Dubbing practices in Europe: localisation beats globalisation1

Frederic Chaume
Universitat Jaume I

Whereas new technologies seem to favour globalisation in many areas of

translation, dubbing shows a reluctance to embrace this trend of globalisa-
tion. Translation memories are now used to make translation easier and
faster all over the world. In the area of audiovisual translation, new subti-
tling software has been developed, which is now widely used among both
practitioners and companies. Also in subtitling, most microtextual practices
(line segmentation, subtitle segmentation, typographical usages, synthesis
of information, etc.) are followed by the majority of professionals. But dub-
bing seems to refuse to bend to homogenisation. Perhaps due to notions of
nationalism and singularity attached to this concept, dubbing still shows
different macro- and microtextual practices in the European countries in
which it is the most popular type of audiovisual translation.
This paper examines four different dubbing practices at a microtex-
tual level -those carried out in Germany, Italy, Spain and France. Before
considering new failed attempts to globalise this practice, and also some
major advances brought by new technologies, the paper focuses on the dif-
ferences in layout, take segmentation and dialogue writing in these four
countries. These differences show that dubbing practices are still very con-
servative, and resistant to change and homogenisation.

1. Introduction: dubbing vs. subtitling in the global world

New technologies seem to favour globalisation in many translation domains

all over the world. Translation memories are now used to shorten transla-
tion processes and avoid repetitive routines in legal, administrative, medical
or scientific translation, to mention just a few areas. In the field of audio-
visual translation new software has been developed for subtitling and is
being used among practitioners and companies, particularly to shorten the
spotting process and to make subtitling a less-time consuming and therefore
more profitable activity. Free software is also available on the web for use
by freelancers, companies or universities, Subtitle Workshop being just one
example. Such software is increasingly used by both translators and transla-
tor trainees, making translation processes more homogeneous across the
globe. Also, beta versions of lucrative software can be downloaded without
charge, enabling translator trainees to become acquainted with the programs
they will be using when they work for subtitling companies. This eventually
results in a more homogeneous and global process, since everybody applies
the same routines, files are compatible all over the world, and professionals
204 Frederic Chaume

complete the same stages to achieve the final product, thereby making
translation more profitable and easier. Moreover, in this specific field of
subtitling, most guidelines for microtextual practices (with respect to line
segmentation, subtitle segmentation, typographical usages, techniques of
information synthesis, etc.) are increasingly being followed by the vast ma-
jority of professionals and teachers. This is due to the excellent manuals in
the market which claim for good subtitling standards (Ivarsson 1992; Ivars-
son and Carroll 1998; Díaz Cintas 2003). Díaz Cintas and Remael (2007)
will become a cornerstone in the field of subtitling in this respect.
However, this is not the case in the field of dubbing. Dubbing still
shows major different macro – and microtextual practices in the European
countries in which it is the most popular type of audiovisual translation, to
say nothing of Asian dubbing countries. In this paper, I examine various
dubbing practices at a microtextual level, and others at a macrotextual level,
followed in France, Germany, Italy, and Spain. I focus on the dubbing
process, text segmentation, and dubbing symbols.

2. The dubbing process

The dubbing process in Western European dubbing countries has been me-
ticulously explained in Whitman (1992). Whereas Germany, Spain and It-
aly generally seem to follow the same routines and use the same human
agents (Spain being the country where fewer human agents are used in the
whole process), France deviates from the norm and constitutes the excep-
tion by using a unique form of text synchronisation, known as la bande
rythmo. Generally speaking, Western European dubbing practices follow
the routine described below.

• A TV channel or a distributor decides to broadcast a foreign audiovis-

ual text (mainly films, TV series, documentaries, cartoons).
• A dubbing studio is charged with the task of dubbing it into the target
language.
• The dubbing studio finds a translator and organises the whole produc-
tion process (dubbing director, dubbing actors, sound engineer, etc.).
• The translator produces a rough translation, although now translators
are increasingly writing dialogues, thus making this activity more prof-
itable by speeding up the process and reducing costs.
• The rough translation is domesticated by a dialogue writer to make the
dialogues sound natural and synchronise the text to the screen charac-
ters’ mouths. These two tasks may also be done by different people.
• The dubbing actors read this synchronised, more natural translation
aloud, under the supervision of the dubbing director and with the help
of the sound engineer. In this process, other people may be involved in
single tasks, such as the dubbing assistant in Germany or the assistente
Dubbing Practices in Europe 205

al doppiaggio (dubbing assistant) in charge of take segmentation and

the working plan in Italy.
• The sound engineer records the dialogues and the dubbing studio sends
the film back to the client with a new dubbing track in which dialogues
(and maybe songs, or special effects) are recorded in the target lan-
guage.

