Subtitling Today

Subtitling Today:

Shapes and Their Meanings

Edited by

Elisa Perego and Silvia Bruti

Subtitling Today: Shapes and Their Meanings

Edited by Elisa Perego and Silvia Bruti

This book first published 2015

Cambridge Scholars Publishing

Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Copyright © 2015 by Elisa Perego, Silvia Bruti and contributors

All rights for this book reserved. No part of this book may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without
the prior permission of the copyright owner.

ISBN (10): 1-4438-8035-3

ISBN (13): 978-1-4438-8035-0
 

CONTENTS

Introduction ............................................................................................... vii

Chapter One ................................................................................................. 1

Subtitling Today: Forms, Trends, Applications
Elisa Perego and Silvia Bruti

Chapter Two .............................................................................................. 15

Audiovisual Translation and Sociolinguistic Adequacy
Gian Luigi De Rosa

Chapter Three ............................................................................................ 33

Reading Cohesive Structures in Subtitled Films: A Pilot Study
Olli Philippe Lautenbacher

Chapter Four .............................................................................................. 57

The Language of Inspector Montalbano: A Case of Irony in Translation
Mariagrazia De Meo

Chapter Five .............................................................................................. 77

Cultural References in Fansubs: When Translating is a Job for Amateurs
Ornella Lepre

Chapter Six ................................................................................................ 99

The Influence of Shot Changes on Reading Subtitles –
A Preliminary Study
Agnieszka Szarkowska, Izabela Krejtz, Maria Łogińska,
Łukasz Dutka and Krzysztof Krejtz

Chapter Seven.......................................................................................... 119

Real Time Subtitling for the Deaf and Hard of Hearing:
An Introduction to Conference Respeaking
Saveria Arma

 
vi Contents

Chapter Eight ........................................................................................... 135

France’s National Quality Standard for Subtitling for the Deaf
and Hard of Hearing: An Evaluation
Tia Muller

Chapter Nine............................................................................................ 171

Telop and Titles on the Japanese Small Screen
Claire Maree

Chapter Ten ............................................................................................. 189

It Ain’t Over Till the Fat Lady Sings: Subtitling Operas and Operettas
for the DVD Market
Adriana Tortoriello

Chapter Eleven ........................................................................................ 203

Subtitling – From a Chinese Perspective
Dingkun Wang

Chapter Twelve ....................................................................................... 221

Learner Corpus of Subtitles and Subtitler Training
Anna Bączkowska

 
 

INTRODUCTION

ELISA PEREGO AND SILVIA BRUTI

Subtitling, a well-known, established and widespread form of audiovisual

translation, nowadays comes in many forms and accomplishes many
purposes. It enables viewers to access, understand, enjoy, interpret and
remember an audiovisual product, but it can at times guide or limit the
viewer’s interpretative options or achieve a decorative function. It can be
invisible and easing–as most manuals desire–or intrusive and taxing but
still adored especially by niche film fanatics.
Given its many forms, functions and audiences, subtitling has lately been
studied from several evolving perspectives. This collection of contributions
wishes to show at least some, and it does so assembling papers that analyse
aspects of subtitling in several audiovisual genres (ranging from TV series
and variety programmes to operas and operettas, including feature films and
live conferences) and combinations of languages (Chinese, English, Finnish,
French, Italian, Japanese and Polish), and welcoming both traditional and
descriptive frameworks and novel methodological approaches in the field of
Audiovisual Translation (AVT).
The volume includes papers reporting case studies on language transfer
strategies in specific situations, e.g. when the source text is challenging
because it heavily relies on multilingualism and it requires to tackle the
issue of translating sociolinguistic varieties or because the subtitler has to
render irony, culture-specific references, comic nuances in peculiar genres,
source language specificities in very distant target languages. It includes
descriptive papers that offer a state of the art overview on cutting-edge
subtitling methods, such as telop, real time subtitling for the deaf and hard
of hearing, subtitling status and policy in a given European and non-
European country. It also includes papers reporting on and empirical
research assessing subtitle reading in particularly challenging situations or
assessing the effect of specific subtitle features on comprehension. It
tackles the issue of teaching translation for subtitling and the importance
of corpus data to make the teaching process more focused and effective.
The empirical papers offer an invaluable perspective on the changes
subtitling research is undergoing. They both resort to eye tracking
methodology thus following a research path that has recently flourished in

 
viii Introduction

AVT and that is ever more interested in the viewer’s reaction to, and
comprehension of, subtitling vs. the translator’s problem-solving
processes. Such papers emphasise the lack of a solid empirical
methodology in a discipline that has long been mainly descriptive and they
raise methodological issues that need to be tackled in future empirical
research in AVT.

Acknowledgments
This research was partially supported by the University of Trieste
Research Fund FRA 2013 (“Towards an empirical evaluation of
audiovisual translation: A new integrated approach”) awarded to Elisa
Perego. The authors would like to thank Francesca Bozzao for her initial
typesetting work, and Christopher Taylor and Serenella Zanotti for their
support throughout the editing work.

 
 CHAPTER ONE

SUBTITLING TODAY:
FORMS, TRENDS, APPLICATIONS1

ELISA PEREGO2 AND SILVIA BRUTI3

1. Recent Developments in Subtitling

When subtitling began to be practiced (1909 at the cinema, and 1938
on TV according to Ivarsson 2004 and Ivarsson and Carroll 1998), the
main concern of practitioners was to convey the dialogue of the actors to
the audience. Technical and translational problems–e.g. how to place
subtitles on the distribution copies or how to distribute the same film in
different languages–soon arose. This paved the way for scholars to tackle
them in the attempt to offer solutions or to perform systematic analyses on
the evolution of subtitles from intertitles, on their economics, distribution,
principles and conventions. The first manuals for subtitlers started to
appear much later (e.g., Ivarsson and Carroll 1998) along with in-house
unpublished guides outlining specific rules and company policies.
The proliferation of audiovisual media, the need to access original
versions of AV products as soon as possible and the newly-acquired
flexibility of dubbing countries have recently led to an increase in the
volume and the nature of such activity. Furthermore, the idea that subtitles
accomplish a mere translational function and that they are simply “a
translation [appearing] at the bottom of the screen during the scenes of a
motion picture or television show in a foreign language” (Merriam-
Webster Online) is nowadays outdated, or at least too restrictive.
Currently, subtitles come in several forms and their applications are
manifold, and they have further contributed to the recent growth in
                                                            
