John Benjamins Publishing Company

his is a contribution from Handbook of Translation Studies. Volume 1.

Edited by Yves Gambier and Luc van Doorslaer.
© 2010. John Benjamins Publishing Company
his electronic ile may not be altered in any way.
he author(s) of this article is/are permitted to use this PDF ile to generate printed copies to
be used by way of ofprints, for their personal use only.
Permission is granted by the publishers to post this ile on a closed server which is accessible
to members (students and staf) only of the author’s/s’ institute, it is not permitted to post
this PDF on the open internet.
For any other use of this material prior written permission should be obtained from the
publishers or through the Copyright Clearance Center (for USA: www.copyright.com).
Please contact rights@benjamins.nl or consult our website: www.benjamins.com
Tables of Contents, abstracts and guidelines are available at www.benjamins.com
 Subtitling

Jorge Díaz Cintas

Imperial College

In the context of today’s multimedia society, audiovisual translation (AVT) is gaining

great visibility and relevance as a means of fostering communication and dialogue in an
increasingly multicultural and multilingual environment. AVT is the umbrella term used
to refer to the translation of programmes in which the verbal dimension is only one of the
many shaping the communication process. he concurrence of diferent semiotic layers
through the visual (images, written text, gestures) and audio (music, noise, dialogue) chan­
nels makes the translator’s task particularly challenging in this ield. Of the several modes
available to translate audiovisual programmes (Gambier 2003), subtitling is arguably the
most commonly used because it is cheap and fast. Other professional practices are inter­
preting*, voiceover and dubbing*.
By way of deinition, subtitling consists in rendering in writing the translation into
a TL of the original dialogue exchanges uttered by the diferent speakers, as well as of all
other verbal information that is transmitted visually (letters, banners, inserts) or aurally
(lyrics, voices of). Subtitling can be seen as a supplement to the original programme,
which, unlike in dubbing, remains intact in the target culture for all to watch and to hear.
All subtitled programmes are therefore made up of three main components: the original
spoken/written word, the original image and the added subtitles. Subtitlers are expected to
come up with solutions that create the right interaction among these components and they
must take into consideration the fact that viewers have to read the written text at a given
speed whilst also watching the images at the same time. he constraining nature of the
audiovisual environment has always been brought to the fore when discussing this type of
translation, leading scholars in the past to label it as an example of ‘constrained translation’
(Titford 1982) or even ‘a necessary evil’ (Marleau 1982).

1. The technical dimension

Generally speaking, subtitles do not contain more than two lines, are displayed
horizontally – usually at the bottom of the screen though in some countries like Japan they
can also be vertical – and appear in synchrony with the image and dialogue. he synchroni­
sation process is known as spotting, cueing, timing or originating and it may be carried out
by the translators themselves or by technicians who know the subtitling program.
he time a subtitle stays on screen depends both on the speed at which the original
exchange is delivered and on the viewers’ assumed reading speed. Tradition had it that

© 2010. John Benjamins Publishing Company

All rights reserved
 Subtitling 345

the best practice should be based on the so­called ‘6 second rule’ (Díaz Cintas & Remael
2007: 96–99), whereby two full lines of around 35 characters each can be comfortably read
in six seconds. For shorter periods of time, proportional values are automatically calculated
by the subtitling sotware, bearing in mind that no subtitle should stay on screen for less
than one second so as to guarantee that the eye of the viewer can register its presence.
Although these parameters still enjoy some currency in the industry, particularly on
television, the viewers’ increased exposure to reading text on screen and enormous techni­
cal advances in recent decades have brought considerable changes. he sacred rule of hav­
ing a maximum of two lines in a subtitle so as to minimise their impact on the photography
is being broken daily by the emergence of three, four and even ive­liners, notably in the
subtitling being done on the internet. he traditional positioning of subtitles at the bottom
of the screen is also being challenged as they are beginning to be displayed on diferent
parts of the screen.
Likewise, restricting the number of characters per line to 35, 39 or even 43 is not an
important factor anymore. Most professional subtitling programs work now with pixels,
allowing for proportional lettering, which means that subtitlers can write as much text as
possible, depending on the font size being used and the actual space available on screen.
Perhaps surprisingly, viewers’ reading patterns and abilities have not been thoroughly
investigated in subtitling, and there is apparently a general consensus in the profession that
the 6­second rule dictates a rather low reading speed. With the advent of DVD and mobile
technology, the mushrooming of screens around us, and the proliferation of audiovisual
programmes, it seems fair to accept that today’s viewers are ‘better/faster’ audiovisual read­
ers than those of previous generations. Besides the lengthening of lines, shorter exposure
times and faster reading speeds are all a consequence of this belief. It is not uncommon to
keep two­liners for a maximum of 5 seconds, and to apply reading speeds that hover around
the 180 words per minute (wpm) or 15 to 17 characters per second (cps), as opposed to the
traditional 140 wpm or 12 cps. As can be expected, all these technical changes have had a
knock­on efect on the way the actual translation is carried out.