Dubbing actors read the final version of the translation aloud. This final
version is printed and placed on a lectern inside the dubbing cabin. Usually,
the dubbing director has one copy, the sound engineer has another copy and
the dubbing actors have a third copy on the lectern, so that they are free to
make gestures with their head and arms (imitating the screen actors’ body
movements), and they can also breath freely either standing up or sitting on
a stool.
This, however, is not the case in France. Here, the translation is
typed onto the screen in a track called la bande rythmo. This track is added
onto the film for dubbing purposes only and will never be seen by the final
spectator. The track is added at the bottom of the screen – like the old black
box in subtitling – and dialogues and dubbing symbols are typed into it, so
that dubbing actors can read the dialogue lines aloud while they are looking
at the screen and hearing the original actors.
Essentially, dialogue lines run from right to left, like scrolling and
paint-on subtitles. When the text crosses a red (green, black) vertical line in
the middle of the track (barre de précision) it signals to the actors that they
should begin reading. The words are meticulously synchronised, so the ac-
tors have to wait until these words cross the red line to begin speaking. If
there is a pause or a silence between utterances, no dialogue lines will cross
the red line until the pause is finished. In addition, the détecteurs (people
who are responsible for synchronising the dialogues) may also add dubbing
symbols to prepare actors to utter a bilabial in a close-up or to speed up
their performance, for example. We examine dubbing symbols further in
the fourth section of this paper.
In summary, although some minor differences in the dubbing proc-
ess are noted across Europe – especially in the number of human agents
involved – the most significant difference is the French bande rythmo, and
the subsequent time and money invested in trying to improve the dubbing
process. Professionals in Spain, Germany or Italy are somewhat critical of
this way of working, since they claim it is too expensive and time-
consuming because a person is required to re-type the translation and add
dubbing symbols in this new track. They also claim that the process does
not necessarily improve the final product, since dubbing actors could read
both the bande rythmo and a printed version of a translation placed on a
lectern without any discernable difference, as the acting does not depend on
where the dialogue lines are located. However, no one appears to be willing
to abandon their trusted practice. In the French case, as technical advances
206 Frederic Chaume

are being developed and applied to the bande rythmo, we cannot envisage
convergence and the dubbing process will remain localised for many years.

3. Text segmentation

Another important difference is text segmentation. If we agree that a subti-

tle is the subtitling unit, then a take is the dubbing unit. In dubbing, the
translation is divided into segments, i.e. chunks of text called takes. The
takes, also called anelli in Italy or boucles in France, are portions of text
with different lengths and layouts in Western European dubbing countries.
Historically, translations were divided into short portions of text because
dubbing actors had to memorise all the dialogue lines they were to utter.
Dialogues were split into these dubbing units, known as takes, enabling
dubbing actors to concentrate fully on the screen actors’ movements and
mouths. If their lines were too long, actors would have to look down at the
printed version of the translation in order to read their lines aloud. This
would result in their missing the screen actors’ mouth articulation and kine-
sic movements and, eventually, in blatant dischronies regarding isochrony
or equal duration of utterances. With shorter dialogue lines, actors could
memorise short portions of text and therefore give their full attention to the
original actors’ body movements and mouth articulation. This eventually
results in better synchronisation and a faster pace in dubbing sessions. The
faster dialogues can be dubbed and recorded, the lower the dubbing costs
are.
For this reason takes have a limited number of lines. However, this
number varies from country to country. Again, globalisation has not spread
to dubbing practices.

