INVESTIGATING TRANSLATION

BENJAMINS TRANSLATION LIBRARY

The Benjamins Translation Library aims to stimulate academic research and training
in translation studies, lexicography and terminology. The Library provides a forum
for a variety of approaches (which may sometimes be conflicting) in a historical,
theoretical, applied and pedagogical context. The Library includes scholarly works,
reference books, post-graduate text books and readers in the English language.

ADVISORY BOARD
Jens Allwood (Linguistics, University of Gothenburg)
Morton Benson (Department of Slavic, University of Pennsylvania)
Marilyn Gaddis Rose (CRIT, Binghamton University)
Yves Gambier (Centre for Translation and Interpreting, Turku University)
Daniel Gile (Universit Lumire Lyon 2 and ISIT, Paris)
Ulrich Heid (Computational Linguistics, University of Stuttgart)
Eva Hung (Chinese University of Hong Kong)
W. John Hutchins (Library, University of East Anglia)
Werner Koller (Department of Germanic, Bergen University)
Jos Lambert (Catholic University of Louvain)
Willy Martin (Lexicography, Free University of Amsterdam)
Alan Melby (Linguistics, Brigham Young University)
Makoto Nagao (Electrical Engineering, Kyoto University)
Roda Roberts (School of Translation and Interpreting, University of Ottawa)
Juan C. Sager (Linguistics, Terminology, UMIST, Manchester)
Mara Julia Sainz (Law School, Universidad de la Repblica, Montevideo)
Klaus Schubert (Technical Translation, Fachhochschule Flensburg)
Mary Snell-Hornby (School of Translation & Interpreting, University of Vienna)
Sonja Tirkkonen-Condit (Savonlinna School of Translation Studies, Univ. of Joensuu)
Gideon Toury (M. Bernstein Chair of Translation Theory, Tel Aviv University)
Wolfram Wilss (University of Saarbrcken)
Judith Woodsworth (Mt. Saint Vincent University, Halifax)
Sue Ellen Wright (Applied Linguistics, Kent State University)

Volume 32

Allison Beeby, Doris Ensinger and Marisa Presas (eds.)

Investigating Translation
Selected papers from the 4th International Congress on Translation,
Barcelona, 1998
 INVESTIGATING
TRANSLATION
SELECTED PAPERS FROM THE 4TH INTERNATIONAL
CONGRESS ON TRANSLATION, BARCELONA, 1998

Edited by

ALLISON BEEBY
DORIS ENSINGER
MARISA PRESAS
Universitat Autnoma de Barcelona

JOHN BENJAMINS PUBLISHING COMPANY

AMSTERDAM/PHILADELPHIA
 TM The paper used in this publication meets the minimum requirements of Ameri-
8

can National Standard for Information Sciences Permanence of Paper for

Printed Library Materials, ANSI Z39.48-1984.

Library of Congress Cataloging-in-Publication Data

International Congress on Translation (4th : Barcelona, Spain : 1998)
Investigation translation : selected papers from the 4th International Congress on Translation,
Barcelona, 1998 / edited by Allison Beeby, Doris Ensinger, Marisa Presas.
p. cm. -- (Benjamins translation library, ISSN 0929-7316 ; v. 32)
Includes bibliographical references and index.
1. Translating and interpreting--Congresses. I. Beeby Lonsdale, Allison. II. Ensinger, Doris.
III. Presas, Marisa. IV. Title. V. Series.
P306.I463 2000
418.02--dc21 00-039824
ISBN 90 272 1637 1 (Eur.) / 1 55619 791 8 (US)
2000 - John Benjamins B.V.
No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other
means, without written permission from the publisher.
John Benjamins Publishing Co. P.O.Box 75577 1070 AN Amsterdam The Netherlands
John Benjamins North America P.O.Box 27519 Philadelphia PA 19118-0519 USA
<TARGET
< R/ RE E F"toc"
F DOCINFO

AUTHOR n""z
"neu1">
"god">
"chu">
""tan">

iTITLE "Table of Contents"

SUBJECT "Benjamins Translation Library, Volume 32"

ntro">
d">
n">
zq">

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150" BOOKMARK "Table of Contents">

Table of Contents

Introduction ix

S I
Investigating Translation Paradigms
1. A Fresh Look at Translating 3
Eugene A. Nida
2. Theory and Practice of Translation Studies Revisited: 25 Years
of Translator Training in Europe 13
Albrecht Neubert
3. The Scope of a Communicative Theory of Translation: An
Attempt at Systematisation 27
Zinaida Lvvskaya
4. Living on the Border: Feminine Subjectivity in Translation 37
Pilar Godayol Nogu
5. Translation Theory in Chinese Translations of Buddhist Texts 43
Chu Chi Yu
6. Self-Translation as an Extreme Case of the
Author-Translator-Dialectic 55
Helena Tanqueiro
7. The Degree of Grammatical Complexity in Literary Texts as a
Translation Problem 65
Isabel Garca Izquierdo and Josep Marco Borillo
< R/ RE E FF

"neu2">
"pac">
"sco">
"zab">
"mun">
"mal">
""gon">
"cla"> g
m

za">
l">

vi TABLE OF CONTENTS

S II
Investigating the Translation Process
8. Opportunities in Conference Interpreting Research 77
Daniel Gile
9. The Computer in Empirical Studies for the Didactics of
Translation 91
Wilhelm Neunzig
10. Acquiring Translation Competence: Hypotheses and
Methodological Problems of a Research Project 99
PACTE
11. Translation Strategies and Translation Solutions: Design of a
Teaching Prototype and Empirical Study of its Results 107
Christopher Scott-Tennent, Mara Gonzlez Davies and Fernanda
Rodrguez Torras
12. From Techniques to Types of Solutions 117
Patrick Zabalbeascoa
13. Translation Strategies: Somewhere over the Rainbow 129
Ricardo Muoz Martn

S III
Investigating Translation and Ideology
14. Language Models and Catalan Translation 141
Joaquim Mallafr
15. Dubbing for Catalan Television: The Acceptable Translation 153
Natlia Izard
16. La traduccin del titulo cinematogrco como objeto de
autocensura: El factor religioso 161
Vctor M. Gonzlez Ruiz
17. The Translation of Mass Fiction 171
John Milton
18. La Traduction des contes de fes: Lenfant entre la tradition et
lavenir 181
Ana Maria Clark Peres
<REF
< /R/TREARGET
E "s
"n
FF

"nor">
"mac">
"ada">
"fue">
"par">
"gar">
"ter">
"i"ref">
"toc"> ns

i">

">

TABLE OF CONTENTS vii

S IV
Investigating Translation Receivers
19. What Do We Know About the Target-Text Receiver? 195
Christiane Nord
20. POSI-tive Thinking About Quality in Translator Training in
Finland 213
Rosemary Mackenzie
21. Towards a More Systematic Approach to the Translation of
Advertising Texts 223
Beverly Adab
22. The Translator as Mediator in Advertising Spanish Products in
English-Speaking Markets 235
Adrin Fuentes Luque and Dorothy Kelly
23. Translation as a Component of Software Localization Projects 243
Joan Parra
24. Traduccin de los nombres vernculos ingleses de animales en
los textos de divulgacin cientca 251
Carlos Garrido
25. A Pragmatic Approach to the Description of Phraseology in
Biomedical Texts 261
Maribel Tercedor-Snchez

References 271
Name Index 289
Subject Index 291
<TARGET "intro" DOCINFO

AUTHOR ""

TITLE "Introduction"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

Introduction

The growth of translation in the last 25 years has lead to signicant changes in
theory and practice. There seems to be a common goal amongst translation
scholars to achieve recognition for Translation Studies as an independent
discipline in its own right. Eorts are being made to nd more objective and
scientic research methods. However, as translation is such a complex phenome-
non, dierent studies choose to focus on very diverse aspects, for very dierent
reasons and using a wide variety of paradigms. Despite these dierences, there
is much common ground and in this volume, Investigating Translation, we hope
to illustrate the unity in the variety.
For a long time translation research was based on the translated text, on the
product, and of course, many scholars still take the translation as the object of
their research. Translations are studied as a social artefact and studied synchroni-
cally or diachronically. The purpose of this research may be to apply dierent
paradigms to nd the most fruitful tools for analysing translation. This may lead
to a critical revision of existing paradigms and the development of new ones.
Alternately, the purpose may be to describe norms prospectively, or, in the case
of minority languages, a prescriptive approach may be taken to nd rules for
linguistic normalisation. The paradigms used may include more traditional
approaches, such as literary criticism, contrastive linguistics, discourse analysis
and descriptive translation studies, as well as sociological paradigms, amongst
which one may nd postcolonial criticism and gender studies.
Translation process studies are a relatively new phenomenon. They take the
mental processes of the translator as the object of their study, concentrating on
the skills, knowledge and strategies needed to carry out this process, or they may
focus on the process in which these skills, knowledge and strategies are acquired.
A key question in this approach is the methodology used in the research, so an
immediate aim may be to nd appropriate tools for empirical research. The
ultimate goal is to throw light on the translators black box, so as to dene
models for translation competence and acquiring translation competence. The
main paradigms used in this research are cognitive psychology, psycholinguistics
x INTRODUCTION

and experimental psychology. However, in describing translation strategies

concepts are drawn from discourse analysis and descriptive translation studies.
There is a growing awareness of the ideological inuence of translation,
particularly through the mass media. Scholars who are concerned about this
invisible inuence look at language models and translation norms through
translation products, both written and audio-visual. Their purpose is to determine
retrospectively the translations underlying ideology (capitalist, fascist, racist,
sexist). Paradigms are taken from socio-linguistics, for example, critical discourse
analysis. The tension between the two languages in contact requires concepts
such as foreignising, domesticating, hegemonic and minority discourses.
As translating becomes more and more professional, with far reaching
eects on international relationships and world markets, attention is being paid
to translation receivers and initiators, as well as the priorities and restrictions
within which the translator has to work. The purpose of these studies is aware-
ness raising in relation to these restrictions, to produce more functionally
appropriate translations, to educate the client, or initiator and to provide tools for
quality control. The paradigms used for these studies vary from functional
translation theories, pragmatics, and discourse analysis to terminology.
Many of these developments in Translation Studies have taken place in the
last 25 years, the lifetime of the Facultat de Traducci i dInterpretaci at the
Universitat Autnoma de Barcelona. The growing importance of translation and
Translation Studies in Spain is reected in the number of translation centres
created over the last 25 years. Some of the research carried out in these centres
is reected in this volume, with contributions from Barcelona, Castelln,
Granada, Las Palmas, Vic and Vigo. We have been made even more aware of
the changes by the papers chosen for this publication and rst presented at the
1V Congrs Internacional de Traducci. The papers were selected bearing in
mind this criteria of change and to stimulate communication amongst translation
scholars all over the world, from Brazil to China. The book is divided into four
sections, following the four major lines of research outlined above: (1) Investi-
gating Translation Paradigms; (2) Investigating the Translation Process; (3)
Investigating Translation and Ideology; (4) Investigating Translation Receivers.
In Section I, Investigating Translation Paradigms, the contributions are all
concerned with the search for suitable paradigms. The rst two are privileged
visions of the evolution in translation studies by experts in the eld. Eugene
Nida charts the transition from a linguistic paradigm to a socio-linguistic one.
Albrecht Neubert traces the eect of this paradigm change on training transla-
tors, and stresses the diculties new translation faculties faced in order to be
recognised as academic institutions. The other contributors to this section focus
 INTRODUCTION xi

on dierent aspects of this paradigm shift.