1
Elisa Perego is the author of paragraph 1 and of the overview on the articles by
Arma et al. in paragraph 2. Silvia Bruti is the author of the remainder of paragraph 2.
2
Università di Trieste, Italy. Email address: eperego@units.it
3
Università di Pisa, Italy. Email address: silvia.bruti@unipi.it
2 Chapter One

subtitling volume all over the globe. The papers collected in this special
issue illustrate such a varied and fluid situation very well.
A first addendum to the traditional definition of subtitling, one which
describes standard interlingual subtitling written in a language different
from the language of the original audiovisual product, would highlight the
presence of a parallel and equally common subtitling form, i.e. intralingual
subtitling, written in the same language as of the original audiovisual
product.
Traditionally, same language subtitling was thought of as a tool to
enable deaf and hard of hearing viewers to access audiovisual products.
Indeed, intralingual subtitling is able to render the dialogue in the same
language along with additional information on the auditory elements of the
soundtrack. As some of the papers show, however, besides being an
invaluable accessible film service (Szarkowska et al. and Muller),
nowadays intralingual subtitling takes different forms and labels, and it
can accomplish several new functions. If produced live, on the fly, through
respeaking techniques, it can be exploited to make AV products other than
films, e.g. conferences (Arma), accessible. If superimposed in post-
productions, especially in Japanese variety programmes, it is known as
telop (Maree) and it accomplishes a peculiar entertaining function for
hearing viewers–it mainly highlights comic hints but at the same time it
contributes to manipulating the source intended meaning and to leading
viewers towards a univocal interpretation of such.
Although very creative uses of intralingual subtitles are possible, this is
certainly best known for its socially relevant and didactic applications–or
at least it has until very recently. Not only do same language subtitles
serve as an aid for deaf or hard of hearing people, but they can also have a
major impact on literacy and reading abilities (making the reading practice
an incidental, automatic, and subconscious part of popular TV
entertainment) (Kothari 1998, 2000; Kothari and Takeda 2000), and on
second language learning and acquisition (d’Ydewalle and Pavakanun
1995; d’Ydewalle and Van de Poel 1999; Kuppens 2010). Standard
subtitles are no less important: beside accomplishing their primary role
(i.e., providing a written translation of the original dialogue), they can be
exploited to “teach, revive and maintain minority languages” (Ivarsson and
Carroll 1998, 7), to distribute art house films from small countries, and to
make AV material accessible before official release. This latter aspect is
very much appreciated in the Chinese world (Wang), which heavily relies
on amateur subtitles (i.e. fansubs, “subtitles made for foreign audiovisual
products in a non-professional environment”, Lepre) to overcome media
censorship often imposed by the government on dubbed productions.
 Subtitling Today: Forms, Trends, Applications 3

Overall, subtitling has evolved and some of its forms have developed
from necessary aids to extra layers added to the original AV product. In
particular, telops (Maree) along with fansubbing (Lepre) are very
particular forms of subtitling which most of all show the extent of
subtitling evolution in terms of functions and conventions. Telops stretch
the traditional idea of same language subtitles to the extent that they no
longer only render what is being said for a deaf audience but they
reproduce part of the dialogue disambiguating it for a hearing audience,
emphasizing it and making it redundant thanks to the graphic conventions
used. Telops do not enable viewers to access, enjoy and interpret an AV
product but they are added to limit the viewer’s interpretative options by
directing his/her attention and by ruling out ambiguity. They also achieve
a decorative function, thus going against the chasteness which has always
been typical of SDH (subtitling for the deaf and hard of hearing) and of
standard subtitling, whose primary feature should be invisibility. Fansubs
can be intrusive and taxing, and they often resort to unconventional
stylistic features which would be unacceptable in professional practice,
which is still more concerned with usability criteria.
Besides coming in different forms and accomplishing several
functions, subtitles have recently been studied from different, new
perspectives. Research on subtitling–just like its forms and uses–has
evolved. From a purely technical and translational perspective, from in-depth
linguistics studies it nowadays includes more modern and interdisciplinary
ways of approaching this subject. The different aspects of subtitling have
been studied via eye tracking and empirical methods, and aspects of
subtitling that are related to its reception, usability, and effectiveness have
attracted the attention of several scholars. This special issue is
representative also in this respect. It includes traditional descriptive papers
reporting case studies on language transfer strategies in specific situations,
e.g. when the source text is challenging because it relies heavily on
multilingualism or the issue of translating sociolinguistic varieties (De
Rosa) needs to be tackled or because the subtitler has to deal with and
render irony (De Meo), culture-specific items (Lepre), comic nuances in
certain genres (opera and operetta, Tortoriello), source language
specificities in a very distant target language (Wang). It includes
descriptive papers that offer a state of the art overview on cutting-edge
subtitling methods, e.g. telop (Maree), real time subtitling for the deaf and
hard of hearing (Arma), subtitling status and policy in a given European
(Muller) and non-European (Wang) country. It tackles the importance of
subtitling corpora to diagnose the competence of subtitle trainees and to
prepare teaching materials (Bączkowska). It also includes papers reporting
4 Chapter One

on empirical research assessing subtitle reading in particularly challenging

situations, e.g. when subtitles are displayed over shot changes, or assessing
the effect of specific subtitle features, such as internal cohesive structure,
on comprehension (Lautenbacher).
The empirical papers offer an invaluable perspective on the changes
subtitling research is undergoing. They both focus on eye tracking
methodology thus following a research path that has recently flourished in
AVT and that is increasingly concerned with the viewer’s reaction to, and
comprehension of, subtitling vs. the translator’s problem-solving processes.
Such papers emphasise the lack of a solid empirical methodology in a
discipline that has long been mainly descriptive and they raise methodological
issues that need to be tackled in future empirical research in AVT.
To conclude, the audiovisual genres tackled in each paper and the
languages involved in the analyses involve a number of languages. AV
genres range from TV series and variety programmes to operas and
operettas, including feature films and live conferences. The languages
examined, as either source or target language, are (in alphabetical order)
Chinese, English, Finnish, French, Italian, Japanese and Polish.