Spain
In Spain the target text is divided every ten lines or eight lines in some
Catalonian dubbing studios. This operation involves certain constraints; if
this were not the case, a computer could carry out the task at no cost. For
instance, a take can be ten lines long, but no one character in the take
should speak for more than five lines. That is to say, a take can have ten
lines when no dubbing actor has more than five lines to dub. Ten-line takes
only occur in the case of two characters speaking five lines each, or three
characters speaking three, four and three lines, respectively, and so forth. At
the same time, if there is a silence or pause in the film where no dialogues
or paralinguistic gestures are heard, the dialogue writer will cut the take and
begin a new one after the silence.
Finally, the audiovisual narrative must also be taken into account.
What in Film Theory are known as audiovisual punctuation marks, i.e.,
wipes, iris, fade-ins, fade-outs, etc., also oblige the dialogue writer to seg-
ment the text, since a new scene or sequence will begin after them. This
will theoretically imply a new mood for dubbing actors, or the introduction
Dubbing Practices in Europe 207

of new dubbing actors for the new characters on screen. An audiovisual

punctuation mark, then, generates a new take. Scene cuts and sequence cuts
marked by these audiovisual devices represent the end of one take and the
beginning of another. Temporal, spatial and narrative cuts in the plot also
require a new take (Chaume, 2004a).
Actors are paid according to the number of takes they dub, accord-
ing to the number of lines they read aloud, or according to the number of
sessions they attend. In Spain, for instance, actors and dubbing directors are
paid according to the number of takes they dub; therefore, the more takes a
translated film has, the more they earn. Thus, dubbing studios prefer longer
takes, since this means that they pay less for the dubbing. In short, whereas
dubbing actors and directors prefer short takes because they are easy to re-
member and record, on the one hand, and because they earn more, dubbing
managers prefer longer takes because their costs are lower for the same
film.
Hence, while all the above-mentioned constraints (total number of
lines, lines per character, silences, scene and sequence cuts, etc.) have to be
taken into account, over-segmenting the text into too many takes must also
be avoided, since managers prefer longer takes and cheaper dubbings. Con-
sequently, a balance must be reached between the number of takes and the
requirements of the audiovisual narrative. Translators and dialogue writers
have to consider all these points in order to submit a product that complies
with both professional constraints and audiovisual narration.
The following example illustrates some of these conventions:

___________________________________________________________
1. 07.46
YURTXENKO (ON) Hola a tots, hola Steve, mira estos són els meus
companys. / Assenteu-vos.
STEVE (OFF) Ens aniria molt bé que ens digueres què saps sobre
les rutes de comerç il·legal en esta zona.
YURTXENKO (ON) Estem en Slavyanka. El tràfic d’animals va cap a
Usserisk o cap a la frontera de la Xina.
STEVE (OFF) D’ací van directament a la Xina, hi ha una carretera
que hi va directa.
___________________________________________________________
2. 08.19
YURTXENKO (ON) Sí, hi ha una carretera, i també una via de tren.
Potser que primer els productes del tigre o del lleopard va-
gen a Usserisk, i que després, a través d’ací, vagen a la
frontera.
STEVE (OFF) Ens centrem en les zones més riques des del punt de
vista de la biodiversitat, zones amb poca regulacio
jurídica, on els caçadors saben que el risc és menor i els
guanys majors.2
____________________________________________________________
208 Frederic Chaume

In the example, the first take has eight lines and no one character speaks
more than five lines. The second take also has eight lines, and the charac-
ters involved have four lines each. There is no audiovisual punctuation
mark that obliges to generate a new take, since the camera focuses Yurtx-
enko for the whole scene.
In Spain, some dubbing studios have developed non-commercial
software to segment the translation automatically. This software usually
counts the total number of lines and characters in the target text, and then
divides the translation according to the maximum number of dialogue lines
allowed per take. This process disregards both the requirements of audio-
visual narration and the film plot. It is merely an easy way to segment the
text automatically, with no consideration for how the action progresses or
how scenes and sequences structure the plot and the way they have been
joined together. Moreover, this software is not available in the market: dub-
bing studios hire a computer science engineer who designs the program.
The software is only used by the dubbing studio that initially commissioned
and paid for it. Again, this is just further evidence to show that dubbing
practices are not globalised, but trapped, isolated and hidden away jealously
within each dubbing studio.