Zinaida Lvvskaya, from a background in the Russian Translation Studies
tradition, focuses on the capacity of a communicative theory of translation to
describe, explain and predict the functioning of translating. Pilar Godayol Nogu
applies deconstructionism and gender studies to the work of feminine subjects in
translation, which she sees as a borderline experience involving dynamic
procedures and tactics in a continuous process of negotiation. Chu Chi Yu
summarises an ancient debate on translation paradigms that ran through the
history of Chinese translations of canonical Buddhist texts: simple translation
versus sophisticated translation. He shows how this debate covered concepts
that are still crucial today, such as, the role of the translator, reader and text type.
Helena Tanqueiro brings new light to the old dichotomy between the creative
author and the imitative translator in her study of self-translation. In this
framework, concepts such as faithfulness and freedom acquire fresh signicance.
Isabel Garca Izquierdo and Josep Marco Borrillo consider the grammatical
complexity of literary texts as a relevant marker of the stylistic intentions of the
author that is often neutralised by the translator. Using systemic-functional
linguistics, they have devised an instrument to measure this complexity in source
and target texts.
Section II, Investigating the Translation Process, begins with Daniel Giles
review of empirical research in conference interpreting. As one of the major
gures in the struggle for a more rigorous approach to research in translation and
interpreting, he points out the strengths and the weaknesses of work to date. He
also outlines areas and problems still pending, categorising them according to the
methodological diculties involved and encouraging young researchers to
contribute.
Wilhelm Neunzigs main concern is to assure the environmental validity of
experimental research in translation. From this point of view he oers a critical
review of the most commonly used methods and instruments and shows how the
computer can be used as a reliable data collection instrument in a normal
working environment. The PACTE group report on their research into Transla-
tion Competence and Acquiring Translation Competence. They are interested in
applying standard procedures in experimental psychology to research in mental
processes in translation. They present their theoretical and working hypotheses
and discuss some of the methodological problems involved in this kind of
research. Christopher Scott-Tennent, Mara Gonzlez Davis and Fernanda
Rodrguez Torras have carried out a pilot study to observe the eects of a
specic training programme to teach explicit translation strategies. Comparisons
between the experimental and the control group showed the eectiveness of the
xii INTRODUCTION

teaching method and provide a basis for future experiments. Patrick Zabalbeascoa
reviews and studies the development of apparently established concepts, such as,
process, technique, method and strategy, defending a more coherent use of
terminology. He suggests how these concepts may be useful to describe certain
phenomena, how they may be used as categories and tools for description and to
make the learning process more ecient. Ricardo Muoz Martn focuses on
complex problem-solving strategies. He has designed a binary decision tree to
help students to focus on certain textual and contextual constraints and develop
potentially optimal solutions.
In Section III, Investigating Translation and Ideology, the authors are
interested in the inuence of ideology on translation. Joaquim Mallafr, one of
the most prestigious Catalan translators, describes the process by which each
translator develops his own language model within a social context. He analyses
two collections of Catalan translations of canonical texts of world literature,
published in the eighties, and relates the translation norms observed to the
ocial language model in force. Natalia Izard describes the use of acceptable
translations when dubbing programmes for Catalan television (TVC). Cultural
and linguistic adaptations are recommended by TVCs style-book so that TV
programmes should seem to have been conceived in Catalan. This domestication
is a defensive reaction from a minority language and is one aspect of the Catalan
governments linguistic policy.
Victor Gonzlez Ruiz studies the translation of lm titles in Francos Spain
(193975) and shows how censorship can be explicit and implicit under a
dictatorship. The lm industry used self-imposed censorship in line with the
ocial Catholic morality to please the State Censorship Boards. John Milton
describes the characteristics of the translation of mass ction for the Clube do
Livro in Brazil. Commercial interests were responsible for many of these
characteristics, but others may be due to the dominant ideology of the military
dictatorship (196489), when the Clube do Livro thrived. Sexual, scatological,
religious and socialist references were neutralised in the translations. Ana
Mara Clarks study of Brazilian translations of fairy tales for children shows a
tendency by the translators to idealise the childs universe. Over-simplication
and a moralising tone indicate the adults image of the child reader, very
dierent from todays children.
Section IV, Investigating Translation Receivers, is dedicated to the
growing eld of research into translation readers. Christiane Nord, one of the
leading gures in functional translation studies, opens this section by making a
distinction between the receiver and the addressee, the abstract notion the
translator has of the receiver. The results of the studies she presents suggest that
</TARGET "intro">

INTRODUCTION xiii

the sender-audience relationship is much more emphasised in Spanish and French

texts than in German.
Rosemary Mackenzie presents the results of a questionnaire given to users
of translation and interpreting (T/I) services in Finland. She draws didactic
conclusions from how the users assessed the quality of these services, their
assessment of T/I training and their suggestions for how training could be
improved. Beverley Adab oers a series of guidelines for translating advertising
within the parameters of globalization and localization. Her work is based on a
study of a corpus of translated texts in English and French. Adrin Fuentes and
Dorothy Kelly are also interested in the translator as mediator in international
advertising. They have analysed a corpus of texts advertising Spanish products
in English-speaking markets, focusing on questions of national stereotypes. Joan
Parra argues that translation as a component of the software localization industry
has specic characteristics that require special attention from theoretical and
applied translation studies. He stresses the need for co-operation between
professionals and translator training centres and suggests that a rst step in this
direction is to collect data on the industry.
The last two papers take a functional approach to terminology in specialised
translation. Carlos Garrido proposes strategies for overcoming problems arising
from the translation of common names in scientic publications. This is a
particular problem when translating from English to Spanish, given the long-
standing English amateur tradition of studying fauna. Maribel Tercedor-Snchez
addresses the problem of terminological variation in Oncology due to wide public
interest in cancer and the dierent social situations in which the terminology is
used. She has built up a corpus of comparable Spanish and English oncological
texts and has classied them according to communicative situations, to facilitate
the selection of the appropriate terminology for each professional context.
In conclusion, we hope Investigating Translation will contribute to a better
understanding of what is going on in translation research today. Translation
scholars, particularly young researchers, need a clear idea of the dierent aspects
of translation that are the objects of investigation, the dierent motivations
behind the research and the dierent paradigms used. This should help us to see
which paradigms are complementary, clarify terminology and develop more
objective research methodologies.

Allison Beeby, Doris Ensinger and Marisa Presas

<TARGET "p1"
</TARGET "p1">DOCINFO

AUTHOR ""

TITLE "Investigating Translation Paradigms"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

S I

Investigating Translation Paradigms

<TARGET "nid" DOCINFO

AUTHOR "Eugene A. Nida"

TITLE "A Fresh Look at Translating"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 1

A Fresh Look at Translating

Eugene A. Nida

The poor rating of translating, in comparison with such other skills as music,
graphic art, and dance, has always puzzled me. If, as I. A. Richards declared
translation is perhaps the most complex event in the history of the cosmos,
(1953) why should translating not have greater prestige and certainly be better
rewarded nancially. But people seem to take for granted that any bilingual
person can interpret or translate, and accordingly, interlingual communication
must not be such a dicult task. Unfortunately, many persons do not realize that
most translations are not only poor but at times even misleading. In fact some
translations into English by professional translators have as many as fteen
expressions per page that are not normal English, either because the translator had
not fully understood the source text or was the victim of a translationese virus.
Translating and interpreting are innate skills because almost any multilingual
can hear or read a statement in one language and respond in another language
with an expression that is at least somewhat equivalent in meaning. Even pre-
school children often serve as eective interpreters for recently arrived immi-
grant parents, and they seem to accomplish the task of interpreting or translating
without any particular frustrations. It all seems so natural, until such children go
to school and begin learning about nouns, verbs, adjectives, etc., at which time
their natural skills seem to be side-tracked by dependency on various grammati-
cal classes of words.
Some translatologists would like to regard translating as a science, but an
important distinction must be made between the specic act of translating and
the study of translating as an important interlingual and intercultural phenome-
non, best referred to as Translations Studies. There is no widely accepted
theory of translation because we simply do not know, and perhaps never will
know, what goes on in the black box of our brains. In a sense we are like
alchemists before the discovery of dierent atomic weights and valences of
diverse chemicals made possible the science of chemistry or like physicists
4 EUGENE A. NIDA

before the discovery of electrons, positrons, neutrons, and quarks, or geneticists

before the discovery of the double helix of the DNA.
But perhaps more surprising than anything else about translation studies is
the fact that so many of the most skillful translators have little or no use for the
theories of translation. They generally regard such theories as a waste of time,
and Translation Review, the leading journal on literary translating, does not
accept articles on translation theory, but they do publish magnicent interviews
with outstanding translators who explain in detail how they deal with the
complex issues of multilingual equivalence.
Even if expert translators have little or no interest in translation theories,
one would expect that students of translation would certainly favor such courses.
But in visiting more than thirty schools of translating in various parts of the
world, I have found that almost always students nd the courses on the theory
of translating to be some of the most boring and useless aspects of their training.
Perhaps the real problem is that most theories of translation are not based on
actual practice, and students sit through too many lecture hours before they have
enough experience to see the relevance of the various theories. Why not begin
with three or four lectures on translation studies and then toward the end of a
three or four year programme provide a comprehensive picture of the various
principles of interlingual communication, based on precisely what the most
competent translators actually do?
Many students of translating can be helped to understand what they are
doing by being exposed to the research of Damasio and Damasio (1992) about
the three major networks of the brain and the manner in which all images and
concepts must be represented by intricate sets of synapses. It may also be useful
to learn about the important concepts of Miller, Galantar, and Pribram (1960) in
highlighting the role of values associated with all experience, including language
texts and usage. Human beings cannot be likened to computers, with the brain
matching the hardware and the mind being the software. What is distinctive
about human beings are the values associated with every experience.
I must confess that I have accomplished very little in assisting translators by
means of lectures on principles of translating, because so few people have been
able to recognize the direct application of the principles, despite the abundant
illustrative examples of similar diculties. I have, however, accomplished much
more by sitting with translators, going over texts word for word, inquiring of
local people about the meaning of sentences, and helping translators discover
how to fully understand the meaning of a text. On one occasion I spent several
days with a translator going over a relatively short passage in one of Saint Pauls
letters in the New Testament. Because of certain presuppositions about divine
 A FRESH LOOK AT TRANSLATING 5

inspiration of the Greek text, the translator simply could not bring himself to
translate meaningfully. At last, he declared, If I only understood exactly how
the words t together, I could translate! And that proved to be precisely the
problem; he knew the meanings of the individual words but not how they
combined to make sense. He had not paid enough attention to the context of this
letter nor to the manner in which the author normally constructed meaningful
utterances, nor what other authors had written about similar matters. Translators
must go beyond dictionaries and grammars to investigate the contexts of related
discourses and of the corresponding cultures.
Because of the numerous examples of translation problems encountered by
translators of the Hebrew Old Testament or the Greek New Testament, some
people have assumed that my ideas about translating developed as a result of
extensive contacts with Bible translators in more than 200 languages. But my
principles of translating were formed long before I began to help Bible transla-
tors. When I did write books and articles to help the more than 5,000 people
world wide engaged in Bible translating, I quite naturally used illustrations drawn
from such experience. But these were not the basis of my understanding of how
and why translations should represent the closest natural equivalence.
My basic ideas about principles of translating developed during my studies
at the University of California at Los Angeles, where as a Greek major I was
fascinated by the literary character of the Greek Classics, especially the timeless
relevance of Euripides, the magnicent style of Thucydides, the shifts from
hilarious humor to exquisite dialogue of Aristophanes, and the incredible
sensitivity in Sapphos poems. Our teachers would never permit a literal
translation of a text because such a rendering would indirectly but inevitably
depreciate the quality of the Greek text and at the same time would encourage a
poor use of ones own language. A detailed comparative study of dierent
translations of the Classics soon revealed that in many respects these translations
were superior to many translations of the Bible which are too often inuenced by
theological presuppositions.
As the result of a course in historical and descriptive linguistics in my
senior year and the reading of Bloomelds Language and Malinowskis Coral
Gardens, I could see how closely language is bound to culture. Sapirs analyses
of Southern Paiute also make language a universal phenomenon, which was
further enlarged by studies in cultural anthropology at the University of Michi-
gan. I began to see how language is layered, not only in structures of words,
grammar, and discourse, but in levels of relevant contexts that provide the
framework for understanding texts. Language became as much a part of culture,
as spirit is a part of body. This means that ultimately people understand a text
6 EUGENE A. NIDA