2. The linguistic dimension

Whilst respecting the technical specifications discussed above, subtitles must provide
a semantically adequate account of the SL dialogue. The fact that viewers do not nor­
mally have the possibility of back­tracking to retrieve information has a great impact
in the way subtitles are presented on screen. Ideally, if they are to be easily understood
in the short time available, each subtitle ought to be semantically self­contained and
come across as a coherent, logical and syntactical unit. To boost readability, both spot­
ting and line­breaking ought to be carried out in such a way that words intimately
connected by logic, semantics or grammar should be written on the same line or
subtitle whenever possible.

© 2010. John Benjamins Publishing Company

All rights reserved
346 Jorge Díaz Cintas

Unless speakers deliver their utterances really slowly, reduction is arguably the main
strategy in use by subtitlers. Reductions can be partial, where condensation of the original
is paramount, and total, when part of the message is deleted. In both cases, decisions have
to adhere to the principle of relevance and make sure that no information of vital diegetic
value is deleted. Any solutions should take the iconic information into account and avoid
translating what is explicitly conveyed through the image. Although subtitles cannot trans­
late absolutely everything that is said, they must strive to capture the essence of what is said.
As aptly put by Gottlieb (1998/2001: 247): “In subtitling, the speech act is always in focus;
intentions and efects are more important than isolated lexical elements”.
he transition from oral to written poses certain challenges and raises the question
of whether non­standard speech, like accents and very colloquial traits, can be efectively
rendered in writing. More oten than not, this type of linguistic variation is neutralised
in the subtitles. Swearwords and other taboo expressions are also particularly sensitive to
this media migration as there is the tacit belief that they are more ofensive when starkly
reproduced in text than when verbalised, which in turn tends to lead to the indiscriminate
deletion of most eing and blinding in the TL subtitles.
Because of the concurrent presence of the original soundtrack and the subtitles, and espe­
cially when translating from a well­known language like English or from one linguistically
close to the TL, subtitling inds itself in a particularly vulnerable situation, open to the scrutiny
of anyone with the slightest knowledge of the SL. One strategy used to deal with this is for sub­
titles to follow, as far as possible, the syntactic structure of the source text so as to reinforce the
synchronisation and to preserve the same chronology of events as in the original utterances.
A worrying practice in the industry is the recourse to English as a pivot language to
translate from some languages (Czech or Japanese) into others (French or German) follow­
ing an English translation rather than the original soundtrack. Errors or misunderstand­
ings in the English translation will most likely be replicated in the other languages, and
nuances and interpretations will also be iltered through English.
he imperative of having to synchronise dialogue and subtitles, the need to stay within
a maximum of two lines per subtitle, and the widespread belief that the best subtitles are
the ones that are not noticed, have been frequently invoked to explain why subtitlers cannot
make use of metatextual devices, such as footnotes or glosses, to justify their solutions. How­
ever, this assumption seems to be being challenged by new practices, where glosses inside the
subtitles and explanatory notes on top of the screen are freely used (Díaz Cintas 2005).

3. Types of subtitling

From a technical perspective subtitles can be open, when they are delivered together with
the image and cannot be turned of, as in the cinema, or closed, when they are optional
and can be added to the programme at the viewer’s will, as on most DVDs. he process of

© 2010. John Benjamins Publishing Company

All rights reserved
 Subtitling 347

merging the subtitles with the images has evolved considerably over the years (Ivarsson &
Carroll 1998: 12–19) and today’s main methods are laser, whereby the subtitles are burnt
onto the celluloid, and electronic, whereby the subtitles are projected onto the ilm.
Subtitles can appear on screen as a block and of again, known as pop-on subtitles,
scroll horizontally, or roll-up. According to the time available for preparation, subtitles can
be pre-prepared ahead of the programme’s release, or (semi/real)live if they are produced at
the same time as the programme is being broadcast.
From a linguistic point of view we can distinguish between intralingual subtitles, also
known as captions in American English, where the language of the subtitles and the pro­
gramme coincide and interlingual subtitles, where the spoken/written message of the origi­
nal programme is translated into a TL. Bilingual subtitles are part of the latter category and
are produced in geographical areas where two or more languages are spoken, as in Finland
(Finnish and Swedish) or Jordan (Arabic and Hebrew).
he best known type of intralingual subtitles is aimed at audiences with hearing
impairment and is widely known as subtitling for the deaf and the hard­of­hearing (SDH).
hey are a step forward in guaranteeing greater democratic access to audiovisual pro­
gramming and, in many countries, their output is regulated by legislation. Although they
share many features with standard subtitling, they also make use of some unique attributes
(de Linde & Kay 1999; Neves 2005). On television, they normally change colour depend­
ing on the person who is talking or the emphasis given to certain words within the same
subtitle, whilst on DVD they resort to labels to identify speakers. It is not unusual to
come across subtitles of up to three or even four lines, and accommodate more than one
speaker in the same line. Crucially, not only do they reproduce the speaker’s dialogue, but
they also incorporate paralinguistic information that deaf people cannot access from the
soundtrack, such as the revving of an engine, steps on a staircase, indications concerning
music, laughter, or whispering. heir positioning is also important and they can be let or
right justiied so that speakers can be easily identiied or to indicate where a given sound
is coming from.
hanks to greater social awareness, SDH is one of the forms of audiovisual communi­
cation which has undergone spectacular growth in recent years on all media. In addition
to a higher turnover, with some TV stations subtitling 100% of their output, SDH has also
crossed linguistic barriers and interlingual subtitling for hearing impaired audiences is
now a reality on some DVDs.