France
In France, takes or boucles can be much longer, but they are usually less
than a minute long. There is no established limit for the number of lines
permitted per take. Instead, dialogue writers meticulously follow audiovis-
ual narrative criteria: action breaks, scene changes, or fade-ins and outs,
segment the target text, and a new take begins after these audiovisual punc-
tuation marks. The French tradition also goes further. It also differs in that
they consider the following actions to be criteria for introducing new takes
as well: a new character comes into the scene (for example, two characters
speaking in a scene are joined by a third character), a character leaves the
scene or there is a flash back or a flash forward.
Like subtitles, takes usually have a time code in and a time code
out. The time in is noted when the first character in the take speaks for the
first time, i.e., begins his or her dialogue lines or utters a paralinguistic sign
(coughs, laughs, etc.). When the last character in the take stops speaking or
uttering suprasegmental features, the time out is noted. The total time dura-
tion is also noted. The following example shows a French boucle.
Dubbing Practices in Europe 209

___________________________________________________
Boucle: 1 de 10:00:00:00 à 10:00:58:22 Durée: 00:00:58:22
___________________________________________________

Akim : (reac)
Mr. Salim : Now watch...
: Watch carefully...
Albert the b : Oooooh!
: Oh!
: Never seen anything like it!
Ali : (reac)
Mr. Salim : And that’s nothing! Wait till you see what’s next!
: Watch...
: what he’s going to do!
: And now he’s going to make a super fast pass!
Ali : (reac)
The Scorpion : (“Yeah! Yeah! We’re awesome!”)
Mr. Salim : See, I told you!
: What do you think?
Albert the b : hhh!
: I think the blues don’t have a chance...
: against...
Mr. Salim : (mm_
Albert the b : against that!
: We have to warn them!
Mr. Salim : _end)
Albert the b : So they are prepared!3
___________________________________________________

Germany
As in the previous cases, it is difficult to generalise too much since conven-
tions can vary from one dubbing studio to another. However, general trends
can be observed in how the target text is segmented. As in Spain, there is a
maximum number of lines per take. German practice does not normally
allow takes with more than five dialogue lines. Regardless of the number of
lines, the translation is also segmented according to time. Thus, takes can
be from 3 to 10 seconds long, and the text is segmented again after a maxi-
mum of 10 seconds.
The German tradition also respects the audiovisual narrative crite-
ria in as far as possible. Therefore, action breaks, scene changes, flash-
backs, fades and audiovisual punctuation marks entail a new segmentation
of the translation, a new take.
As with the French tradition, a take ends when a new character comes into
the scene, or when a character leaves.
The following example shows the layout of a German take.
210 Frederic Chaume

____________________________________________________________
023
KOVAC: (OFF/ON) Sam!
SAM: (OFF/ON) Wir fahren noch mal zu Shelley’s, (.) zur Union
Station (.), zum Park...
KOVAC: (über.) Sam! (..) Sam, der Wagen steht...4
___________________________________________________________

This short review shows that dubbing practices are far from unified or ho-
mogenised in dubbing countries at this time. On the contrary, it seems that
the differences between text segmentation and layout are progressively be-
coming more pronounced. The number of dialogue lines per take varies
from one country to another and even from one dubbing studio to another.
The use of time codes is different in each country. The length of the lines is
different in each country. Time constraints (duration of takes) are different
in each country. In a word, globalisation has not yet reached the dubbing
industry.

4. Dubbing symbols

Another example of heterogeneity surfaces when one considers dubbing

symbols. Dubbing symbols are indications that dialogue writers or dubbing
assistants include in the translation to help dubbing actors imitate the screen
actors’ gestures, paralinguistic signs, pitch, volume, primary qualities of the
voice, etc. The most important ones refer to where the voice comes from:
(ON), (O/S), etc., mean that the actor is speaking on-screen, i.e., that the
audience can see the actor’s face or body, whereas (OFF), (V/O), etc., indi-
cate that the actor is not in field, that the audience cannot see the actor
speaking on screen. Other symbols provide more clues to help the dubbing
actor.

Spain
As with layout, dubbing symbols may even depend on the conventions and
preferences of individual dubbing studios. A few commonly used symbols
are listed below:
• (OFF) Voice Off, voice uttered out of camera
• (ON) Voice uttered On Screen, in field
• (DE) The character has his/her back to the camera
• (DL) The character is far from the camera
• (SB) The character is on screen, but we cannot see his/her mouth
• (G) Paralinguistic gestures
• (A la vez) Two or more characters speak at a time
• (P) (T) (X) A character interrupts another character
Dubbing Practices in Europe 211

• (Ambiente) A general shot (schoolyard, hotel hall, etc.) with no dia-

logues in the original script, but original voices have to be dubbed to
create possible coherent dialogues, predictable social interactions etc.
• (R) Laughter
• (LL) The character cries or weeps
• (ATT) The voice comes from a telephone
• (ATR) The voice comes from a radio
• (ATV/ATTv) The voice comes from a TV set
• / Silence or pause in the original dialogues
• // Long silence or pause in the original dialogues

These symbols, among others, are inserted in between the dialogue lines, so
that dubbing actors know when they must be uttered – like laughter or cry-
ing– or how these dialogue lines have to be uttered – paying attention to lip-
sync in ON voices, interrupting another character, pausing between two
sentences, etc.