only on the basis of the relevant contexts, a fact integrally related to all types of
verbal communication.
In one language of East Africa in which certain tonal distinctions had been
missed by a translator in speaking and writing the language, what was supposed
to mean this cup of blessing do we bless actually meant this cup of poison,
do we bless. Under other circumstances most people would have immediately
recognized that this was merely a mistake in pronunciation and orthography, but
the local people never recognized such a mistake, because they had a practice of
drinking a poison cup as an ordeal to prove their innocence. Accordingly, the
people assumed that the Christians each Sunday drank a bit of miraculously
transformed poison in order to show that they were innocent. These indigenous
people were doing what everyone does in trying to make sense of a strange
statement: they immediately try to determine the meaning in terms of their own
culture.
In other instances the local culture can provide a fascinating context for a
correct understanding and appreciation of a biblical text. For example, in
explaining to some West Africans the creation story in Genesis, Chapters 1 and
2, the translator told the people that the name Adam actually means earth or
ground and that the name Eve means life, to which one insightful person
exclaimed, Then this means that the story is not just about two people, but
about everyone of us. This alert African had immediately associated the biblical
account with the mythic context of his own culture. Fortunately, he did not have the
literalist verbal background that is typical of so many people in the Western World.
Since all verbal communication is based on the two fundamental processes
of selection and arrangement and since the original author has already selected
certain basic concepts and has arranged these in a text, presumably a translator
only needs to duplicate what the author has already done. But translating is not
that simple. In the rst place, the ideas represented in the source text are a
reection of a particular culture, and the intended audience of a translation may
not share such concepts nor place the same value on the corresponding ideas.
Furthermore, languages do not dier in what they can say, but in how they say
it. What is linguistically obligatory in one language may not be so in another.
For example, singular versus plural number is obligatory in Indo-European
languages, but not in Chinese. In English all sentences must have a formal
subject, even if it is no more than a functional zero, e.g. there, as in there is a
worm in that apple or an anticipatory it in it is a shame for him to feel defeated, but
in Chinese a subject that is clearly evident from the total context is simply omitted.
Because language is linear, arrangements must also be linear, but they also
need to express dierent levels of relevance. That is precisely why all languages
 A FRESH LOOK AT TRANSLATING 7

show dependency relations in written texts by means of such features as conjunc-

tions, participial qualiers, quotation marks, parentheses, indentations and
subtitles, while in spoken texts speakers often employ louder volume to highlight
certain features, rapid utterance to indicate less importance, and stage whispers
to suggest a high degree of relevance.
Because the author of a text has both selected and arranged the units of a
language, to the framework of dierent types of discourses, this does not mean
that a translator does not also have to be concerned with processes of selection
and arrangement. A translator rst needs to discover what were the cultural
contexts that evidently prompted the author to write as he did in order to
communicate eectively to an audience that he evidently intended to inuence.
But a translator must bear in mind the dierent audience and the diverse
features of the language-culture into which he is translating.
The presumed audience of a translation is not limited to those who will buy
a published translation. For the translator a more critical and crucial audience
includes those who pay for the translation and those who judge it at various
stages, including the responsible person in the agency that contracts for the
translating and one or more editors of the agency or of the organization paying
for the translation process. If these persons are not pleased with the results,
further assignments will soon die up.
A translator must engage in thousands of decisions involving both selection
and arrangement to t another culture, a dierent language, diverse editors and
publishers, and nally a reading audience.
If the source and receptor cultures are quite similar, the easiest texts are
those that are composed primarily of highly specic vocabulary, e.g. novels and
short stories, but if the two cultures are very dierent, these same kinds of text
are often the most dicult, e.g. from English into Chinese. If, however, the
content of a text is largely ideational, e.g. philosophical or juridical, even within
a very similar culture there may be vast dierences of vocabulary and rhetorical
devices, e.g. the translation of philosophical texts from German into English. But
diversities of language are never as great as the dierences of culture. For
example, translations from Hungarian into English (two languages from dierent
language families) are relatively simple in comparison with translations from
English into Hindi, in which religion, philosophy, and cultural values dier
greatly, despite the fact that both languages are Indo-European.
But a translator has a further factor that often inuences his or her work,
namely, the degree to which a translator can identify with the evident value
system of the original author. Professional translators are often called upon to
translate texts that are contrary to the translators set of values and word view.
8 EUGENE A. NIDA

The capacity to do justice to something which seems so culturally abhorrent

demands a degree of objectivity beyond the power of many translators. For an
interpreter the emotional strain of representing what one knows is entirely wrong
can be even more dicult, especially since the oral form of communication
seems to be so much more personal and self incriminating.
The role of context is immediately evident in understanding the meaning of
words. In fact, the English word turn has no meaning apart from some context,
but in contrast with the listing of 110 dierent meanings of turn (in addition to
32 idioms) in The Random House Dictionary of the English Language, second
edition, unabridged, there are actually only a relatively limited number of
dierent types of contexts, and these contexts are precisely what combine with
turn to produce the variety of meanings.
In the past lexicographers assumed that each word had an inherent number
of dierent meanings and the contexts simply pointed to the right meaning, but
it is more realistic to regard the combinations of words and contexts as providing
the basis for the intended concept. This means moving up from the atomic level
of analyzing words as isolatable units to the molecular level of words in context,
which is a much more meaningful level of analysis and of comprehension.
The verb turn combines with certain contexts to indicate a movement of an
entity on an axis, e.g.
(1) He turned to see the boat.
(2) He turned around.
(3) They turned toward her.
(4) The actor turned a somersault.
The verb turn may also combine in a causative relation with a number of objects
that turn completely or partially on an axis, e.g.
(5) He turned the screw.
(6) She turned the key.
(7) The servant turned the handle of the door.
(8) They turned the wheel rapidly for ten minutes.
But turn may also occur in contexts indicating movement of some entity with
respect to some hypothetical axis, e.g.
(9) The driver turned the corner.
(10) They turned around the block.
(11) He turned right
(12) He turned up the hill, meaning He turned and went up the hill.
 A FRESH LOOK AT TRANSLATING 9

Turn may also occur in contexts expressing a change of state, e.g.

(13) They turned Republican.
(14) She turned Catholic.
(15) The stock market turned bearish.
(16) The poor man turned pale.
(17) The business turned protable.
(18) He turned eighteen.
In (18) turn becomes equivalent to an attributive verb in an equational relation,
e.g. become.
As a noun, turn may combine with various contexts to refer to an alternat-
ing period of activity, e.g.
(19) They took turns in playing tennis.
(20) It was his turn to speak.
There are also a limited number of highly specic combinations of turn and a
context, e.g.
(21) He turned the garment, meaning to turn something inside out.
(22) He turned the knife blade, i.e. he dulled the blade.
(23) The leaves turned, referring to a change in color.
(24) He turns wood, referring to the occupation of shaping wood on a lathe.
In trying to determine the meanings of words in context, it is important to look
rst to the syntagmatic contexts, those contexts by which most people learn at
least 99 percent of the meanings of words in their mother tongue. But it is also
useful to examine the paradigmatic contexts that can highlight dierences of
meaning by placing dierent words in the same syntagmatic contexts, e.g.
(25) He turned the handle.
(26) He twisted the handle.
It is also useful to study the meanings of words in a closely related semantic
eld or domain, e.g. turn, twist, rotate, revolve, spin. For this purpose monolin-
gual dictionaries are usually superior to bilingual or multilingual dictionaries. But
when the meaning of a term seems very elusive, an encyclopedic article may be
indispensable. With so many encyclopedias in CD ROMs a translator often has
an incredible amount of data available in his or her personal computer.
For understanding a text the next level of contexts is the grammar, but
unfortunately most traditional grammars are based almost exclusively on formal
10 EUGENE A. NIDA

and distributional classes and not upon referential classes. This means that such
expressions as
(27) jumped the fence vs. broke the fence
(28) climbed the mountain vs. blew up the mountain
are considered to represent the same semantic relations. But in jumped the fence,
the so-called object marks the location of the jumping not the entity aected by
the jumping. The same type of distinction applies to the set climbed the mountain
and blew up the mountain. Compare also the following set:
(29) he followed the road vs. he wrecked the road.
Most grammars list be and become as attributive verbs in so-called equational
sentences, but other verbs serve the same purpose, e.g.
(30) They felt sick.
(31) They turned pale.
(32) They ended up tired.
(33) They seemed listless.
(34) They appeared dejected.
Some languages, however, have additional obligatory categories, e.g. degree of
reliability of information. Several indigenous languages of South America have
at least three degrees of certainty indicated by enclitics. The rst degree marks
information known personally by the speaker. The second degree indicates
information obtained from reliable sources, and a third degree takes no responsi-
bility for the information. In other words, the information is legendary or mythic.
The potential for building grammatical constructions is so enormous that at
times bureaucrats become more interested in their way of speaking or writing
than they do in producing understandable texts. In large international organiza-
tions the multiplicity of languages and the traditions of juridical texts tend to
produce grammatical absurdities in many texts that defeat the very purpose of
eective communication. The following sentence is not an unusual one in
documents of the Commission of the European Union:
(35) The Diario Ocial de las Comunidades Europeas (DOCE) of 13.1.98
publishes a notice (98/C 804) of initiation of an investigation of
imports of canned tuna originating in Ecuador with a view to
reintroducing the Common Customs Tari duty applicable to that
product, following the Commissions receipt from Spain of a com-
plaint pursuant to Article 14(1) of Council Regulation (EC) No
 A FRESH LOOK AT TRANSLATING 11

1256/96 concerning canned tuna falling within CN codes 1604 14

11, 1604 14 18, 1604 14 90, 1604 19 39, and 1604 20 70 of the
Combined Nomenclature originating in Ecuador and put into free
circulation within the Community with preferential benet under the
Communitys generalized system of tari preferences.
This one-hundred-plus word monstrosity violates not only a number of rules of
English grammar but is a serious barrier to eective communication. This is,
however, precisely what happens when specialists are only concerned with
talking among themselves, something that applies also to linguists, anthropolo-
gists, and translatologists.
Like every aspect of language the signicance of a discourse or part of a
discourse depends on the context of the culture. It is, however, a serious mistake
to think that one can neatly separate the semantic from the pragmatic elements.
There is only one continuum with varying degrees of semantic and pragmatic
factors throughout. But the most relevant manner for discussing such issues
requires that a distinction be made between (1) the relation of discourse types to
cultures and (2) the cultural dierences that may require marginal notes or
introductory comments.
Dierent cultures have distinctly dierent ideas about what certain
discourses should be like. In China both short stories and novels generally have
sad endings, while Americans usually prefer happy endings. In some cases,
however, publishers have tried to improve the sales of translations from Chinese
by altering the endings of novels so as to please an English-speaking audience.
Even a translation of the Bible in which great attention was paid to the
poetic qualities of the Hebrew texts has been published exclusively in prose
form. One translator and publisher explained his use of a prose format as trying
to assure readers that the texts were true, rather than merely derived from the
poetic imagination of ancient prophets.
Even within certain cultures there are important changes taking place in
discourse structures. Milan Kunderas novel Immortality (1990) represents a
radical change from traditional plot-dominated stories. This novel is based more
on dialogue and monologue, while also being intensely personal and philosophi-
cal. As a result, it is largely nonsensical to a society accustomed to gang warfare,
fast car chases, and sexual intrigue.
What the future will produce no one can tell, but with increasing attention
to voice texts, automatic transcribers, and a ood of internet communication,
aesthetic literary creativity will probably be appreciated by an increasingly
restricted public.
</TARGET "nid">

12 EUGENE A. NIDA

In many cases, however, the diversities of cultural practices and the

underlying presuppositions are so diverse that only notes, whether on the same
page or in an appendix, can bridge the cultural gaps. For example, the concept
of trial through ordeals by being forced to thrust ones hand into boiling water,
to touch ones tongue to hot iron, or to drink water mixed with ashes from a
burnt sacrice is so strange to modern readers that it usually requires some form
of explanation, whether inserted into the text or in footnotes. The old idea that
footnotes are a sign of failure on the part of a translator is nonsense.
The concept of limited good is even more dicult for the Western World
to comprehend because most people have the idea that there is no limit to riches
or fame. In many societies, however, any person who seems to have far more
than his or her share of goods, reputation, and strength must have acquired such
vital force by robbing others of their share. For most societies there is not only
a ceiling, but it is not too far above the oor.
If there is any one crucial procedure for revealing the meaning of words,
sentences, and texts, it is the indispensable role of combined linguistic and
cultural contexts.
<TARGET "neu1" DOCINFO

AUTHOR "Albrecht Neubert"

TITLE "Theory and Practice of Translation Studies Revisited"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 2