2. The Contributions to the Present Volume

Given the variety of themes tackled by the authors, ordering papers has
not been easy. We have decided to arrange them according to the main
themes they explore, starting from those that deal with more general
matters in translating audiovisual texts and then delving into more
specialised topics.
Gian Luigi De Rosa investigates one of the thorniest problems in
translation, i.e. the rendering of different varieties of language and of
socio-pragmatic elements. His specific focus of attention is on two
Brazilian TV series, Mandrake (2005-2007) and FDP (2012). The former
was broadcast on Sky TV in Italy in 2009 in its dubbed version in Italian
and the latter was subtitled by the students of Portuguese in an MA course
at the University of Salento. The two case studies offer ample material for
extending the reflection to all the texts that share these properties.
De Rosa aims to analyse a series of issues related to the audiovisual
translation of mixed-language texts, i.e. texts that are strongly characterised
by the presence of different varieties and of socio-pragmalinguistic
elements. The investigation focuses particularly on the two aforementioned
TV series, Mandrake and FDP. Both of them partly conform to the
Brazilian Portuguese neo standard, but they also contain a
 Subtitling Today: Forms, Trends, Applications 5

“strongly connoted spoken language with elements from sub-standard

varieties belonging to popular varieties of BP which are used in less
monitored contexts and situations and in order to characterize characters”.

This is possible, the author contends, because TV language tends to

use exaggerated sociolinguistic features in order to more precisely connote
some of the characters. When marked texts need to be translated, a main
strategy seems to emerge, especially in dubbing: choices that are marked
in the original are turned into either a neutralised non-standard or, at best,
into monitored informal speech. Markedness powerfully reflects the
relativity of sociolinguistic situations that are couched in very different
linguistic forms across different lingua-cultural pairs. In this regard, De
Rosa shows that the main difficulty in translating the two Brazilian TV
series revolves around the frequent overlap of the diastratic and diaphasic
dimensions. Choosing popular varieties of Italian cannot be a palatable
solution, in that popular varieties correspond to dialects or regional
varieties of Italian and their use would thus be diatopically marked and, as
such, unacceptably domesticating. A compromise solution needs to be
pursued, i.e. one that combines both neutralisation of markedness and
consequently a standardisation of the text, and the dislocation of
markedness from one dimension of variation to another or from one level
of analysis to another.
Olli Philippe Lautenbacher discusses the results of a small scale pilot
study to ascertain how comprehension takes place when watching a
subtitled film, submitting the same sample with different types of subtitles
(L1, L2 or no subtitles) to three different informant groups. The author
aims to evaluate the combined impact of filmic cohesive structures and
interlinguistic (L1) or intralinguistic (L2) subtitles on the comprehension
process of the audience, presupposing, in line with the main literature on
the topic, that bimodal input, i.e. input deriving from both dialogue and
subtitles (but also from image and subtitles), is profitable for recollection
and comprehension corroboration. The experiment is carried out with a
short excerpt from a French film, which was shown to three groups of
Finnish university students who were asked to answer limited response
items in a questionnaire. Their viewing experience was also recorded by
means of eye tracking measures. The best results were obtained by the
students who watched the extract with Finnish subtitles, with almost 75%
of the answers being satisfactory, against slightly less than 60% for those
who watched the clip with French subtitles and some 40% for those who
watched it with no subtitles. Differences in results between the groups that
were shown intra- and interlinguistic subtitles are actually confined to just
a few questions. Careful multimodal analysis of the various questions
6 Chapter One

suggested that the outcome of the experiment strongly depends on the kind
of cohesive links that are exploited. The types of cohesion can thus be:
minimal cohesion, in which subtitles support the dialogue alone; narrative
cohesion, when several expressions in the excerpt develop the same theme
and thus cohesion may be in the form of more or less explicit redundancy;
and multimodal cohesion, when dialogical elements are sustained by
image and sound. In the first case subtitles in Finnish seem to be helpful,
whereas in the case of narrative cohesion both subtitles produce
comparable results. When multimodal cohesion is involved, a distinction
needs to be drawn, i.e. when an expression is completely supported by the
audiovisual mode, so much so that the images repeat the content of the
utterance, both the L1 and L2 subtitles strongly support the viewer’s
understanding; when, instead, cohesion is only partial, the understanding
hinges more on the linguistic utterance, so the students who could access
the L1 subtitles had better results.
An equally challenging topic, the translation of irony, is the object of
Mariagrazia De Meo’s contribution. Translating irony is always
problematic in that it is a phenomenon that aims at producing an emotive
response in the audience, who need to be actively involved. De Meo
chooses an inductive and descriptive approach to investigate the
translation of verbal irony in the English subtitles of the detective TV
series Il Commissario Montalbano, a successful series based on the
eponymous novels by Sicilian author Andrea Camilleri. Much of the
fortune of both novels and TV series rests on the main protagonist,
Montalbano, a fractious Sicilian detective who works in the police force of
Vigata, an imaginary Sicilian town, lives a single life, is a gourmet, and a
long-distance swimmer with a wonderful ocean-front house. He is often
confronted with puzzling crimes that necessitate his wits, stamina and a
special ability to deal with bureaucratic and political pressures that in most
cases require him to close the case quickly, without stepping on the wrong
toes. Salvo Montalbano’s speech, in line with Camilleri’s own literary
jargon, is a mixture of standard Italian and Sicilian dialect, heavily imbued
with ironic remarks and tones.
After reviewing the main approaches to irony in translation studies, De
Meo opts for a dynamic and pragmatic approach and underlines the double
role of the translator as interpreter and ironist, whose main task is to re-
codify and re-contextualise the ironic triggers for the target audience. The
analysis highlights the fact that both metafunctional and structural ironic
triggers are generally translated, included the case of echoic utterances.
Repetitions, which are usually avoided in subtitles as redundant elements,
are instead retained also when they are not the main ironic triggers, but
 Subtitling Today: Forms, Trends, Applications 7