France
As in Spain, different dubbing symbols are used in different dubbing stu-
dios. They are inserted in the bande rythmo, and they also help dubbing
actors act.
• ON / OFF are only marked on the bande rythmo
• (réac) Gestures (especially those made with a closed mouth)
• Gestures (especially those uttered with an open mouth) can also be
transcribed: hum, mm, hhh, ahh…
• (rire) Laughter
• (sanglot) Sobbing
• (ambiance) A general shot (schoolyard, hotel hall, etc.) with no dia-
logues in the original script, but original voices have to be dubbed to
create possible coherent dialogues, predictable social interactions etc.

In addition, some studios use particular dubbing symbols to call the actor’s
attention to some important dubbing constraints. Thus, bilabials can be
marked on the bande rythmo with a symbol placed below the consonant, so
that the dubbing actor does not miss the closing of the original actor’s lips.
For example, in a French dubbing studio the symbol of a bone is placed
below those bilabials that are uttered in a close-up, so that the dubbing actor
pays attention to make a good lip-sync at that time.

Italy
Fortunately, in Italy symbols are standardised as they all appear in the col-
lective agreement signed by the dubbing actors’ association and the dub-
bing companies. This goes a long way to facilitating the dubbing actors’
familiarity with dubbing symbols:
• IC On Screen (in campo);
• FC Off Screen (fuori campo);
212 Frederic Chaume

• inIC, finIC Beginning and end of voice on screen, respectively;

• inFC, finFC Beginning and end of voice off screen, respectively;
• SOVR A voice is heard over another voice;
• / Pause between two sentences;
• // Pause between two scenes;
• EFF A voice effect (through a TV set, etc.);
• VOCE Inner monologues, the voice of thoughts, dreams, etc.;
• Ant. The dubbing actor has to begin speaking shortly before the origi-
nal actor begins his or her lines;
• DS The character has his/her back to the camera ;
• Orig. Advises actors and engineers not to dub the original voices, spe-
cial effects, etc.;
• VERSO Gestures (coughing, kissing, crying, whistling...);
• FIATO Gestures (sighing, huffing, puffing);
• RIDE Laughter;
• RIS Smiling, giggling, etc.

Apart from the fact that dubbing symbols are different because they are
abbreviations or words in the target language, which seems reasonable from
the point of view of the industry, the most worrying issue is that not all
countries use the same symbols for the same occurrences, and that some
symbols are used differently in each country. Again, unification of symbols
in dubbing countries is difficult to foresee.
The above examples show the difficulties of homogenising the
process of translation for dubbing. In subtitling, there are also some minor
differences between subtitling companies and countries, but these differ-
ences do not prevent professional translators from working for foreign
companies. The process of adaptation to foreign companies’ conventions is
easier, since the differences between subtitling practices are less significant.
However, in the field of dubbing these differences are so marked that dub-
bing studios cannot easily take on work from outside their own countries.
A German subtitling company located in Berlin can translate from
English into Italian, or an Italian subtitling company based in Rome can
translate from English into Spanish. In contrast, it is not usual to translate
from English into Italian outside Italy for dubbing purposes. Indeed, when a
German dubbing studio receives a translation brief consisting of, for exam-
ple, translating a film from English into Italian, they look for an Italian
partner in Italy to do the dubbing, and vice-versa.
Indeed, in the field of dubbing, historically, the struggle between
globalisation and localisation has taken a different path. Just after talking
movies came onto scene, creating the need for translation, the first attempts
to solve this problem were not at all localised. Multilingual movies, films in
the late twenties and early thirties in which a scene was shot in one lan-
guage, and then, taking advantage of the same sets (and sometimes actors,
and even directors), the same scene was shot in a different language, were
all produced in the United States (Izard 1992). However, the beginnings of
Dubbing Practices in Europe 213

a process of localisation were noted when all European language versions

of these movies were moved to Paris, Joinville. A second localising step
placed all dubbings in their respective target language countries, creating
some local dubbing conventions such as those outlined above. This process
seems to contradict the general trend towards globalisation that can be ob-
served in other translation domains, and is particularly significant when
compared to the subtitling globalisation process; for example, the increas-
ing use of templates in subtitling all over the world allows translators to
work for any subtitling company in the world from home.