Theory and Practice of

Translation Studies Revisited
25 years of Translator Training in Europe

Albrecht Neubert

Translation studies as applied linguistics

A quarter of a century ago I was instrumental in putting translation studies on

the map of applied linguistics. And there was a concrete occasion to do just this.
I was asked to deliver the rst ever plenary paper on the subject at a world
congress of AILA, the Association Internationale de Linguistique Appliqu which
took place in Copenhagen (Neubert 1973). I did not do anything, but I reflected
what was an agenda, what, in fact, had to be done. And it was by no means
accidental that in several other countries in the world translation successfully
aspired to become a respectable academic discipline.
The reason AILA were recruiting a German professor was that in my
country, both in its Eastern and Western political realms, translation studies was
a little in advance with respect to the academic status of translation. In fact, I
had the privilege at that time not at all thought to be one by my mostly
philological colleagues of being oered the directorship of the Leipzig
Dolmetscherinstitut in 1962. This school, originally founded as a private school
in Leipzig in the twenties, had been incorporated into the university just ve
years earlier. There were similar developments in the Federal Republic of
Germany, where training centres, which were highly regarded as practical
schools for translators and interpreters, were trying hard to get rid of their stigma
of being looked upon as non-academic Berlitz-school-type establishments.
How dicult and often frustrating their struggle for acceptance by their
academic peers could be can be learnt from an experience I myself had on the
occasion of a rare visit to Heidelberg University. I think it was in 1963 or 1964.
Our colleagues at the School for Interpreters and Translators, which was
nominally part of Heidelberg University, had invited my colleague Otto Kade
14 ALBRECHT NEUBERT

and myself to participate in a colloquium on translation studies. It turned out that

we shared most of our cherished concepts of how to combine theory and practice
in our curricula at Heidelberg and at Leipzig. But during a reception organised
by the Philosophical Faculty, which considered itself as a kind of guardian of the
newcomers from the lowlands of translation practice, the Dean, a professor of
Romance philology, who happened to know me from his Leipzig years, secretly
took me aside. He remembered my work as a philologist in the department of
English and American Studies of Leipzig University in the late fties.
We had the same teachers who thought little about translation, not to speak
of translation studies. He advised me, earnestly, not to take the academic
aspirations of the translation colleagues in his faculty too seriously, hinting that
they were not real scholars, but just language practitioners. They were doing
their job alright, but, honestly, their work was a far cry from traditional academic
standards, which, of course, were the norm for French or English studies, and the
research carried out by his professors in the philosophical faculty. Translation
graduates, he exhorted me, would therefore not be allowed to take a Ph.D. course
without doing, at least, an extra two to four semesters of philology, in particular
the heavy load of the linguistic and literary history of their respective languages.
These subjects had to be taught by proper scholars, not by the language experts or
polyglots, however excellent, which constituted the majority of the translation sta.
Such was the atmosphere of the sixties. In fact, the misgivings vented by
our philological peers lingered on for a long time. This episode is not just an
anecdote. I rather believe that my story is symptomatic of a much more impor-
tant aspect of this early stage of translation studies, at least in Europe. Of course
there were local circumstances and the personal idiosyncrasies that accompanied
the specic situations which prevailed in dierent countries and at dierent
universities at the times when translation studies made their rst tentative
academic steps. But there was a general pattern, which was pretty much the same
all over Europe. It resulted from the reaction of the translation students and
scholars against the discouraging tutelage of their new academic hosts. In
consequence, many of us were convinced that the only way to gain academic
standing was to look for a frame of reference that would facilitate our acceptance
in the prevailing scholarly setting.
What could help us to step on the academic ladder and climb up, however
slowly, was by studying the medium of our subject, namely language. Transla-
tion entered higher education as a new eld of language studies. This is why the
first scholarly paradigm of translation studies was borrowed from linguistics, at
least what came to be practised as linguistics among European academics in the
sixties and early seventies. In this sense Catfords timely book A Linguistic
Theory of Translation (1965), which appeared in the mid-sixties was symptomat-
ic. But I think that another book which appeared in the Netherlands at the same
 THEORY AND PRACTICE OF TRANSLATION STUDIES REVISITED 15

time was just as, if not even more inuential. I am, of course referring to Eugene
Nidas, seminal Towards a Science of Translation (1964). Though written from
the specic perspective of the practice of Bible translation, it must be looked
upon as the work of a well-known American linguist turned translation
student.
Linguistics was hailed as a science pilote that seemed to promise ample
rewards for all those students and scholars who had decided to make translation
their main academic occupation. Their professed goal to describe, explain and,
last but not least, to teach translation, that is, the task of rendering a product of
a source language into one of the target language, had to be formulated in terms
of linguistics. And this corpus of ideas about language was equally thought to be
the main source of knowledge that was to inuence the various components of
the students curriculum. Thus the institutionalisation of translation studies
occurred under the auspices of the study of languages, specically, as this
academic eld was understood in many philology departments. This latter
ingredient included, quite signicantly, much of literature plus culture and
civilisation of the respective countries, the languages of which happened to be
the working languages of the translation students. But everything was dominated
by how this was expressed by language.
More aptly, the linguistics of translation always referred to a language pair.
This is why I focused my Copenhagen paper on the correspondences between
source and target languages. The underlying idea, supposed to condition and
shape whatever translators would produce as a result of their endeavours, was the
replacement of L1 linguistic units and structures by L2 units and structures.
Individual renderings at sentence level, which came to be regarded as tertium
comparationis, were understood as being enabled by the resources of the target
language system in relation to the source language system.
In eect, a diagram of the then dominant linguistic model of translation
could simply be as follows (actually this was precisely the diagram I used in my
Copenhagen lecture!): L1 gets restructured by L2. For every L1 word or meaning
and every L1 grammatical, i.e. morphological and syntactical pattern, there are
one or several potential L2 correspondences. These correspondences make sure
that L1 is completely recovered by L2. The source system becomes projected
onto the target system, which acts as the L1s match. Or the linguistic means of
the L2, the translators tool kit, reect those of the L1. For instance, L2 tenses
step in for L1 temporal distinctions, L2 words rephrase L1 lexical originals. As
can be expected, the resulting system of correspondences exhibits very few one-
to-one relationships. One-to-two, -three, or one-to-many correspondences abound,
only to be disambiguated when the words and structures are translated on the
sentence level. Part of these linguistic juxtapositions were, of course, also one-to-
zero relations, which, though more intriguing, could mostly be treated with the
16 ALBRECHT NEUBERT

help of paraphrasing L2 means. The more elements of the two language systems
were included in the overall network of correspondences the more comprehensive
was the repertoire of building blocks accessible to the prospective translator.
However, the upshot of all this was the realisation that knowledge of
language was clearly not enough for successful translation. Students were asked
to develop a keen awareness of the matching potentials of the working languag-
es. Exercises might include methods of how to convert English progressive
tenses into languages where other constructions were supplied by the system. Or,
lexical elds characteristic of the German educational system had to be reordered
in terms of French, Spanish, or Russian when particular words were translated
into their respective sentence contexts.
Obviously, all considerations guiding the study and practice of translation in
the European universities tted perfectly into the contrastive brand of linguistics
well established as a new and ourishing branch of applied linguistics. Thus,
for translation studies to become a legitimate member of the academic family in
the arts sector seemed to be a foregone conclusion. Translation, by the mid-
seventies, had achieved the status of a linguistic paradigm, more precisely, a
contrastive linguistic paradigm.
And this is where the founding of European translation departments in the
academe found its higher justication. It looked as if translation studies lled
an existing slot. That the training of translators was a practical necessity for
society was one thing, but to install it in an academic setting was another. And
translation scholars had more or less persuaded their philological colleagues that
contrastive language studies, which had previously also transformed foreign
language teaching, supplied the necessary scholarly credentials.

A new paradigm translation textology

The historical picture I have drawn so far is not complete. Convincing as it may
seem that translation students were able to make academic headway as contras-
tive linguists, the reality of the translation process had always proved much more
complex. As translators had always known and translation teachers had to
admit to their students in class an adequate target version of an original was
the result of system-linguistic recoding only to a certain extent. Often enough
target sentences which were neatly recoverable in terms of the correspondence
relations of contrastive linguistics turned out to be quite correct but not entirely
adequate renderings of their target originals. Evidently, the information as to
how to hit upon a more satisfying translation, which is both correct and
adequate is provided by insights from a corpus of knowledge that goes beyond
the translators language qua systemic competence.
 THEORY AND PRACTICE OF TRANSLATION STUDIES REVISITED 17

Grammatically and/or lexically correct translations often sound strange to L2

users, or at least translation readers. In the terminology of the information
engineer, they produce noise that obstructs the smooth passage of information
in the communication channel. In a nutshell, the awkwardness or the inappropri-
ateness of the translation, linguistically proper as it may be, is not a feature of
the individual sentences. It is a quality, and for that matter, a negative quality of
the text. It clashes with the textual expectations shared by target language users
as text users, specically by those readers who had expected the translation to
come up to the textual standards that are normally associated with L2 texts of
this particular kind.
For renderings into L2 to conform to L2 textuality translators knowledge
of language has to be supplemented by knowledge of texts. This means that
translators must be experts in both L1 and L2 discourse. In addition to what they
know about their respective working languages as such they need to have
recourse to the two textworlds which the speakers of L1 and L2 are used to and
are habitually conversant in. And it is these genres and text-types that signi-
cantly condition the words and structures of the two language systems. No doubt
translation studies have to incorporate this textuality, in order to be a more
realistic reection of what goes on when we translate. Consequently, this has to
be taught to the translation student, who has to learn why and how an adequate
translation deviates from an otherwise expected rendering that takes only into
account considerations that arise from the systems of the source and target
languages.
In other words, the textual paradigm of translation does not disregard
linguistics but supplements it by what may be called a science of texts. In
analogy to paradigm changes in the history of science, where Thomas Kuhn rst
introduces the term in his famous The Structure of Scientic Revolutions (1970)
inadequate translations may be regarded as anomalous texts. It is true, they pass
the lter of systemic linguistics, that is, they cannot be explained by a linguistic
paradigm of translation. They call for elucidation by a new approach. The
anomaly disappears if textual norms are brought into play. Thus the textual
paradigm followed the linguistic one.
Of course, there were forerunners to the textual approach, such as what I
called the pragmatics of translation practised, for instance, in our Leipzig school
as early as the mid-sixties (Neubert 1968). But they had not acquired paradigm
status, though they prepared the stage for future changes.
One could claim and in fact the claim was made (Hartmann 1980;
Hansen 1983) and is still being made (Malmkjaer 1997) that what has
actually happened is the extension of applied linguistics to include textual
contrasts, actually the systematics of textual contrasts. But when we look at what
was going on outside translation studies it becomes quite evident that the
18 ALBRECHT NEUBERT

decisive input came from the study of texts and not from contrastive linguistics,
where the topic of the text was not thematised until the mid-eighties or nineties.
By contrast, a fairly independent discipline was rapidly developing in the
seventies, whether it was usually called text linguistics, discourse analysis (van
Dijk 1985) or, more rarely, text science (Beaugrande 1980, 1981). Generally this
was not to the liking of the true-blooded linguists with structuralist and, soon
enough, Chomskyian leanings, who viewed this extension of the linguistic eld
of study to the text as impure, risky or, at least highly premature. Premature
and perhaps even misguided, was the tenor of the reproaches that were levelled
against the textlinguists in the light of the fact that so much about the subtleties
of the inner sentence had not yet been properly described and rigorously
explained. There is, I think, quite a lot to be said in favour of the implacable
attitude of the dyed-in-the-wool linguists when they regarded the inclusion of the
text into the realm of language studies with extreme suspicion. They feared, and
I think rightly, that the properties of discourse required much more than linguis-
tic expertise. There is more between sentence and text than is dreamed of in a
linguists philosophy.
This is why I wonder whether the methodology of linguistics as linguistics
is able to cope with this enormous expansion of scope. What was involved
turned out to be much more than the stu of language because texts are held
together by realms of knowledge that spread over many elds, in fact, text
structures are patterned by world knowledge. Linguistic aspects of the text
become overlaid by encyclopedic domains.
The textual or rather the text-scientic paradigm involves an explanatory
frame that is inseparable from the mindset of the expert who produces the text.
Translators, in exchange, must try to emulate real-life communication not only
between people with everyday or general interest, whose cultures may share
textual features, but more importantly and, of course, more intriguingly, between
specialists, whose knowledge and cultural worlds may be wide apart.
It is a well-known fact that a dicult L1 text can tax the L1 readers
immensely if they are not the normal L1 addressee. Imagine then the enormous
extra problems encountered by translators if they believe they can approach such
texts primarily as a linguistic construct. Instead, source texts should be, and
normally are, read and analysed as complex semantic loads coded by unique
discourse signals. These markers allow the allocation of a text to a particular text
class. At the same time they ensure that the text is interiorised as a unique
exemplar of its text class. And text classes are constituted by more than formali-
ties. They presuppose the competence of knowledgeable text users. Text mean-
ings are intricately interwoven with text shapes, both on a macro- and a micro-
textual level.
Think of the textual intricacies, such as the isotopic chains linking the words
 THEORY AND PRACTICE OF TRANSLATION STUDIES REVISITED 19