simple ironic cues in the unrolling of dialogue. The subtitler pays careful
attention to guarantee maximum effect with minimum effort, relying in
part on the support of the paralinguistic and prosodic features of the text
and on the audience familiarity with Montalbano’s patterns of behaviour
(e.g. the target of his ironic remarks are more or less always the same
characters).
Specific types of subtitles and some related problems are investigated
in the contributions by Lepre, Tortoriello and Maree. Lepre takes a close
look at fansubs, an increasingly popular and globally spread type of
amateur translation and aims to observe whether the strategy adopted by
fansubbers in translating cultural references differs from that of
professionals; Tortoriello is concerned with DVD subtitles for operas and
operettas, a genre which was in the past destined for connoisseurs, but
which, thanks to better availability, now reaches the more general public;
and Maree reports on the use of telops in Japanese television.
Adriana Tortoriello aims at analysing the distinguishing features of a
rather new product, i.e. operas that are filmed during a live performance in
order to produce a DVD which is later subtitled, intra- and/or
interlinguistically. This novel type of opera subtitling is distinct both from
live opera surtitling and from more conventional types of subtitling for
DVDs.
The starting point of the contribution is a careful account of the nature
of opera subtitles for the DVD: if opera surtitles are not available either
before or after the opera itself is staged, DVD subtitles can rely on better
viewing conditions and can be read several times, as the DVD can be
stopped and replayed according to the viewer’s liking. Furthermore, they
seem to be more similar to fansubs than to both traditional subtitles and
opera surtitles, thanks to their length, the use of repetition and the
tendency to repeat the features of the source text especially in the rhyming
pattern of the lyrics. Tortoriello also explores another sub-genre, the
operetta, or light opera, which developed around the middle of the 19th
century, with different themes and different language features. Operettas
can be produced in two alternate ways, i.e. in the original language with
surtitles, or having the libretto translated and adapted and then producing
the opera in the language of the audience, as happens with the English
National Opera, to which DVD subtitles could be added as a third,
compromise solution. They provide a written and thus more permanent
text but at the same time they allow the audience to follow the opera in the
language in which it was written. In the case of operettas, whose content is
light and includes humour and satire, longer subtitles are more helpful in
making sure the audience captures the gist of the message.
8 Chapter One

The author offers examples of how an operetta, The Mikado (1885,

Sullivan and Gilbert) was both intra- and interlingually subtitled for a
DVD release based on its Australian staging. The various subtitlers
(Tortoriello was responsible for the Italian subtitles) worked together to
give the final product a certain consistency despite language differences.
The analysis of various examples point to some characterising elements:
first of all DVD subtitles for the opera/operetta need to take into account
the so-called “musical constraint”, entailing issues of rhyming and rhythm
requirement; secondly, they are produced for a different kind of audience,
with better viewing conditions and higher reading speed, and aim to render
the text more closely; but perhaps, more importantly, they need to consider
the presence of the theatre audience in the filmed text, which leaves traces
in terms of reactions to both the performance and to the surtitles that were
projected (to which the subtitler may also have no access). As for the
content, the nature of the examples analysed proves the necessity of
carefully considering the farcical nature of the plot, which needs to be
understood, and the tendency to update content whenever possible but to
leave out very specific cultural (Australian) references that, given the
Japanese setting of the operetta, would probably puzzle the audience. Also
of relevance are the metareferences to the act of translation, which testify
to the rather “subversive” nature of these subtitles (Nornes 1999).
Still devoted to a form of subtitling that was, and still is, considered
abusive and subversive, is the contribution by Ornella Lepre, who deals
with fansubbing. Fansubbing developed out of a practice that fans adopted
to translate and spread Japanese anime, but it was soon extended to all
genres (with a special preference for TV series, which are the object of
investigation of this article) and soon acquired gigantic proportions. The
growth of the phenomenon–Lepre argues–brought about a decisive
improvement in the quality of translations, as communities of amateur
subtitlers became more organised and devoted more attention to quality
control. In particular, Lepre investigates, both quantitatively and
qualitatively, how cultural references–adopting Pedersens’ (2005: 2)
definition of “extralinguistic culture-bound reference” (ECR)–are rendered
in both fansubs and official translations of two non consecutive seasons
(for a total of 43 episodes) of the US comedy series 30 Rock, one which is
particularly rich in cultural references (with an average of 45 per episode).
Cultural references were first identified, then classified into eleven
categories, drawing on previously proposed taxonomies; their translations
were also classified as either source language- or target language-oriented
and examined for frequency by using statistical instruments, e.g.
 Subtitling Today: Forms, Trends, Applications 9

“regression models that estimate how one or more independent variables

affect a dependent variable, by estimating the parameters that define their
relationship”.

For example, the category of measure has the largest positive

coefficient, which means that, quite expectedly, ECRs referring to this
topic are usually adapted for the target language. Conversely, categories
such as people, fiction and geo display the largest negative coefficients,
which means that they are less likely to be adapted than other categories.
In order to compare the translating strategies used by fansubbers with
those adopted by professional translators, Lepre used the translation of the
same ECRs in the dubbed version of the series, as it has not been subtitled
in Italian. Results for dubbing are on the whole in line with those of
fansubs, yet there is variation in time, from season 1 to season 3, in the
direction of an evident decrease in target language procedures. Although
more data are needed to corroborate the emerging trend, the research has
highlighted that, as SL-oriented procedures appeared in fansubs first and
only later in official translations, fansubs are very likely to provide
guidelines and preferable trends in audiovisual translation.
Another recently emerged form of subtitles, i.e. telops, is the object of
Claire Maree’s paper. As the author makes clear, heavy use of texts and
graphics of a variety of colours, sizes and fonts, which may slide into view
diagonally or also pop up from a celebrity’s talking mouth, is a major
characteristic of Japanese variety programming. Telops are thus a form of
inscription of text onto the screen, which derives from television opaque
projector.
Although applied linguistics research has evidenced that text on screen
common to Japanese TV serves the purpose of summarising, framing,
highlighting or also embellishing the action which is taking place, telop
seem to have taken on the more specific function of the tabloidisation of
news broadcasts and the highlighting of comic content in entertainment
broadcasting, by selecting only some of the talk enunciated on screen.
They tend in fact to capture and maintain the viewers’ attention through
engaging their gaze. In the course of time they have expanded from
sporadic text in white lettering with a black edge to more varied forms that
exploit a vast array of animations, orthographic variants, colours, symbols,
special effects and graphics, including pictures and emoticons.
By analysing a set of examples coming from different genres, Maree
shows how text-on-screen successfully constructs salient “media personas
anchored in identifiable social identities” and gets rid of those which are
not in line with the favoured ideological scheme. The exaggerated use of
10 Chapter One