5. Attempts at globalisation

Nevertheless, in the age of globalisation, some attempts have been made to

globalise the process of dubbing. In the acclaimed Languages and the Me-
dia conference, which takes place in Berlin every two years, major techno-
logical advances on audiovisual translation are regularly presented. In its 5th
edition in 2004, brand new software was presented by Elio Zarmati, CEO,
VoxWorks Technologies, Inc. A voice conversion technology called Reel-
Voice was designed by the company to morph any voice into another voice.
The basic idea of the software is to imitate the voice of original actors on
screen. Thus, dialogues in the target language are recorded, and, later on,
the pitch of the voice is manipulated by this software to resemble the origi-
nal screen actor’s voice. The voice of the dubbing actor is morphed to
sound exactly like the original actor while retaining all the characteristics of
the dubbing actor’s performance. The aim of the program is to convey a
voice similar to that of the original actors to the target audience; that is,
since dubbing prevents the target audience from enjoying the real screen
actors’ voices, this software is able to imitate the original actor’s voice, so
that the target audience does not completely miss out on the original voices,
which for the original audience are inescapably linked to the original char-
acters.
Although it must be recognised that the demo shown was excep-
tional in its replication of the original actor’s voice, the experiment in itself
turned out to be a bit of a let-down. The idea of replicating the original
voices focused all the attention on the voices. Equivalence was set accord-
ing to voices. But the effect of imitating the original voices made the target
product sound artificial, strange, and foreign. Dubbing is in itself a domes-
ticating kind of translation, especially compared to subtitling. In the demo,
the imitation of voices, pitches and tones, which did not comply with the
dubbing tradition in the target language (in this case, Spanish), took the
audience outside the film, reminded them that they were listening to a trans-
lation, and made them aware that the product was a palimpsest. The voices
sounded unnatural, exaggerated, overacted. The imitation itself resulted in a
foreignising device which deviated from the (implicit) norms of acting in
the dubbing target culture.
214 Frederic Chaume

Fandubs are another technological innovation that have opened up

a new route to globalisation. Fandubs are dubbings of audiovisual texts
broadcast on the web, made by fans. Like fansubs, fandubs are mostly cre-
ated for anime, Japanese cartoons that are not yet being broadcast in a cer-
tain target culture. Fans watch their favourite cartoons on the web and then
produce a dubbed version at home, using free software (see below) and car-
rying out the whole dubbing process: translating and writing the dialogues,
lip-sync (Chaume 2004b), revoicing the dialogues, recording them in a new
soundtrack, and finally uploading the new product back on the web. Al-
though Japanese anime is the most popular fandub genre, fandubs of other
genres can be found, such as sitcoms or TV series. The software used by
fandubbers to separate the sound tracks from the original audiovisual text,
and then record the fan’s own voice on a new virgin track can be, among
others, Virtual Dub or the more professional software, Adobe Premiere.
This makes the whole process of fandubbing homogeneous, and the results
very similar worldwide. It should be emphasised, however, that fandubs are
not a commercial threat to the dubbing industry. The two can coexist with-
out any risk to the traditional dubbing industry. Furthermore, the appear-
ance of a popular fandub on the web may be a signal for the dubbing indus-
try to make a professional dubbing of the same source text and to distribute
it in the target culture in which the fandub was originally made. Fandubs, in
this sense, can act as a catalyst for professional dubbing.
New York University has also developed an application known as
Video Rewrite. Video Rewrite uses existing footage to create a new clip of
a person mouthing words that she did not speak in the original footage. The
application automatically labels the phonemes in the training data and in the
new audio track, and then reorders the mouth images in the training footage
to match the phoneme sequence of the new audio track. When particular
phonemes are unavailable in the training footage, Video Rewrite selects the
closest approximations. More information can be obtained from the Video
Rewrite webpage.
In the field of teaching (Chaume 2003), new software can also help
students to be more aware of the dubbing process. As recently as five years
ago, students had to read aloud their translation in order to check whether
their target text matched the original actors’ mouth movements. These days,
software can be used to check the lip-syncing process. Windows Movie
Maker, for example, free software which is now available to all Windows
users, allows the student to cancel the original soundtrack, and to record his
or her own translation on a new soundtrack following the screen actors’
mouth articulation and movements,. Once finished, students can check
whether their recorded translation fits the original actors’ mouth move-
ments by playing the clip with their new soundtrack. If the translation fits,
the dubbing is satisfactory and he or she can go on to translate and lip-sync
the next utterance. If it does not fit, the student can try rephrasing the trans-
lation, shortening the sentence, or making it longer, or maybe adding some
bilabials to match the closing movements of the original actor’s mouth. Al-
Dubbing Practices in Europe 215