4. New trends

Subtitling is so dependent on technology that any technical advances have the potential
to encroach both on the subtitling process from the practitioner’s perspective as well as
on the perception that viewers have of subtitling as a product. In this sense, digitisation

© 2010. John Benjamins Publishing Company

All rights reserved
348 Jorge Díaz Cintas

and the availability of free subtitling sotware on the net have made possible the rise
and consolidation of translation practices like fansubbing, which in turn are having an
incidental efect on how formal conventions are applied. Subtitles have been traditionally
rather humdrum in terms of positioning, font type and layout. his new way of approach­
ing subtitles as part of a budding participatory culture is pushing the boundaries of cre­
ativity and shaking the foundations of traditional subtitling. Only time will tell whether
these conventions put forward by the so­called ‘collective intelligence’ (Lévy 1997) are
just a mere leeting fashion or whether they are the prototype for future subtitling. he
impetus provided by 3D technology may well open the door to more interactivity and
cr3aTVty in subtitling.
Recent developments in voice and speech recognition have made possible the appear­
ance and booming of respeaking as a professional practice to subtitle programmes that
are broadcast (semi/real)live, such as the news or sports. he wider breadth and scope of
genres being distributed audiovisually – corporate videos, scientiic and technical docu­
mentaries with a high level of lexical repetition – makes the incipient use of translation
memory systems and automated translation in subtitling a very promising development.
In terms of research, the didactic potential of subtitling to learn and consolidate a
foreign language has been a particularly active line of enquiry in recent times (Díaz Cintas
2008). In an attempt to bolster their quantitative indings and gain an insight into the cog­
nitive eforts presupposed by reading subtitles, some researchers are resorting to the appli­
cation of new methodologies and tools, like corpus studies and eye­tracking.
Since the late 1990s subtitling has been a most inspiring ield in which to conduct research
and more recently also for netizens to communicate in cyberspace. Far from waning, this inter­
est is still alame and as strong as two decades ago, if not stronger.

References

Díaz Cintas, Jorge. 2005. “Back to the future in subtitling.” In MuTra 2005 – Challenges of Multidimen-
sional Translation: Conference Proceedings, Heidrun Gerzymisch­Arbogast and Sandra Nauert
(eds). www.euroconferences.info/proceedings/2005_Proceedings/2005_DiazCintas_Jorge.pdf
[Accessed 10 May 2010].
Díaz Cintas, Jorge (ed.). 2008. he Didactics of Audiovisual Translation. Amsterdam & Philadelphia:
John Benjamins.
Díaz Cintas, Jorge & Remael, Aline. 2007. Audiovisual Translation: Subtitling. Manchester:
St Jerome.
Gambier, Yves. 2003. “Introduction: Screen transadaptation: Perception and reception.” he Translator
9 (2): 171–189.
Gottlieb, Henrik. [1998] 2001. “Subtitling.” In Routledge Encyclopedia of Translation Studies, Mona
Baker (ed.), 244–248. London & New York: Routledge.
Ivarsson, Jan & Carroll, Mary. 1998. Subtitling. Simrishamn: TransEdit.

© 2010. John Benjamins Publishing Company

All rights reserved
 Subtitling 349

Lévy, Pierre. 1997. Collective Intelligence. Mankind’s Emerging World in Cyberspace. Transl. by Robert
Bononno. Jackson: Perseus Books.
de Linde, Zoe & Kay, Neil. 1999. he Semiotics of Subtitling. Manchester: St Jerome.
Marleau, Lucien. 1982. “Les sous­titres... un mal nécessaire”. Meta 27 (3): 271–285.
Neves, Josélia. 2005. Audiovisual Translation: Subtitling for the Deaf and Hard-of-Hearing. London:
Roehampton University. Ph.D. hesis. http://roehampton.openrepository.com/roehampton/
handle/10142/12580 [Accessed 10 May 2010].
Titford, Christopher. 1982. “Sub­titling: constrained translation”. Lebende Sprachen 27 (3): 113–116.

© 2010. John Benjamins Publishing Company

All rights reserved