘dubbing is not a transparent but an opaque form of translation. It enables manipulation – and even censuring, as happened above all in the past – of the source message while not allowing the audience to verify during the projection the original content’], a truth epitomised by some of the ingenious dubbing strategies adopted in Eat, Pray, Love. It is striking, however, that in this film such virtuosity is triggered by something as circumstantially innocent as a foreign language lesson, and one cannot help wondering if the acrobatics are really necessary. As viewers we all know that in reality Liz doesn’t speak a word of Italian, we all know that two Portuguese people would not normally address each other in English if no-one else can hear them, just as most of us know that rapidly-delivered New York slang will be way beyond the understanding of fifteen-year-olds in a small French village. When we watch films we suspend belief constantly, it is part and parcel of the viewing experience. So in Eat, Pray, Love is it really necessary, simply because Liz is dubbed into native Italian, to block out altogether the notion that she might have some difficulty at least with the finer points of the Italian language (such as the passato remoto), rather than have her extol the euphony of the sounds of Latin or be told jokes about the Romans being pigs? Such radical outcomes are directly contingent upon the directionality strategies dominating this section of the film (native → native, non-native → native), which have the effect of squeezing out the foreign in favour of the local.

What, on the other hand, makes the dubbing of Under the Tuscan Sun so striking is that, despite the undeniable awkwardness of its directionality strategies, it resists opacity, it creates generous gaps in the dubbing screen and allows alterity to come through. Foreignness is screened but not screened, so to speak. And that, one might tentatively propose, is what the celebrated acrobatics of dubbing should be all about.

Alongside the analysis of other factors such as the rendering of culture-bound realia, jokes and lexis, the study of directionality in dubbing affords insights into the domesticating or foreignising thrust of multilingual products, helping us to gauge and to decide to what degree foreignness is toned down or foregrounded in the process of transfer from one language and/or culture to another.


Baccolini, Raffaella, Bollettieri-Bosinelli, Rosamaria and Laura Gavioli (1994) Il doppiaggio; trasposizioni linguistiche e culturali, Bologna, Clueb.

Beeby, Allison (2009) “Directionality” in Baker, Mona and Saldanha, Gabriela (eds) The Routledge Encyclopedia of Translation Studies, London and New York, Routledge: 84-8.

Bollettieri-Bosinelli, Rosamaria (1994) “Doppiaggio e adattamento nel cinema” in Baccolini, Bollettieri-Bosinelli and Gavioli: 13-26.

Bollettieri-Bosinelli, Rosamaria (2002) “Tradurre per il cinema” in Zacchi, Romana and Morini, Massimiliano (eds) Manuali di traduzione dall'inglese, Milano, Mondadori: 76-88.

Bruti, Silvia (2009) “From the US to Rome passing through Paris: accents and dialects in The Aristocats and its Italian dubbed version” in Giorgio Marrano, Michela, Nadiani, Giovanni and Rundle, Christopher (eds) The Translation of Dialects in Multimedia, InTRAlinea special issue: [url=][/url]

Bucaria, Chiara (2008) “Acceptance of the norm or suspension of disbelief? The case of formulaic language in dubbese” in Chiaro, Heiss and Bucaria: 149-163.

Chaume, Frederic (2012) Audiovisual Translation: Dubbing, Manchester, St Jerome.

Chiaro, Delia (2008) “Issues of quality in screen translation: problems and solutions” in Chiaro, Heiss and Bucaria: 241-256.

Chiaro, Delia (2009) “Issues in audiovisual translation” in Munday, Jeremy (ed.) The Routledge Companion to Translation Studies, London and New York, Routledge: 141-165.

Chiaro, Delia, Heiss, Christine and Chiara Bucaria (2008) Between Text and Image: Updating research in screen translation, Amsterdam, John Benjamin. 

Danan, Martine (1991) “Dubbing as an expression of nationalism” in Meta: 36:4, 606-614.

Davies, Alan (2003) The Native Speaker: Myth and Reality, Clevedon, Multilingual Matters.

Denton, John (2000) “The domestication of otherness: film translation and audience intercultural awareness assessment” in Taylor: 145-155.

Fink, Guido (1994) “Essere o non essere: la parola e i suoi codici” in Baccolini, Bollettieri-Bosinelli and Gavioli: 29-39.

Galassi, Gianni (1994) “La norma traviata” in Baccolini, Bollettieri-Bosinelli and Gavioli: 61-70.

Heiss, Christine (2004) “Dubbing multilingual films: a new challenge?”, Meta: 49:1, 208-220.

La Trecchia, Patrizia (1998) “Dubbing: An Italian case study”, Perspectives: Studies in Translatology: 6:1, 113-124.

Lionello, Oreste (1994) “Il falso in doppiaggio” in Baccolini, Bollettieri-Bosinelli and Gavioli: 41-50.

Paolinelli, Mario (2000) “Le regole del doppiaggio ideale” in Taylor: 51-4.

Paolinelli, Mario and Di Fortunato, Eleonora (2005) Tradurre per il doppiaggio: la trasposizione linguistica dell'audiovisivo: teoria e pratica di un'arte imperfetta, Milano, Hoepli.

Pavesi, Maria (2005) La traduzione filmica, Carocci: Roma.

Perego, Elisa and Taylor, Christopher (2012) Tradurre l’audiovisivo, Roma, Carocci.

Pokorn, Nike (2005) Challenging the traditional axioms: Translation into a non-mother tongue, Amsterdam, John Benjamin.

Taylor, Christopher (2000) Tradurre il cinema, Università degli Studi di Trieste, Dipartimento di scienze del linguaggio, dell'interpretazione e della traduzione.

Ulrych, Margherita (1996) “Film dubbing and the translatability of modes of address. Power relations and social distance in The French Lieutenant’s Woman” in Heiss, Christine and Bollettieri-Bosinelli Rosamaria (eds), Traduzione multimediale per il cinema, la televisione e la scena: 137-160.

About the author(s)

Dominic Stewart is Associate Professor in English Language and Translation at the Department of Literature and Philosophy of the University of Trento, Italy. He has published extensively in the field of translation and corpus linguistics, and is the author of Semantic Prosody: A Critical Evaluation (2010) and Translating Tourist Texts from Italian to English as a Foreign Language (2012).

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

©inTRAlinea & Dominic Stewart (2016).
"Screening the foreign: directionality strategies used in the dubbing of Eat, Pray, Love and Under the Tuscan Sun", inTRAlinea Vol. 18.

This article can be freely reproduced under Creative Commons License.
Stable URL:

Go to top of page