though this cannot be considered a globalising trend in the field of profes-

sional dubbing, this kind of software can undeniably homogenise the proc-
ess of training in academic settings.
Academic proposals of dubbing standards, as far as translation
quality is concerned, can also be understood as a globalising trend attempt-
ing to make this type of translation homogeneous. As well as Dries’ guide-
lines (1995), I presented an open list of dubbing standards which can be
summarised as follows (Chaume 2006: 5-12):
• good lip-sync (‘good’ depending on the degree of tolerance in each
dubbing culture);
• credible spontaneous dialogues (but not too spontaneous, since while
the language of dubbing pretends to be spontaneous, it is very norma-
tive indeed);
• coherent translation (coherent as a global text, that is, coherent from a
linguistic point of view, but also coherent with the visuals);
• equivalence to the source text (equivalence as defined by Toury 1995);
• technical rigour, for example, avoiding noises in the recording, hearing
clearly audible and distinct voices, etc.;
• credible acting, i.e., neither overacting, nor underacting.

These standards, understood as ideal objectives, affect the translation itself,

the technical recording process, and the dramatisation in the acting, but they
do not influence the dialogue writing process, splitting the translation into
takes, the layout of the translation, or the use of dubbing symbols.

6. Some final remarks

An initial conclusion can be drawn: dubbing countries follow different con-

ventions in how they split the translation into takes, prepare the translation
layout, and use dubbing symbols. While translation for subtitling is becom-
ing increasingly homogeneous and globalised in terms of layout and subti-
tling conventions, translation for dubbing is still heterogeneous and local-
ised in these terms.
This raises the following question: does dubbing refuse to bend be-
fore globalisation? Is it a question of conscious resistance? In my view, it is
not a question of resistance to globalisation. The dubbing industry has his-
torically been limited to a small number of dubbing studios that have fo-
cused their attention on the art of dubbing, rather than on technological ad-
vances. Dubbing studios have concentrated their efforts on acting, direction
of actors, and lip-sync. For many years, a good dubbing has been regarded
as one in which the acting was neither monotonous nor too exaggerated.
Unfortunately, translation has not been given a starring role in this process.
On the other hand, the dubbing industry is a very conservative sec-
tor, and reluctant to change. One example is the software developed by each
individual Spanish dubbing studio to segment the translation into takes
216 Frederic Chaume

automatically. Or the software used in France to type the translation and

dubbing symbols on the bande rythmo. This software is not available on the
market; it is commissioned by a single studio and only used there. More-
over, translators do not have their own copy of the software; as a rule, they
simply submit the translation and someone in the dubbing studio uses the
software to segment it. Dubbing companies are afraid of industrial espio-
nage; they fear that other companies may take advantage of their findings.
Finally, little research on dubbing technology is being carried out in univer-
sities. This also obstructs the flow of information among dubbing compa-
nies.
Sticking to old practices and being impervious to change can also
be understood as a planned agenda to prevent academic intrusion. In my
opinion, a fear of academic intrusion is evidenced by the fact that some
common sense changes have not been implemented. For example, I firmly
believe that the task of dialogue writing can be better done by the transla-
tors themselves. Whitman (1992) also claimed that these two tasks are best
carried out by just one person. Dialogue writers manipulate the draft trans-
lation submitted by translators in order to make the translation sound
fresher and more natural, and also to attain lip-sync. However, dialogue
writers do not always master the source language. While they are very good
professionals in what they do and have an excellent command of the target
language, especially regarding its oral registers, they can completely change
the meaning of the film for the sake of naturalness and lip-sync. Moreover,
in economic terms, the whole process would be much faster and cheaper if
both the translation and the dialogue writing were commissioned to just one
individual.
On the other hand, translators are trained in universities, whereas
dialogue writers have no official training. They learn their trade by watch-
ing other professionals. Courses on dialogue writing for dubbing are very
scarce. This increases the risk of atomisation, in the sense that each profes-
sional and each dubbing studio have their own conventions. By contrast, in
universities translators can be trained, with more or less success, not only
with respect to translation strategies and techniques, but also in new techno-
logical advances. If translators were allowed to do the dialogue writing,
they would share their know-how and experience more easily. Indeed, in-
formation flows better among translators than among dialogue writers and
dubbing companies. Moreover, if dialogue writing is taught in universities
this will no doubt change the pace of old localised practices as well. Un-
doubtedly, dialogues written by translators would help to homogenise and
globalise dubbing practices and cast off old local and capricious conven-
tions.
Can nationalism also be blamed for this heterogeneity? As a do-
mesticating practice, dubbing is more sensitive to nationalistic influences
than subtitling, for example. Furthermore, each dubbing country is proud of
its dubbing industry, claiming that dubbing is done better there than in other
dubbing countries; some European dubbing studios despise the perfectionist
Dubbing Practices in Europe 217