and, above all, the technical terms of a textual superstructure (van Dijk 1980), or
of the characteristic sequences of morpho-syntactical conventions distinguishing
the discourse of expert communications. These have to be grasped and properly
identied as what they are to contribute to the global meaning to be translated.
Furthermore, the construction of the target text must be based on an intimate
knowledge of the text-building principles of the L2 culture. All this is part and
parcel of the overall message that has to be put across.
The textual paradigm of translation has as its background the textworlds of
the two communicative cultures involved. These cultures branch into vast
networks. They exhibit textual domains characteristic of all or more often
groups of L1 and L2 users. In particular, it is the vast framework of L1 and L2
parallel texts that lends itself as material support. By this term we mean texts
that are not directly related to translations. They are the normal products of L1
and L2 discourse, outside the realm of translation. What makes them parallel are
the identical or similar communicative conditions under which they have been
produced time and again. This is why they share signicant textual markers
although, and this is the point that makes them so informative and helpful for
translators, they may and often do exhibit striking lexical, syntactic, and stylistic
dierences in their respective L1 and their L2 realisations. Translators can take
parallel texts as their model if they are intent on producing target versions that
can be accepted by L2 recipients as genuine constituents of their familiar native
textworld. As an aside, Id like to point out that the adjustment of translations to
L2 parallel texts does not only apply to linguistic restructurings of various kind,
lexical, grammatical and textual. It also encompasses more substantial alter-
ations such as, for instance, additions and cuts as well as shifts, which are
deemed recommendable or even necessary in the L2 genre corresponding to its
L1 counterpart.
At this point I hasten to add that the enculturation of translations into the L2
communicative culture is by no means the only way to apply the textual model
of translation. A particular translation assignment may call for quite the opposite
procedure, viz. the target version should openly betray L1 textual features
because not familiarisation but alienation is expressly intended, maybe by the
commissioner of the translation or by the circumstances of the concrete trans-
lational situation. I think examples can easily be supplied.
What is essential in this context is the obvious fact that this method of
bringing the L2 reader closer to the L1 textworld is not a deviation from the
textual paradigm. It is rather a consequence because this approach oers a rich
diversity of textual options to the discourse-sensitive translator.
The way originals are approached is a function of textual considerations
couched in the realities of the translation market. The impact of discourse is
manifold. Generic provisions, aecting the classes of text, as well as holistic
20 ALBRECHT NEUBERT

concerns, arising out of the needs of any particular text, yield a multifaceted
repertoire of guidelines for prospective translators (Neubert 1996: 9194). As a
matter of fact, the integration of all translation activities into the prevalent
discourse habits of the receptor community, or, to paraphrase Even-Zohar (1978),
into the L2 communicative polysystem, present a most varied pattern with regard
to the relationship of translations to their translated source texts and, more
signicantly, the multiplicity of L1 discourse situations. This relationship is
anything but single-minded. It has nothing to do with such utterly misleading
notions as a translation must be such and such. On the contrary, thinking in
term of textboundness implies there can never exist such a thing as the
translation. Translations always vary over time and place.

The rise of an interdisciplinary paradigm

(a) The translators knowledge prole

There is no doubt that the textual paradigm of translation has transformed the
teaching of translation from the mid-seventies to the eighties and onwards. It can
be said that with the overcoming of system-linguistic constraints the discourse
orientation has also strengthened the need of relying on the actual experiences of
the professional translators, past and present. The textual paradigm has opened
the way towards the empirical reality of translation and translators. Arm-chair
views gradually gave way to complex perspectives of what is actually going on
not just when we translate but how translators have been doing their job
through history (Delisle and Woodworth 1995). Scholars could no longer
proceed from their wonted academic position deducing the features of the
translation process from an abstract model, be it couched in linguistic or textual
terms. Teachers, as well, could no longer remain within the bounds of their
favourite textual choices, which were but a replica of their own restricted
experience, if this experience was realistic at all. Often enough it had been just
an extension of their experience in foreign language teaching into translation
teaching. By contrast, the translation classroom was being slowly transformed
into a workshop, practising, if only in simulation, how professionals develop
their translation competence by marshalling a vast array of knowledge compo-
nents and transfer skills.
This broadening of the scope of teaching as well as the underlying study
and research aimed, eventually, at including the whole gamut of translation
activities in society. This could not mean, of course, a reproduction of the total
scope of what was going on under the name of translation. But no doubt the
spectrum of the corpus of texts to be incorporated into translation studies and
teaching set out to achieve prototypicality. This meant that the extent of the
 THEORY AND PRACTICE OF TRANSLATION STUDIES REVISITED 21

texts should be representative of the country, in fact, the materials to be put up

for translation exercises and for scholarly investigation were likely to be the best
examples of the professional scenario. To give just one example, viz., the long
cherished, admittedly textual distinction between general and technical
translation did not really include specialist texts, which, by denition, could only
be fully understood by specialists in the eld. The new approach, which so far
has perhaps not been realised by the majority of teachers, is going much further.
It takes the term technical seriously, including highly sophisticated originals,
which require in-depth pre-translational analysis, an undertaking that would
normally bet a subject expert. And the target versions, in order to be accepted
as adequate, should strike the addressee as making expert sense in the eld.
I have in mind, for instance, legal documents or papers in scientic journals,
technical reports or political treaties, and any other kind of specialist discourse.
They do not only call for expert formulations but genuinely expert understanding
of the subject matter. There are enough surveys available in many countries,
which give detailed data as to the kind of specialised texts that are prototypical
of a certain period. They display the country-specic reality awaiting the
translation graduates (Schmitt 1990).
In turn, translation studies are now trying, more or less directly, to achieve
a fully-edged model of this translational reality. Scholars and teachers alike are
making serious attempts to integrate the professional expertise of translators and
interpreters into their institutional projects and programmes. They take into
cognizance how practitioners outside the classroom master their daily routine,
how they go about their highly specialist work preparations. And, perhaps most
importantly, translation students are taught how to concern themselves with the
many-faceted knowledge profile, which represents the foundation of translational
skills, and they are acquainted with a diversity of knowledge components.
The best way to achieve this is by introducing students to a set of structured
modules. These are cognitive objects comprising, e.g., terms and denitions,
categories and relations, products and processes, illustrations and diagrams, in
short, proles of a knowledge eld or subject area, grouped in such a way that
they reect the typical procedures of experienced translators, their strategies as
to how to access and digest new topics. They bring into the classroom what
out there in situ gets continually integrated into the practitioners cognitive set
as a result of their endless endeavours to cope with ever new translational themes
and challenges. Translation studies have actually set themselves the aim to live
up to what translation practice is all about.
Of course, it would be futile for academic institutions to try to copy all
translation situations. They cannot but demonstrate the integration of textual
prociency and expert knowledge as a matter of principle, both in research and
in the curriculum. But whatever selection they may decide upon, the professional
22 ALBRECHT NEUBERT

competence of the subject expert has to be included in the form of prototypical

cases. Case studies, carefully chosen from the general practice in a country or in
an international organisation, acquaint the prospective translator with the subject
study and analysis that must be brought to bear on the translation process.
Students cannot possibly be made expertly familiar with all the elds of
knowledge that might be on a countrys translation agenda. Here again, less than
half a dozen protoype areas are chosen, which are then dealt with in detail. The
aim can only be to train prospective translators in the light of these cases, how
to go about any other translation that they may happen to have to cope with as
professionals.
However, it was soon discovered that this brave attempt at modelling
professional expertise was like opening Pandoras box. It meant taking into
account everything practitioners needed to know in order to rise to a multiplicity
of occasions, in response to what their commissioners and clients wanted in ever
new variations. If we add to this the translators own professional ethos to
produce quality whenever possible, their responsibility as experts, we end up
with a new, extremely comprehensive agenda. Consequently, translation students,
scholars, teachers and learners had to recognise that acting out the textual
paradigm was not nearly enough. In fact, discourse options by themselves
presupposed accessing entirely new items of information. Evidently, in order to
do their job properly, translators had to know how communication could be
successful. Starting with the analysis of the source text, its formal and its
semantic-pragmatic properties, in short, with text comprehension they have to
research the corresponding material, structural, cognitive and cultural conditions
prevailing among the intended addressees in the target community. Based on this
knowledge (which is largely declarative knowledge) they have to bring to bear
their repertoire of practical resources to eect an adequate transfer (which is
primarily procedural knowledge).
In the attempt to assess and bring order into this heterogeneous complex of
knowledge components, which make up the experienced translators competence,
almost inadvertently translation studies have become involved in many elds of
knowledge, actually in all the other disciplines that are aected by the content
or subject matter of the texts to be handled. As a result, the seemingly self-
sucient concepts of translation studies, originally linguistic and later textual,
were extended to incorporate, or at least, adjust to what are non-translation-
oriented elds. Out of this ongoing amalgamation grew a new paradigm.
Translation studies are being transformed into an interdiscipline. Accordingly,
the rst congress of the newly-founded European Society for Translation Studies
held in Vienna in September 1992 was about Translation Studies: An Inter-
discipline (Snell-Hornby et al. 1994).
This new interdisciplinary paradigm has, in fact, only conrmed what had
 THEORY AND PRACTICE OF TRANSLATION STUDIES REVISITED 23

tacitly been taken for granted, namely, that in a time of growing specialisation
when everyone was proud to admit that they knew more and more about less
and less, translators by contrast were expected to know everything about more
and more. With the spectacular penetration of translation activities into practical-
ly all areas of the global village, quantitative as well as qualitative demands on
translators and interpreters were raised dramatically. Though in the public eye the
translators craft is still largely misconstrued as that of a mere language
mediator, the normal pattern of expectation, both from the angle of the sender
and from that of the addressee, is by no means that restricted. Everyone expects
to be served with the real thing, that is, translators are always supposed to rise
to the occasion of oering their clients nothing but the true meanings of the
message across the language bar. By this demand, translation users insist that
translators be equipped with all the knowledge ingredients necessary to safeguard
smooth communication in politics or industry, in science and technology, in the
arts as well as in the world of sport, in short, in everyday matters as well as in
specialist contexts. Often enough the expert in a eld protests or, at least is taken
aback when a translator misses a point allegedly simple enough for a novice in
the specialist eld. We all have experienced cases like this.

(b) Dierent perspectives on translation

What I have so far pointed out as supporting the idea of an interdisciplinary
paradigm is, I am afraid, not the whole story. Technical translators have always
felt the need for a high degree of specialist knowledge. Experienced translators
are now saying that often more than seventy or even eighty percent of the time
required for such translations is taken up by researching the subject matter. Thus
the disciplines that are often translated certainly become part and parcel of the
interdisciplinary paradigm.
But the thematisation of translation studies from the angle of the subject
matter is just one process that threatens to jeopardise the supposed unity of
translation studies. Without being able to give an exhaustive overview of the
other disciplines that have been drawn to the translation scholars attention, let
me just mention disciplines such as psychology, economics, sociology, history,
criticism, cultural studies, and computer science. In a recent book (Neubert and
Shreve 1992) I have myself tried to show how translation has been progressively
modelled from the point of view of more than half a dozen dierent disciplines,
among them linguistics and text science as early conceptualisations. What is most
important about these models is they are complementary. They do not exclude
each other, though when one particular point of view is pursued, for instance, in
the now quite fashionable psychological model, the other approaches are
disregarded or outright neglected. Studying translation as it happens in the minds
of people when they translate must leave out many features such as quality,
24 ALBRECHT NEUBERT

subject matter, experience, language, textuality, etc. They are placed in the
background, whereas the forefront is given over exclusively to the mental
processes going on after reading the original and before formulating the target
version.
Of course it is quite another matter, how these disregarded features are
brought in at a later stage when insights of the mental model are used to shed
light on inquiries leading into other directions. It is certainly rewarding not only
to study the mental strategies adopted by professionals compared with beginners
(Krings 1988), but to apply the striking results in teaching novices.
Similarly the modelling of translation from the angle of comparative culture
studies, valuable as it is in its own right, aords elucidation and, in fact,
explanation of an approach that usually has translation criticism as its main
agenda.
Furthermore, the implication of the translation market, that is, the economics
of the profession, has very clear implications for the practical model of the
translation classroom. Translators are not only expected to produce perfect
translations t for the printer. Clients often demand a whole battery of working
and gist translations, to be completed in very dierent time spans.
Last but not least, there is a revolution under way as a result of the use of
electronic means to enhance the work of translators. I am not primarily thinking
in terms of automatic translation, but rather about the much more important and
promising perspective of computer-aided translation and the signicant way it
changes and expands the horizon of the craft. This new dimension is already
changing translation teaching, turning the classroom into computerised work
stations. It goes without saying that this transformation has an important impact
on translation studies.