non standard orthography and non-normative speech forms are

instrumental to depicting language ideologies.
The contribution by Wang also offers an interesting and up-to-date
account of the fortunes of subtitling in China, a country where both the
production and interest in audiovisual translation and in subtitling in
particular are rocketing to meet the ever-increasing demand of the
audience to access the latest productions of world cinema. Domestic
productions have also increased dramatically, at the expenses of
Hollywood films, mainly because of the strict censorship the government
imposes. Yet, this strict control has largely favoured audiovisual piracy,
both because it eschews the lengthy process of approval and also because
the products are much cheaper. The author reports that one of the most
popular ways of accessing audiovisual material is downloading it at
Internet cafés, which may or may not be legally authorised. Fansubbing
was born in 2003, but it is by now a well-established phenomenon that
often turns into a real business: YYeTs, the largest Chinese fansubbing
group, is turning form a fansubbing group to a sanctioned provider of
translation services. Before the growth of fansubbing, however, dubbing
was also used, but contemporary viewers often find it at odds with the
visuals and prefer subtitling, despite the many difficulties involved by the
linguistic fragmentation and the still high rate of illiteracy among the
masses, especially in rural areas. Other complications arise from the
Chinese writing system, in which each character is semantically
independent but can also combine with other characters to form words and
sentences.
In order to widen the availability and the quality of subtitling, the author
advocates more scholarly attention to audiovisual translation from Chinese
researchers and possible contacts with foreign academics to establish
common terminology and procedural rules, especially for subtitling.
Three contributions in the volume deal specifically with different
aspects and applications of intralingual subtitling, and one deals with
audio subtitling, i.e. all forms of accessible translation.
In her paper, Saveria Arma gives a detailed account of quite a new and
very interesting type of real time subtitling for the deaf and hard of
hearing, i.e., conference respeaking generated by speech-to-text (STT)
technology. STT refers to the translation of spoken words into text. The
spoken words are those uttered by a conference participant or by a
professional operator, the respeaker, who listens to the source text and re-
narrates it, condensing and rephrasing it. Such vocal input is then
transformed into written words which take the form of subtitles and make
the event accessible to a specific audience (persons with hearing
 Subtitling Today: Forms, Trends, Applications 11

impairments) which is also a very heterogeneous one in terms of

background knowledge, education, reading behaviour, and language
competence. The paper illustrates the technical, linguistic and professional
aspects of respeaking-based live subtitling for the deaf and the hard of
hearing in a conference setting–although this technique can be used in a
wide array of different contexts–and it shows that the job of the respeaker
shares a number of aspects with that of the interpreter and of the subtitler.
However, it is quite particular and much more based on the ability to know
and communicate with the target audience without patronising them.
Tia Muller offers a French perspective on subtitling for the deaf and
hard of hearing. In particular, the author thoroughly describes a document
(Charte relative à la qualité du sous-titrage à destination des personnes
sourdes ou malentendantes, that is, in English, the Charter relating to the
quality of subtitles addressed to the deaf or hard of hearing) and all the 16
rules for good subtitling it includes, and she evaluates them
“in relation to SDH addressees’ opinions captured in a 2010 survey, other
European guidelines, and empirical studies, in order to assess the validity
of the components it sets out for all the stakeholders involved”.

Each of the Charte’s 16 rules is analysed resorting to Arnáiz-Uzquiza

(2012)’s typology. On this basis, the author considers pragmatic
parameters (which include the addressees’ characteristics, SDH
production’s aim, the production date, and its authoring), technical
parameters, aesthetic-technical parameters (e.g., those which pertain to the
visual aspects of subtitling and that are a direct consequence of the
production process and of the configuration of the finished product, such
as reading speed and delay in live subtitling), purely aesthetic parameters
(which refer to the purely visual aspects of subtitles, e.g., number of lines,
subtitle placement, box usage, shot changes, font style and size, number of
characters per line, subtitle justification, line spacing, synchrony with the
image), linguistic parameters (editing and segmentation) and
extralinguistic parameters (aspects that represent non-verbal information
included in the audiovisual text, e.g. sound effects, music, paralinguistic
elements and character identification). After the thorough analysis which
places France on the European audiovisual map, the author concludes that
much work is still needed to improve the real effectiveness of the rules
that govern French SDH even after their implementation in 2011.
Szarkowska, I. Krejtz, Łogińska, Dutka, and K. Krejtz present a
preliminary eyetracking study on the influence of shot changes (i.e. cuts)
on reading subtitles. The aim of the study is to demonstrate empirically
whether the presence of a shot change really forces viewers to re-read the
12 Chapter One