lip-sync practised in the dubbings of the United States, for example. How-
ever, this is not a question of nationalism, as dubbing itself is not only a
nationalistic practice (Chaume 2004a). This is more a question of conserva-
tive behaviour, a fear of industrial espionage, and a firm belief that dubbing
is ultimately a question of art and craft, which cannot be improved by any-
one other than the professionals already involved in this industry.

Bibliography

Chaume, Frederic (2003). “Teaching audiovisual translation: Some methodological proposals”.

Pérez González, L. (ed.) Speaking in Tongues: Language across Contexts and Users.
Valencia: Publicacions de la Universitat de Valencia, 271-302.
Chaume, Frederic (2004a). Cine y traducción. Madrid: Cátedra.
Chaume, Frederic (2004b). “Synchronization in dubbing: a translational approach”. Orero, Pilar
(ed.) Topics in audiovisual translation. Amsterdam: Benjamins, 35-52.
Chaume, Frederic (2006). “Los estándares de calidad y la recepción de la traducción audiovisual”.
Puentes, 6. Hacia nuevas investigaciones en la mediación intercultural. Atrio:
Universidad de Granada, 5-12.
Díaz Cintas, Jorge (2003). Teoría y práctica de la subtitulación. Inglés-Español. Barcelona: Ariel.
Díaz Cintas, Jorge and Aline Remael (2007). Audiovisual Translation: Subtitling. Manchester: St.
Jerome.
Dries, Josephine (1995). Dubbing and Subtitling Guidelines for Production and Distribution.
Düsseldorf: The European Institute for the Media.
Ivarsson, Jan (1992). Subtitling for the Media. A Handbook of an Art. Stockolm: Transedit.
Ivarsson, Jan and Mary Carroll (1998). Subtitling. Simrishamn: Transedit.
Izard, Natàlia (1992). La traducció cinematogràfica. Barcelona: Publicacions de la Generalitat de
Catalunya.
Whitman, Candace (1992). Through the Dubbing Glass. Frankfurt: Peter Lang.
Zarmati, Elio (2004). “Who you see is who you hear in any and all languages: A glimpse into the
future of dubbing”. Paper given at the 5th International Conference Languages & The
Media, Berlin, November, 2004.

Filmography

ER - Season 12, Episode 1 (2005) Street Soccer - Episode 19

Canon City Trop d'amour tue l'amour / Hot stuff
USA Australia / France
Dir. Christopher Chulack Dir. Margaret Parkes & Bruno Bligoux

1
This research has been granted and conducted as a part of the research project HUM 2007-
65518/FILO of the Spanish Ministerio de Educación y Ciencia.
2
Although the translation shown here is in Catalan, the same conventions apply to Spanish trans-
lations.
3
I am very grateful to Cécile Carpentier, from Chinkel, Paris, for granting me access to this won-
derful dubbing studio and providing this sample. The excerpt comes from Street Soccer, Epi-
sode 19, and was commissioned by Tele Images Kids.
4
I am most indebted to Matthias Münterfering, from Deutsche Synchron, Berlin, for providing
this sample. The excerpt comes from ER, Episode Title “Canon City”.