A new synthesis?

What are the implications of this unending story of dierent perspectives?

Have translation studies not rather come to a halt in view of this fatal embarras
de richesse that appears to have befallen their eld? How can we cope with the
implications of the interdisciplinary paradigm without losing a sense of coher-
ence? Are these many facets of interdisciplinary development not rather a sign
of disintegration? Have we nally lost our overall view? This is the crucial
question that we have to ask ourselves after twenty-ve years of progress in
translation studies. Certainly, it has been a success story. Just think of the almost
fabulous increase in publication in and about the eld. Publishing houses in
several countries are competing for new books on translation and interpreting.
The trickle of writing in the fties and sixties has widened into a massive stream
 THEORY AND PRACTICE OF TRANSLATION STUDIES REVISITED 25

with new authors and new topics entering the scene almost every month.
To conclude, it was no accident that I have not once used the term theory.
I do believe that the rise of the term translation studies has been positive. It is
true, the term translation theory appears now and again. But, I think, this usage
is usually meant to be a synonym for translation studies, depending upon the
author more than upon the language.
But I dont want to quarrel about words. If I bring up the theory issue now
it is for a more serious reason. Theory for me is an ambitious term and should
be used with discretion, at least in its dominant sense current in Britain. It is well
known, especially in American English, that theory can also be taken to mean a
particular conception or view of something to be done or of the method of
doing it, or even just a guess or conjecture (The Random House Dictionary of
the English Language, 2nd ed. 1987). Theory of translation, however, should
aspire to be a coherent group of general propositions used as principles of
explanation for a class of phenomena (ibid.).
It is my rm conviction that, in spite of their remarkable ndings in many
areas, so far translation studies have not achieved theoretical status. Maybe the
time is ripe, or perhaps just riper, to string together the various strands of our
knowledge about translations and translation processes. To be more specic, I
think the interdisciplinary paradigm has not only brought to light the pressing
need to look at the complexities of our eld from various angles. It has at the
same time cleared the ground that had been notoriously obscured by a host of
theorists who claimed to describe the translation as such from their favourite
vantage point, discouraging the views of others as o the point. In actual fact,
what we have been facing in the last quarter of a century has been an often
embarrassing and even irritating talking at cross-purposes.
Perhaps a new synthesis could spring up from the interdisciplinary para-
digm. Perhaps it would lead to a genuine theory. What might be feasible is the
development of a number of partial theories or subtheories, by systematically
amalgamating individual models or perspectives of translation. This process
could then be improved or speeded up by combining two or more mini-theories
to achieve more comprehensive conceptualisations, which would not only cover
more ground but would yield more overarching explanations by including more
classes of the phenomenon. In my view a genuine or comprehensive theory of
translation would be an exercise in dialectics. On the one hand it could be built
up by any number of specific theories concentrating on and explaining the
categories arrived at by looking at translation from dierent angles. This
development is certainly under way. On the other hand, and to some extent as a
result of integrating these partial views we would have to inquire into by which
factors can be brought together to yield a general theory of translation. Are there
any binding forces at work that can be recognised as holistic patterns? To
</TARGET "neu1">

26 ALBRECHT NEUBERT

deserve this name they would occur whenever we speak of translation or

interpreting. They have the explanatory capacity which the minitheories could not
achieve, partly because of their limited scope, their restricted perspectives.
A theory, by denition, is an interrelated concern. I have myself tried to
delineate some of the postulates of a theory of translation (Neubert 1997). But,
and I want to make this point quite clear, the past twenty-ve years have not led
us to a general theory of translation. Nevertheless, these have been rich years as
regards our specic knowledge about the various aspects of what constitutes
translation, including how to teach it in the university. Maybe the next twenty-
ve years will eventually bring translation studies closer to a theoretical status
and the scientic rigour a general theory demands. I think we can and perhaps
must make this point because the rising role translation and interpreting are
playing in the world will demand that our research and teaching be based on
ever more stable scholarly and practical foundations. As we have seen often
enough, theory without practice is empty, but, equally, practice without theory
is blind. The world as global village in an age of ever faster information
highways just cannot aord translators who are inferior to all the other workers
who manage our survival.
<TARGET "zin" DOCINFO

AUTHOR "Zinaida Lvvskaya"

TITLE "The Scope of a Communicative Theory of Translation"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 3

The Scope of a Communicative Theory

of Translation
An attempt at systematisation

Zinaida Lvvskaya
Universidad de Las Palmas de Gran Canaria

Our dierences of opinion are not due to the fact that some opinions are more
reasonable than others, but rather that our thoughts run along dierent lines
and we do not contemplate the same things.
(Ren Descartes. Discourse on the Method)

Introduction

The structure of scientic knowledge implies a system of dierent elements

which are logically linked together. This systemic nature of science is one of the
basic premises for the development of any of its branches. The scope of a theory
depends not only upon its capacity to describe, explain and predict the function-
ing of the object under study (the activity of the translator, in our case) but also
upon whether the scientic knowledge which forms the basis of the theory
actually constitutes a system.
In our opinion, the analysis of the scope of a communicative theory of
translation and the systematisation of scientic knowledge are now a priority.
There are at least two reasons for this. First of all, we need to unite the eorts
of dierent translation theoreticians who, indeed, seem to have more in common
than not. Moreover, if they dier, it is usually because they do not contemplate
the same things. For example, Skopos theory supporters consider that any
bilingual activity, regardless of its aim, can be tted within one sole framework,
whilst many other scholars centre exclusively on communicatively equivalent
bilingual activity. Secondly, although many recent publications have oered
valuable practical observations, they have not always made any great contribution
28 ZINAIDA LVVSKAYA

to translation theory, precisely because the research was not based on a scientic
hypothesis that could be conrmed, amended or rejected by empirical data.
The communicative approach to translation (where translation is seen as an
intercultural verbal activity) has thrown open a whole new realm of possibilities
for the Science of Translation. However, at the same time, the approach requires
the systematisation of dierent aspects of scientic knowledge, so that they can
be tested to see whether they really constitute a system and can be applied in
practice.
Here, we shall analyse the scope of a communicative theory of translation
(Lvvskaya 1997) using the scientic methodology and logic used by those
sciences which study objects which cannot be observed directly. The aim of this
analysis is to show that the scientic knowledge on which our communicative
theory of translation is based, does constitute a system. To that end, we shall use
the inductive-deductive method and scientic reection. We shall also indicate
the various stages in the development of our hypothesis.

Methodological premises

Theory about an object that cannot be observed directly does not use data which
can be analysed by the senses, or premises which are in themselves evident, or
axioms. On the contrary, this kind of theory has to cope with the absence of
completely reliable sources of scientic knowledge. This is not only true for the
Science of Translation, but also for other sciences which study objects that
cannot be observed directly or could not be observed during a certain period in
the development of this scientic knowledge. In such cases, scientic hypotheses
are advanced, based on the data provided by other sciences and scientic
reection.
Scientic methodology advances hand in hand with scientic knowledge
itself. For example, Descartes considered the only valid source of scientic
knowledge to be those premises which were beyond all reasonable doubt (what
we would now call common sense). Spinoza maintained that science could
deduce other truths beyond those supplied directly by the force of reason. Now,
these reasonable ideas seem to have lost ground with the development of
experimental sciences.
Science has made enormous progress and has discovered new truths.
Therefore, when attempting to develop a theory, it is important to bear in mind
the history of science, not only the history of one particular science, but also
other related sciences. In our case, these are all those sciences which study
human behaviour (psychology, the general theory of communication, sociology,
pragmatics, the theory of action and others).
 THE SCOPE OF A COMMUNICATIVE THEORY OF TRANSLATION 29

Each science requires the development and the support of the rest and,
therefore, no one science can be considered fundamental or basic. Even philoso-
phy, which has always been considered to be the science draws upon other
elds. Thus, the interdisciplinary nature of the Science of Translation, which is
often stressed by translation theorists, is not an exclusive or distinctive character-
istic. Furthermore, all sciences start with questions, not certainties. Certainty
itself is always relative; it is an end, not a beginning.
The rst task faced by all research is to decide what we really want to
investigate. Once this has been done, we look for the factors of the problem,
that is, those factors which determine how the object of research functions. The
analysis of these factors may open up possible ways of solving the problem. In
the process of evaluating the various suggestions and suppositions, we may nd
new factors which are then incorporated into our hypothesis. This is the way
scientic hypotheses are developed, as the result of a chain of events. A system
may be discovered, if the original a priori is valid and the logic of the research
has not been infringed upon.
Any hypothesis should be applicable. That is, it should be possible to test
it in practice to see whether it can solve a given problem: in other words, that
it leads towards the prescribed goal. During the testing of a scientic hypothesis,
new suppositions or reasons will appear, which will either conrm or invalidate
the whole hypothesis or a part of it. The validity of a hypothesis can also be
conrmed if it is possible to deduce one part of the whole from another. It was
Kant who said that scientic knowledge can only exist when the human brain
orders and organises data.

Stages in the development of the hypothesis

Therefore, research begins with uncertainty and takes as given, or a priori, the
premise considered most adequate to solve a problem. Thus, the linguistic theory
of translation worked upon the supposition that the process of translation was a
linguistic operation. During the practical process of testing, this hypothesis was
proved to be mistaken. It could not explain the behaviour of the translator when
faced with no linguistic equivalences in the target language, or the impossibility
of always using linguistic equivalences in the target text even when such exist,
or the need to expand, or reduce, the target text. In short, it could not explain all
the partial or total changes in the linguistic (semantic) structure of the target text,
in comparison with the original (9\&@&F8"b 1985: 674).
In the communicative theory of translation, we start with a dierent
hypothesis: translation is deemed to be an intercultural communicative activity.
The activity of translating is like any other activity as dened by action theory:
30 ZINAIDA LVVSKAYA

it has a structure which includes its aims, the means to achieve them, the
dierent stages of the process, sub-phases, and the results of the process. The
aim of the activity of the translator will depend on how translation is dened.
We dene translation as equivalent intercultural communication, the product of
which can be dened as a communicatively equivalent representation of the
original text in another cultural milieu (Lvvskaya 1992: 313).
By comparing the structures of equivalent and heterovalent intercultural
activities (the latter being that which does not pursue communicative equivalence
between the ST and the TT) and noting their dierences, we have reached the
conclusion that dierent types of bilingual activity exist. This, in turn, has led
us to the following conclusion: dierent activities cannot be explained by the
same theory, given that they are determined by dierent factors. Each activity
requires its own scientic explanation, that is, its own theory.
Thus, in our opinion, the key problem of translation is the communicative
equivalence of two texts. The principal factor in this problem is how sense is
made and interpreted. Advances in the study of pragmatics, semantics and
information theory have suggested the importance of extralinguistic factors in
making and interpreting sense. However, these studies have not been sucient
to describe and dene the nature of the activity of the translator and the factors
which determine it. Pragmatics was right in linking the sense of a text into its
authors proposed intention. However, there is still a certain lack of denition of
the concept of sense and its relationship with a texts content/meaning. Obvious-
ly, these problems are directly related to equivalence in translation.
This presents us with a new problem to solve (9\&@&F8"b 1985: 76143;
Lvvskaya 1997: 342). Research into the sense of a text has allowed us to
discover other factors which inuence translating. The sense of a text is a
subjective and extralinguistic category, the product of the purpose of a given
individuals communicative activity in a given situation. The meaning (of a word,
a phrase or a series of phrases logically linked together) is a linguistic category
and, therefore, is objective for all the individuals who share the same culture.
The relationship which exists between the linguistic meaning and the sense of a
text is similar to that which exists between the philosophical categories of form
and content. If, within a cultural poly-system, the same sense can be expressed
with dierent meanings, all the more reason for this to be true when it is
transferred from one culture to another.
The sense of a text has a hierarchical structure that is made up of three
components: the situational component (the communicative situation), the
pragmatic component (the authors intentional-functional programme in the text)
and the linguistic component (the content of the text, which is the result of the
interaction between the lexical and the grammatical meanings). The communica-
tive situation dominates the hierarchy since it participates both in shaping the
 THE SCOPE OF A COMMUNICATIVE THEORY OF TRANSLATION 31