subtitles–as maintained in literature–or not. The study has been conducted

in Poland with deaf, hard of hearing and hearing participants watching
two-line subtitles for the deaf and hard of hearing (SDH). Participants
were exposed to short excerpts from various audiovisual material: a
feature film (Love actually, 2003, Richard Curtis) and two documentaries
(Super Size Me, 2004, Morgan Spurlock and Roman Polanski: Wanted and
Desired, 2008, Marina Zenovich). The results, although preliminary, are
counterintuitive and show that, overall, reading subtitles is effective also in
challenging situations, and that the type of programme being watched and
its structural complexity can influence the subtitle reading behaviour of
viewers.
The paper by Bączkowska shows instead some applications of
interlingual subtitling in language teaching. The work illustrates some of
the results of the Learner Corpus of Subtitles (LeCoS) project developed
by the author and some collaborators at Kazimierz Wielki University,
Bydgoszcz, Poland, whose participants are students of English Philology
(MA level) and Modern Languages with an English major and a Russian
minor (BA level). The project aims to identify the preliminary subtitling
competence of modern language students, to prepare materials to be used
to teach subtitling, and develop a complete subtitling module for
translation students. As the name of the project reveals, one of the steps is
the compilation of a corpus of interlingual subtitles produced by Polish
students of Modern Languages, which is structured as an increasable
database. The contribution focuses however on one corpus component,
namely a stand-alone subcorpus (Corpus B), which is qualitatively
analysed. Discussion of data provides ample evidence of overreliance on
the source text, which often results in literal translations and calques, and
excessive lexical and syntactic precision. Students very often take little
notice of the typical diamesic shifts required in subtitles and reproduce
many features of orality, such as interjections, expletives and backchannel
cues. Apart from interesting hints as to the areas where students need more
guidance and training, the paper also shows that corpus data can provide
valuable insight into the translation difficulties and pitfalls that subtitling
trainees are likely to meet, and could thus prove to be beneficial for the
design of subtitling courses.

References
d’Ydewalle, Géry and Pavakanun, Ubolwanna. 1995. “Acquisition of a
Second/Foreign Language by Viewing a Television Program.” In
Psychology of Media in Europe: The State of the Art, Perspectives for
 Subtitling Today: Forms, Trends, Applications 13

the Future, edited by Peter Winterhoff-Spur, 51-64. Opladen,

Germany: Westdeutscher Verlag GmbH.
d’Ydewalle, Géry, and Van de Poel, Marijke. 1999. “Incidental Foreign
Language Acquisition by Children Watching Subtitled Television
Programs.” Journal of Psycholinguistic Research, 28: 227–244.
Ivarsson, Jan. 2004. “A Short Technical History of Subtitles in Europe.”
Retrieved from www.transedit.se
Ivarsson, Jan, and Carroll, Mary. 1998. Subtitling. Simrishamn: TransEdit.
Kothari, Brij. 1998. “Film Songs as Continuing Education: Same
Language Subtitling for Literacy.” Economic and Political Weekly, 33:
2507–2510.
—. 2000. “Same Language Subtitling on Indian Television: Harnessing the
Power of Popular Culture for Literacy”. In Redeveloping
Communication for Social Change: Theory, Practice and Power, edited
by Karin Wilkins, 135–146. New York: Rowman and Littlefield.
Kothari, Brij and Takeda, Joe. 2000. “Same Language Subtitling for
Literacy: Small Change for Colossal Gains.” In Information and
Communication Technology in Development, edited by Subhash
Bhatnagar and Robert Schware, 176–186. New Delhi: Sage.
Kuppens, An. 2010. “Incidental Foreign Language Acquisition from
Media Exposure.” Learning, Media and Technology, 35: 65-86.
Nornes, Abé Mark. 2007. “Cinema Babel: Translating Global Cinema”,
Minneapolis, MN: University of Minnesota Press.
Pedersen, Jan. 2005. “How is Culture Rendered in Subtitles?”. In MuTra:
Challenges of Multidimensional Translation, edited by Heidrun
Gerzymisch-Arbogast and Sandra Nauert. Saarbrücken. 2-6 May 2005.
Accessed 25 January 2013.
www.euroconferences.info/proceedings/2005_Proceedings/2005_Pede
rsen_Jan.pdf
 CHAPTER TWO

AUDIOVISUAL TRANSLATION
AND SOCIOLINGUISTIC ADEQUACY

GIAN LUIGI DE ROSA1

1. Introduction
The language of TV fiction is created and organized (Nencioni 1983)
through a process of reconstruction and representation of language in
context. The language of TV fiction may be intended both as a reproduction
of a face-to-face dialogue (Bazzanella 1994, 2002) and as a variety
transmitted with Multiple Senders – Heterogeneous Receivers (Pavesi 2005;
Rossi 1999, 2002). The dialogues of the Brazilian contemporary TV series
show a number of features of Brazilian Portuguese (hence BP), which
reveal, as can be seen from the structure of the BP language (Fig. 1-1), an
advanced process of re-standardization (Bagno 2005, 2012, De Rosa 2011,
2012, Lucchesi 2004, Perini 2007, 2010).
However, despite the presence of recurrent features, the language used
in TV fiction–such as sit-coms, serials (including costume serials) and
soap operas–shows to be less homogeneous and more varied than film
language. This is due to the fact that each fictional subgenre has different
communicative aims and textual and dialogic features. In fact, texts
characterized by formal to highly formal registers with a tendency towards
standard and/or neo-standard language (mainly in TV fiction) can be found
together with texts which show (sometimes excessive) tendencies towards
sub-standard varieties and features. On the basis of the verisimilitude
agreement between sender and receiver, the process of writing dialogues
should not have external constraints with reference to the text or to the
context, nor should they have external pressures. If this is only partially true

1
Università del Salento, Italy. Email address: gianluigi.derosa@unisalento.it
16 Chapter Two

Fig. 1-1 Architecture of Brazilian Portuguese

 Audiovisual Translation and Sociolinguistic Adequacy 17

for the creation of film language, it is even less true for the language of TV
fiction, where the excessively concentrated use of sub-standard
(stigmatized or non-stigmatized) features is possible because of the hyper-
connotation and hyper-characterization of fictional characters.
Both TV series chosen for analysis use elements and oral varieties
which tend towards the neo-standard BP but they also show a strongly
connotated spoken language with elements from sub-standard varieties
belonging to popular varieties of BP (hence PBP) which are used in less
monitored contexts and situations and in order to characterize characters.
In the series Mandrake, the scriptwriters José Henrique Fonseca, Tony
Bellotto e Felipe Braga, have freely adapted the character created by the
novelist Rubem Fonseca. The series focuses on Mandrake, a criminal
lawyer from Rio de Janeiro and most of the series’ characters speak a sub-
standard variety which makes the original linguistic and stylistic choices
of the writer even more characterized. The second series chosen for
analysis, FDP, tells the vicissitudes of a soccer referee. The original
screenplay has been written by Adriano Civita, Francesco Civita and
Giuliano Cedroni and the series is directed by Katia Lund (among the
others), who is also co-director of Cidade de Deus (2002). The language
used is a neo-standard variety of BP and the use of PBP varieties is only
sporadic.
Despite the presence of some limits due to the strong characterization
of characters, the language of TV Brazilian fictions makes use of neo-
standard, non-standard and sub-standard varieties and elements with the
aim of showing the contemporary diamesic variation which is,
consequently, recognized by the audience as the reproduction of language
in context. The features of the spoken language used can be distinguished
according to those elements which show the existence of a grammar of
spoken language which is different from the grammar of written
language 2 . However, text types, genres and the features of the target
audience (which in the case of networks such as HBO–where the two TV
series are broadcast–is large and heterogeneous) represent two further
extra-linguistic variables which are to be added to the diamesic variation