sense of the text and its interpretation by the receiver.1 The pragmatic component
is subordinate to the situational component, and the linguistic component is
subordinate to the other two in the hierarchy of sense, since the author of the text
chooses the way of expressing his ideas according to his intentional-functional
programme and the given communicative situation.
Since no two individuals share the same knowledge, experiences or values,
the interpretation of a texts sense by a receiver will always be somewhat
dierent from the original sense intended by the author of the text. No message
is ever perfectly understood. However, the more knowledge author and receiver
share, the easier communication will be.
Research into the structure of sense has allowed us to deduce some charac-
teristics of the real nature of the activity of the translator and equivalence. These
deductions seem so clear that they could almost be regarded as axioms:
(1) Given the importance of extralinguistic knowledge (encyclopaedic, situational
and background) and its primacy over the linguistic component of sense,
translation, like any other verbal communication, involves intellectual activity.
(2) Given the inter-subjective nature of any verbal communication and the
relative nature of the receptors comprehension of the sense of the text (in
translation there are two receptors: the translator and the person for whom the
TT is designed), communicative equivalence will always be relative. (3) Given
the intercultural nature of bilingual communication, the communicative situation
always changes when the translator produces the TT. There is a basic contradic-
tion in trying to produce a TT which will preserve the pragmatic component of
the ST in a new communicative situation (in another cultural environment), given
the possible lack of coincidence between the knowledge, experience, values and,
naturally, the verbal and non-verbal behavioural norms of the two cultures (the
factor of cultural intertextuality). Clearly, the only possible way to solve this
contradiction is to modify the linguistic component of the TT with respect to the
ST in order to t the new communicative situation.
These deductions allow us to look more closely at the factors which
determine the activity of the translator and, thus, to formulate new suggestions.
All the preceding arguments lead us to conclude that these factors are situated in
two closely related spheres: cognitive and cultural. Cultural-cognitive competence
includes knowing how to use linguistic resources correctly within certain
conceptual frameworks and in accordance with the norms of verbal and non-
verbal behaviour. Without this competence, the translator cannot understand the
sense of the ST, de-objectivise it, nor produce the TT, objectivise it.
If we accept the mechanism described above for creating and interpreting
sense, the cognitive-cultural nature of the factors that determine translating, and
the conclusions reached by other authors (3.&,>8@ 1992) as the basis for our
research, we can use deductive logic and scientic reection to suggest, albeit
32 ZINAIDA LVVSKAYA

sketchily, what these factors are. They include the following: (1) The translators
presuppositionary, background and encyclopaedic knowledge, together with his
ideas about what is known by the other two communicants, seen as representa-
tives of their respective cultures. (2) The translators axiological knowledge and
his ideas about the axiological knowledge of the other two communicants, seen
as representatives of their respective cultures. (3) The translators knowledge
about the norms of verbal and non-verbal behaviour in both cultures, including
textual conventions. (4) The translators knowledge about other aspects of the
author of the ST and the receiver of the TT (seen as individuals) which may be
relevant (place of birth, life history, historical period, social condition) as well as
other circumstances relevant to both communicative situations (where, when and
in what circumstances the ST was produced and the TT is being produced).
(5) The translators ideas, reached by de-objectivising the ST, about the
authors own method of achieving the goals of his verbal activity (his idiolect).
(6) The translators ideas, resulting from all the above knowledge, about the
diculties the new receiver may have in interpreting the TT.
As translation is a poly-determined activity, some factors may conict with
others. Logically, the translator resolves this contradiction in favour of the factors
which are most important in any given situation (Lvvskaya 1997: 73).
In addition to the cognitive-cultural factors there are at least two more
evident factors. The rst is the translators professional experience, which
requires no further comment here. The second additional factor is working
conditions and it will be considered in relation to the dierent branches of the
science of translation (specic theories).
These suggestions about the factors which determine the options open to a
practitioner of equivalent intercultural activity are necessarily hypothetical and do
not pretend to be complete, even though they have been tested in practice in
translation classes and in several doctoral theses. A scientic hypothesis should
be able to indicate which lines of research should be followed, even though each
step of the research may suer modication or the hypothesis itself change. This
has been my experience. In a book about theoretical translation problems, written
fourteen years ago (9\&@&F8"b 1985), we claimed that translation competence
included expertise in the so-called translation techniques. Later, we rectied
this conceptual error and we are pleased to know this new vision of the problem
is shared by other researchers (Elena 1994).
The hypothetical nature of a theory does not mean certain deductions cannot
be made, so long as they do not contradict the remaining suppositions, that is,
they are logically sound. This method has allowed us to reach a further deduction
about the nature of translating. This is related to the normative nature of
translating (Lvvskaya 1997: 5860) and clearly contradicts the widely held
opinion that the Science of Translation is not normative. We believe this opinion
 THE SCOPE OF A COMMUNICATIVE THEORY OF TRANSLATION 33

is based on a conceptual error, since the norms of communicative activity were

thought to be linguistic. Logically, the norms of any activity should be of the
same nature as the activity itself. Obviously, the norms of translating should not
contradict the general norms of verbal activity, formulated in the theory of
Speech Acts (conversational maxims), while maintaining the specic characteris-
tics of any intercultural communication.
After all, these norms constitute the operational principle of translation, that
is, maximum faithfulness to the intentional-functional programme of the author
of the ST, combined with TT acceptability in the target culture. We propose the
following norms. (1) A translation should not contradict the intentional-functional
programme of the author of the ST, i.e., it should not lead to false implicatures
which are not in the ST. (2) A translation should not contradict the relevant
factors of the communicative situation created in the target culture, i.e., it should
be appropriate for the new situation. (3) A translation should not contradict the
norms of verbal and non-verbal behaviour in the target culture, in general, and
the textual conventions, in particular (unless the violation of the norms are part
of the intention of the author of the ST and, even when this is the case, the
violation will be appropriate for the target culture). (4) A translation must
capture all the implicatures (with or without explicit markers) that are relevant to
the intentional-functional programme of the author of the ST and necessary for
its comprehension by the receiver of the TT. (5) A translation must capture the
communicative and stylistic peculiarities of the ST authors idiolect regardless of
the kind of linguistic resource used to that end.

The branches of the Science of Translation

A communicative theory of translation also permits a logical approach to the

problem of branches of the science. All the above factors are general in nature.
Their relevance may change from one situation to another, but, they are always
valid and they provide the basis for a general theory of translation. However, one
of these factors, working conditions, can be used to develop specic theories
of translation.
There are at least two aspects to this factor. The rst is translation modality
(written translation, consecutive, simultaneous and liaison interpreting). The
general factors apply equally in all of these dierent modalities. However, other
factors appear which inuence the translation option, for example, the time
factor in simultaneous interpreting. This requires that the TT be reduced with
respect to the ST, thereby reducing the natural redundancy characteristic of any
text and omitting secondary information. Another example is the memory
factor in consecutive interpreting, which requires note-taking skills. Yet another
34 ZINAIDA LVVSKAYA

is the constant linguistic-cultural code switching in liaison interpreting.

Specic theories are needed to study these additional factors produced in
function of the translation modality. In some cases, work on these theories has
been limited and in others, there has been no commitment to any one general theory
of translation and, therefore, they lack scientic value. Fortunately, this is not the
case with simultaneous interpreting (Seleskovitch 1978, 1985; Q,D>&@ 1987).
The type of text to be translated also conditions specic factors which,
together with the general factors, determine the translators strategies and
options. Obviously, it is not the same to translate a literary text as it is to
translate a scientic or administrative text, a lm script or the instructions for
some domestic appliance. The appearance of new factors in function of the text
type implies the need for additional competences and these should be studied.
In our opinion, this is more than sucient reason for developing specic
theories to be applied to literary translation (poetry is another matter), scientic
and technical texts, lm scripts and so on. These theories should correspond to
the functional styles and genres of the texts. We already have valuable studies on
literary translation (Lev 1967; Popovic 1973) and they are increasingly commit-
ted to a communicative theory of translation (eg. '@>R"D,>8@ 1988; Garca
Lpez 1996; Pascua 1996; et al.).
However, we do not agree with the development of partial theories to deal
with specic problems, such as equivalence, metaphor or proper names (Holmes
1988), as we do not consider this to be methodologically sound. It is clear that
equivalence (if, in reality, we are dealing with communicative equivalence) is the
basic problem in general translation theory, and the modality of the translation or
the type of text to be translated does not alter this. Today, most authors agree
that communicative equivalence is what denes translation (Rabadn 1991) and
that this implies the double operational principle mentioned above (maximum
loyalty to the ST authors conceptual programme combined with the TTs
acceptability in the target culture). This concept is valid for all modalities of
translation and text types. Nevertheless, the existence of additional factors which
characterise each of the specic theories, may entail changes in the linguistic
structure of the TT when compared with the ST (Lvvskaya 1997: 4360).
The translation of metaphors, humour or proper names cannot be studied out
of the context of the modality of a specic translation or a specic text type. It
is not the same to translate a metaphor, or humour, in a literary text as it is in a
scientic text, in a written translation or as it is in simultaneous interpreting.
Therefore, these kinds of problem should be studied within the framework of
their respective theories.
Neither do we consider it to be methodologically correct to develop partial
theories around pairs of languages, because the problem of translation does not
depend upon language or language combinations, since the factors which
 THE SCOPE OF A COMMUNICATIVE THEORY OF TRANSLATION 35

determine the activity of the translator are cognitive and cultural and, therefore,
the same for any combination of languages.

Teaching Translation

It is clear that specic teaching methods cannot be developed without connection

to the theory of the given activity since the teaching methods depend upon the
characteristics of the activity which is to be carried out. Therefore, no translation
teacher can aord to leave out aspects of such importance to the communicative
theory of translation as the structure of the activity of the translator, in all of its
phases and sub-phases, the contents of which depend upon general and specic
factors of the text, and which directly corresponds with the competence of the
translator. (Lvvskaya 1992; 1997: 107126). Since translation is an intellectual
activity, the teaching of the same is designed at inculcating skills rather than
habits, a method which implies searching for the relevant factors to be taken
into account in the translation of any specic text which will allow for a solution
to be found to the dialectic contradiction which exists between the two requisites
to be fullled by a TT in order to be communicatively equivalent to an OT.
We should here underline the dierence between the activity of a profes-
sional translator and the activity of teaching translation. The rst process is
internalised and the greater the professional, the more dicult it will be to
subdivide his activity into phases and sub-phases. In teaching, the situation is
radically dierent. To teach method, albeit in the rst stage, we must external-
ise as far as possible the successive phases and sub-phases of the process. For
that reason, precisely, we subdivide the two phases of translation (the interpreta-
tion of the OT and the production of the TT) into the following sub-phases:
The phase of interpretation of the OT: the sub-phase of comprehension of
the semantic contents of the text; the sub-phase of comprehension of the
intentional-functional programme of the author; the sub-phase of compre-
hension of the meaning to be received (with forecasts as to the diculties of
a cognitive-cultural nature that such may present for the receiver of the TT).
The phase of production of the TT: the sub-phase of development of a
strategy for the translation in function of the type of text and idiolect of the
author; the search for functional equivalences; the production of the TT;
self-correction of the TT.
Although teaching of a foreign language for translation purposes is not a branch
of the Science of Translation, it is of enormous importance in vocational training
for translators whenever it is couched within the culture of the language and
</TARGET "zin">

36 ZINAIDA LVVSKAYA

against the backdrop of comparative functional stylistics. The ideal aim of the
teaching of a language, be it the mother tongue or a foreign language, is to train
the student to understand and produce texts of dierent styles and functional
genres by applying the verbal and non-verbal norms of behaviour and, more
specically, the textual conventions of each culture. Unfortunately, this aim is
still very much of an ideal and not precisely because of any fault on the part of
the language teachers. Comparative functional stylistics still has a long road to
travel from the perspective of research despite the enormous practical importance
of the same, since translation of conventional texts represents almost one hundred
per cent of all the translations on call in the labour market. We cannot aord to
waste more time in this eld and, precisely for that reason, many of the research
projects which are currently underway in the various faculties of Translation and
Interpreting have to do with the problem of comparative functional stylistics.