2
Textual and interactional features of spoken language are frequent hesitations,
interruptions, false starts, editing, repetitions, paraphrases, overlaps, etc. Syntactic
features of spoken language are short sentences, juxtapositions, non-clausal units,
ellipsis; discursive markers are: então, ora at the beginning of a turn or of an
utterance; textual or pragmatic connectives such as isso, aí; attention getters such
as olha. In the lexical domain, there is a wide use of slang and present-day slang
(calão, gíria comum or gíria de grupo), and also of obscene and offensive words,
which are frequent elements of colloquial language.
18 Chapter Two

of language and account for the presence of highly marked elements from
sub-standard language, as in the case of the detective fiction, Mandrake, or
of the “mixed genre” TV series set in the world of soccer such as FDP.

2. Audiovisual Translation and Sociolinguistic Adequacy

Translation is usually defined in terms of transposition from a (variety
of) standard language to another (variety of) standard language and most
of the problems which are object of discussion among translation theorists
imply this default situation (Berruto 2010, 899).
However, if this default situation may be considered arguable in the
domain of literary translation because literary language is more and more
often being characterized by sociolinguistically marked elements and
varieties, it is even more questionable in the domain of audiovisual
translation (Díaz Cintas 2009, Díaz Cintas and Anderman 2009, Pavesi
2005, Perego 2005, Perego and Taylor 2012). In this domain, the
transmitted language varieties, such as the spoken language used in the
Brazilian audiovisual products, show a high degree of diastratic and
diaphasic markedness which is even higher in TV products which tend to
characterize and connotate sociolinguistically some of the characters by
exaggerating factitiously their linguistic and expressive features.
In fact, in many Brazilian TV series (in this case, mainly in the series
Mandrake) the tendency towards a factitious recreation of spontaneous
spoken language is particularly visible in those characters who speak the
PBP variety. The result is the presence of an excessive use of those
features which usually have a lower frequency in spoken language.
From a translational point of view, problems related to the
sociolinguistic adequacy of translation are due to the fact that in Brazilian
TV fiction products, many spoken varieties of BP are used and that sub-
standard elements and/or diastratic and diaphasic varieties are used
alternatively with neo-standard3 language.
“Guardando alla traduzione dalla prospettiva sociolinguistica, il problema
centrale è appunto quello dei testi sociolinguisticamente marcati per la
compresenza di più varietà di lingua, ciascuna delle quali per definizione è
portatrice di significati sociali intrinseci alla comunità linguistica della

3
The standard variety of BP is highly codified and acquired through formal
teaching; it is used only in formal contexts and for some written genres (mainly
academic writing). In practice, the diasystem of the BP has an overt prestige
variety (standard variety), a covert prestige variety (neo-standard variety), and a
sub-system of popular stigmatized varieties.
 Audiovisual Translation and Sociolinguistic Adequacy 19

lingua di partenza. Si tratta quindi della traduzione del significato sociale

associato agli elementi (forme, parole, costrutti) di una lingua che lo
veicolano.” (Berruto, 2010, 900).

On the basis of what Berruto says, by sociolinguistic adequacy is

meant the rendering of “the social meaning of linguistic signs” (2010,
900). The rendering of meaning implies the use of a number of different
strategies because sociolinguistic equivalence may be hardly achieved due
to the differences in the structures of source and target languages (Cf.
Berruto 1987, 1995, 2006).
Indeed, a sociolinguistically equivalent rendering of marked elements
would imply the identification of a similarly marked translation equivalent
along the diaphasic or diastratic continuum or in the dimension of diatopic
variation. This process which should tend towards the naturalization of the
language of the sociolinguistically marked varieties–and not towards the
neutralization–may lead, on the other hand, to the total neutralization of
the degree and type of markedness of translation equivalents, which in this
way, will result standardized in the target text. This latter solution is the
most used in the Italian dubbing of TV series and has, as a result, a target
language which may be seen as a type of dubbese tending towards a neo-
standard or a monitored informal spoken language.
Besides the achievement of sociolinguistic equivalence and the
neutralization of markedness, in the domain of audiovisual translation,
other translation compensatory strategies can be considered and applied
with reference to sociolinguistic adequacy, such as those proposed by
Berruto (2010, 902):
“[S]i può (bi) rendere un elemento marcato nella lingua di partenza per una
certa dimensione di variazione con un elemento marcato nella lingua
d’arrivo per un’altra dimensione di variazione, o, (bii) rendere un elemento
marcato a un certo livello di analisi con un elemento marcato per un altro
livello di analisi, o ancora, (biii),rendere l’elemento marcato nella forma
neutra standard e tradurre in altro punto contiguo del testo un elemento non
marcato nella lingua di partenza con un elemento marcato nella lingua
d’arrivo; con eventuale somma o combinazione di (bi), (bii) e (biii): resa
di un elemento di un certo livello di analisi marcato per una dimensione
mediante un elemento di un altro livello di analisi marcato per un’altra
dimensione di variazione e/o in un altro punto del testo.”
20 Chapter Two

The translation strategies discussed by Berruto propose to create and

keep a variational opposition between the marked element and the rest of
the text4.

2.1. Mandrake and FDP

The two TV series analyzed in this paper show a partially specialized
language. In the Mandrake series, where the main character is a lawyer,
the language of law is frequently used. In the FDP series the main
character is a soccer referee and the use of the language of soccer is
limited. Nevertheless, in both cases explanations and sometimes
trivializing reformulations are used as a narrative strategy and in order to
make the two specialized languages less opaque.
For this reason, the analysis will focus particularly–both in the original
and the subtitled and dubbed versions–on a series of scenes and
communicative situations from Mandrake and FDP where different
varieties of language and situations of multilinguism with code-switching
and code-mixing (BP/Spanish) phenomena are present.
In the first episode of Mandrake, titled A cidade não é aquilo que se vê
do Pão de Açúcar, (TV transposition of O caso de F.A. by Rubem
Fonseca 5 ) different varieties of BP with a certain degree of diatopic,
diastratic and diaphasic markedness are present together with code-
switching and code-mixing phenomena, which are neutralized in the
dubbed version.
However, before considering the translation strategies used by
translators in the adaptation and dubbing of the first episode of Mandrake,
a series of closer analyses are needed. Scenes should first be analyzed
through a macroanalysis of TV fiction conversation, through an analysis of
the historical and geographical context (diatopic variation), and an analysis
of a series of extralinguistic factors and their possible influence on
characters, such as their level of education, job, social status, age and sex
(diastratic variation). Scenes also need to be analyzed through a
microanalysis of TV fiction conversation in order to identify details related
to the situation where the interaction is taking place (socio-situational

4
The strategies indicated by Berruto can be referred to Nida’s concept of dynamic
equivalence (1964) even though they are oriented towards sociolinguistic adequacy.
5
O caso de F.A is included in the collection of tales titled Lúcia McCartney
(1967).
 Audiovisual Translation and Sociolinguistic Adequacy 21

variation) 6 , which involves those pragmatic elements which occur in

spoken language and generate contrasts such as proximity/distantiation,
clarity/opaqueness, power/submission, together with other conversational
strategies used by speakers during the dialogue (Cf. Orletti 2008, Preti
2004, 2008).
In the first segment analyzed, two lawyers are among the participants
of the dialogic interaction: Mandrake and Wrexler, and Jorge Fonseca,
who is the promoter of the girls of the Sun Shine, the place where
Mandrake goes to meet Pâmela, the girl one of his clients is in love with:

Mandrake – Original version

Mandrake: Como é que você me achou aqui?

Jorge Fonseca: O problema não é como é que te achei aqui, é o
que é que eu tô fazen(d)o aqui!
Mandrake: É, o que é que cê tá fazendo aqui?
Jorge Fonseca: Cê tá a fim da Pâmela, não tá?
Mandrake: Tô!
Jorge Fonseca: Yo quiero la plata!!!
Mandrake: La plata é com esse aqui ó!
Wrexler: Aqui não tem plata nenhuma…mas o que é isso
aí?

The first thing to be noticed is that despite some differences in terms of

status and role of the participants, the interaction can be defined
symmetrical. The language spoken by Mandrake is gradually adapted to
the language spoken by Jorginho through a series of conversational
strategies such as the informal pronoun of address VOCÊ which is
replaced by its aphaeretic form CÊ (which can be used only as a subject).
The language spoken by Jorginho reveals his low-intermediate
educational level which reflects the diasystem of urban popular varieties.
Furthermore, the presence of some slang words should be interpreted with
reference to Jorginho’s job and the place where he works and may be
considered as a sort of a sub-code or a jargon.
From a morphophonetic point of view, besides the use of the aphaeretic
form CÊ, it is interesting to notice the non-articulation of the dental voiced
occlusive /d/, which is made agglutinated by the nasal consonant in the
forms of the gerund <fazendo>vs<fazen(d)o>. Conversely, from a
morphosyntactic point of view, the alternation between TU/(VO)CÊ in the

6
Relevant factors of the diaphasic variation are the level of formality or
informality of the communicative situation and the speaker’s degree of attention
and control in the linguistic production.
22 Chapter Two

language spoken by Jorginho, the use of the clitic forms for the direct second
person TU (“O problema não é como é que TE achei aqui…”) associated to
the (VO)CÊ may have the stylistic-expressive function of connoting the
language spoken by Jorginho as sub-standard. However, the interpretation of
this linguistic trait has radically changed over the last thirty years thus
contributing to the restructuration of the forms of the pronominal paradigm
with the implementation of the grammaticalized pronominal forms
VOCÊ/VOCÊS and A GENTE. This restructuration has brought a number
of grammatical implications at different levels. In fact, the neo-standard BP
shows the alternation of the two indirect pronouns of address TU/VOCÊ but
the pronoun TU with the verb in the third person is still considered a marked
element. The alternation may occur both for the syntactic form of subject
and object and for the other syntactic functions with the unstressed forms
and with those forms introduced by a preposition (for example: Eu queria
levar VOCÊ no show/Eu queria TE levar no show; Isso é para TI!/Isso é
para VOCÊ). The way and the frequency this alternation occurs in the
language spoken by Jorginho together with some lexical elements from the
semantic field of sex (Preti 2010) are, obviously, to be considered as part of
the process of characterization and connotation of the character through
language. This process becomes more marked if the alternation is compared
with the constant use of the pronominal form VOCÊ in the language spoken
by Mandrake and Wrexler and when, later in the episode, the alternation will
include the subject pronoun: “Agora se VOCÊ qui ser ficar 3 horas com ela
dá pra fazer por 2.5. Mas o melhor mesmo é TU dormir com a mulher,
entendeu? TU dorme com ela, 4 contos, passa a noite inteira CONTIGO, TE
ama, TE dá beijo na boca, entendeu? Olha nos TEUS olhos e fala meu
amor”. However, it needs to be said that although this alternation is
considered marked, its markedness is now interpreted as a diaphasic
variation and as a feature of the informal BP.
The situation radically changes in the Italian dubbed version, as visible
from the dialogues reported below:

Mandrake –Italian dubbed version

Mandrake: Come ha fatto a trovarmi?

Jorge Fonseca: La domanda non è come ti ho trovato, ma perché
sono venuto qui!
Mandrake: Perché è venuto qui?
Jorge Fonseca: Stai cercando Pamela, giusto?
Mandrake: Giusto!
Jorge Fonseca: E io voglio la grana!!!
Mandrake: La grana ce l’ha lui!
Wrexler: Qui non c’è nessuna grana…e questo cos’è?