Conclusion

The systematisation of scientic knowledge is of enormous importance to the

development of the Science of Translation. However, this task can only be
carried out if it is based on a scientic hypothesis that implies a system of
interlinking knowledge, the scope of which is sucient to explain the process
of translation in all of its phases and sub-phases. We have tried to show, albeit
partially, and by way of approximation, how knowledge relating to the activity
of translation forms a system when framed within a communicative theory of
translation.
No science is exclusively theoretical or practical. However, it is theory
which guides us in our research and allows us to systematise our knowledge. In
any case, the scope of a theory can be determined from its possible applications.

Translated by Margaret Hart

Notes

1. The motivations and intentions of the author of the text are shaped as a result of the interaction
of individual characteristics relating to his individual mental framework and factors pertaining
to the communicative situation in itself. The person receiving the text, likewise, manages to
make sense of the communication by tallying the linguistic / semantic meaning of the text with
his own knowledge and with the factors which are relevant to the communicative situation in itself.
<TARGET "god" DOCINFO

AUTHOR "Pilar Godayol Nogu"

TITLE "Living on the Border"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 4

Living on the Border

Feminine Subjectivity in Translation

Pilar Godayol Nogu

Universitat de Vic

A borderland is a vague and undetermined place created by the emotional

residue of an unnatural boundary. It is a constant state of transition. The
prohibited and forbidden are its inhabitants. Los Atravesados live here: the
squint-eyed, the perverse, the queer, the troublesome, the mongrel, the mulatto,
the half-breed, the half-dead; in short, those who cross over, pass over, or go
through the connes of the normal. (Anzalda 1987: 3)
The border experience can happen whenever and wherever two or more
cultures meet peacefully or violently. (Fusco 1995: 169)
I think of my own people, the only people I can think of as my own are
transitionals, liminals, border-dwellers, world-travellers, beings in the middle
of either/or. (Lugones 1994: 469)

Commenting on these three epigraphs, which reect on the problems deriving

from cultural transmission, is perhaps the best way to introduce the concept-
metaphor of Living on the Border. These reections mark the beginning of a
journey which will lead us on to dialogue and discussion about the borderland
experience, that liminal existence which characterises the work of all feminine
subjects in translation, and which, as we proceed to study the matter, will spark
moments of reection and self-criticism.
Boundaries, says Gloria Anzalda in the rst of the epigraphs, have been
established throughout history with the objective of distancing those who live
inside from those who live outside, separating us from them. Those who create
boundaries maintain their integrity by dierentiating themselves and their space
from that which lies beyond their limits, at the same time expelling all signs of
strangeness from their territory. For the person who dwells on the inside,
everything that is from outside is strange and foreign. The exclusion of these
distorting elements and their banishment to alien lands is then the way of
38 PILAR GODAYOL NOGU

maintaining the power of insiders and legitimising its expression. In precisely

this way, dominant cultural discourses have constructed and sustained the
marginality of feminine intellectuals down through the course of history.
However, the borderlands may also be seen not as simple limits to an inner
space, but rather as spaces in their own right, indeterminate areas in which
emotional residues are in a state of constant transition. This does not mean,
though, that our people are those who reject the same reference points as we do
that would be to fall into the same trap as the dominant cultural expressions.
Rather, it is a question of accepting ourselves, and accepting others in all their
complexity, living in a state of permanent change and transfer. Modifying,
changing, (re)creating ourselves, travelling through many voices, territories and
identities is, then, the only option for those who dwell within these imaginative
and creative spaces, neither on the inside nor on the outside.
Despite this, we live in a world where the borderlands are not only abstract
and intellectual but also physical. To be precise, Coco Fusco, in the second of
the epigraphs, distinguishes two forms of contact between cultures. On the one
hand, there are apparently peaceful expressions, the response of intellectual
activity to ideological developments or to cultural reconciliations; on the other,
there are violent forms, imposed by either colonial expansion or by social
situations characterised by oppression and submission. We can see that all
borderland experience, whether it owes its origin to violent attitudes or not,
implies leakage and contamination between cultures. Within this space the
multiplicity of identities, the breakdown of dichotomies, the openness of readings
and the uidity of circulation of textual processes are celebrated. Only here can
one understand the necessity for all subjects to learn to be mobile, plural, erratic,
polyglot and resistant if they do not want to perpetuate the practices of power
developed by dominant groups throughout history.
Similarly, for Maria Lugones, the author of the third of the epigraphs which
open this text, only those who reside in borderland states, and who therefore feel
obliged to build bridges of communication between the variant positions they
occupy, live in open and heterogeneous structures. Contrary to those who see
borderlands as arid, unsafe and uninhabited areas, Lugones claims that it is only
in these borderland situations that the best cultural interchanges are produced. To
live on the border is, then, neither threatening nor indesirable. However, as
Anzalda points out in the rst epigraph, only los atravesados (the daring) dwell
in the borderlands. Only those who do not defend their own integrity by barring
the way to those from outside, who have no fear of occupying indeterminate
positions and who celebrate the opportunity to travel, evolve, mutate, metamor-
phose, adopt new disguises and abandon xed identities, only those people live
with the knowledge that they are neither inside nor outside.
The positions of Anzalda, Fusco and Lugones, with whom we have begun
 LIVING ON THE BORDER 39

this textual journey, are based on the conviction that it is possible to construct a
cultural object from the borderlands, from a space in which both subject and
object sense themselves to be in contact and in mutual contamination. In large
part, this is also the conviction of the person who translates as/like a woman.
What, however, does translating as/like a woman mean? Does it imply a non-
stable subject who is at home in all those borderland states found in life? Is it
possible to construct a politics of identity that presents an absolute, immobile,
silent subject? Above all, how can a translation survive if it does not question its
subject?
The theory and practice of translating as/like a woman, being a political and
social discourse that criticises and subverts the patriarchal practices which render
women invisible, assumes a feminine subjectivity. That is, it makes plain that the
common basis of its activity is a subject who, in the words of Milagros Rivera
(1994: 62), vive en un cuerpo sexuado en femenino (lives in a feminine body).
However, despite sharing a common politics of identity, the dierent feminisms,
among them those in the eld of translation, interpret and express feminine
subjectivity, also known as feminine identication (Fuss 1995), in dierent
ways. Similarly, they also dier in their denitions of their universal categories,
such as women, identity, gender, sex, experience and history. As a
result, some translators, like Lori Chamberlain (1988) and Amy Kaminsky
(1993), suggest that these are unstable starting points for developing either a
theory or a practice of translation. For this reason, they cast doubt on the
possibility of building a feminist theory of translation given the contingency and
mobility of its universal categories.
This raises an urgent question: is it feasible to believe in a politics of
identity which, in its denition of itself, claims that its categories are premises
which are neither problematic nor questionable? How can a politics of identity
survive if it does not take into account the idea that its universal categories must
be permanently open and questioning in order to lay the ground for the inclu-
sions or exclusions of its future demands? Thus, if we wish to move closer to the
unresolved question of the feminine subject in the practice of translation as/like
a woman, as well as in all elds of general feminist study, we must rst inquire
into the theoretical problems and the practical limitations of the concept itself.
The rst problem challenging any attempt to establish the concept feminine
subject translator is the essentialism into which the intersection gender/
translation falls. Were a stable subject to become the starting point for a social
and political theory, such as the practice of translating as/like a woman, this
would imply the irrevocable closing of the debate regarding the construction of
the subject itself, and, therefore, the immobilisation of the subject/object relation-
ship. The desire to elaborate a more denitive theory embracing a more deni-
tive subject in order to facilitate the reconstruction of textual complexity in
40 PILAR GODAYOL NOGU

translation is easily understood. However, there are several advantages in reading

the feminine subject as a permanently contingent and contentious territory, one
which shuns the immobility of fundamentalist formulations and categorical
universalities. Only in this way can we avoid falling into a dangerous homogeni-
sation and simplication of the hermeneutic processes of translation. For this
reason, I believe that the fact of questioning and problematising the feminine
subject in the practice of translating as/like a woman means accepting that
contingency can never be eliminated in the interweaving of gender and transla-
tion. Consequently, no translation can ever share the same feminine subjectivity
as another since this would involve the stabilisation of its meaning. Translating
as/like a woman implies, then, proposing a non-stable and contingent representa-
tion of the feminine subject, recognising, at the same time, that one cannot write
about a subject without essentialising it (Scott 1992: 33). Nevertheless, it is easy
to imagine the diculties such an approach entails. How, for example, can we
reconcile the representation of heterogeneous subjects with the universality of
their social contract? How can we lay claim to a general politics of identity
without eliminating dierence?
Singularity and plurality in feminist study practices are terms which are in
constant conict (Cixous 1986; Kaminsky 1993; Scott 1992, 1996; Suleri 1992).
It is important, then, that the universal category women, being both a particular
and shared representation, be kept completely open. Even so, one of the most
important problems faced by feminisms is the fact that the category women
seems to have to describe the base of its social and cultural identity. This has
given rise to debates among the advocates of essentialist and constructionist
postures (Fuss 1989, 1995). On one hand, the essentialists maintain that there is
a feminine essence which lies outside social and cultural limits. The construction-
ists, on the other hand, insist that the feminine identity is a historical construct
and that therefore the category women is the result of complex discursive
processes.
This paradox regarding the status of woman, which has for many years been
the cause of ideological conicts among feminisms, cannot be made non-
problematic. Rather, we should embrace it as a concept which involves constant
transfer, and in this way avoid the polarisation of feminist study into antagonistic
camps. Only by accepting that the category women represents the uid base of
its practices can translation as/like a woman free the concept from the binary
notions which currently frame all discussion. In this way, the category women
may be used in future discourses, including those of translators, as one which is
open to revision and (re)signication. Only if, in contrast to the static meanings
of hegemonic discourses, the idea frees itself from formulation in terms of
dichotomies where, mistakenly, binary oppositions are simply mutually exclusive
and never mutually communicative, will we be able to speak about and reect on
 LIVING ON THE BORDER 41

the (im)possibilities of identifying the feminine subject in translation.

If the feminine subject translator eludes denition due to its mobility, we
may also add that believing in a stable reference point in order to create a theory
of translation which is non-problematic amounts to paralysing the (re)signicat-
ion of the concept itself, throwing up a hurdle to dialogue between the subject
and object of translation. Further, the feminine subject translator can never be
fully represented, fact which brings to the fore the multifaceted character of its
existence. The true representation of the feminine subject translator is, then, the space
which is dierence. That is, the denition which is at once included in and excluded
from the representation itself the space which permits (re)consideration, (re)-
modelling and (re)formulation of the previous congurations of meanings.
We should remember that the practice of translation as/like a woman and
here lies the heart of the controversy between feminist translation theorists
must be based on a permanent criticism of the subject itself since, as Carol
Maier (1994: 67) points out, translation must problematize identity. Translation
as/like a woman does not mean translating bearing in mind the multiple identities
the translator has accumulated throughout her professional career, but rather
translating from a borderland, a reective and self-critical space in which the
representations of the feminine subject translator are constantly modied and
recreated. Perhaps, translating as/like a woman should not be understood as a
utopian or chimerical space of political and social emancipation. Perhaps, instead,
we should begin to read it as a borderland in which identity and textuality are
constantly (re)written from a point of view of commitment and negotiation.
It is important to bear in mind that the existence of a feminine subject
translator, albeit complex, spontaneous, contingent and dynamic, cannot ascertain
the truth of the translated text, but neither can it fail completely to ascertain it.
In reality, translating as/like a woman means accepting the implications of
reading and interpreting from the interweaving of gender/text in which every sign
of gender is negotiable and every translation of a text is unique and unrepeatable.
Thus, if gender is not xed in the text, we may say that all hermeneutic transla-
tion practice as/like a woman carries with it a tug-of-war between text and
author, between text and translator. The gender signs shape the work, which is,
on the one hand, a source of tension which problematises, and on the other, a
negotiable practice on which we can (re)conduct our own reading.
For this reason, if we remove the conventional baggage from the concepts
of gender, sex and women, treating them instead as points which are open
to political debate, the translator can opt for a form of translation which not only
questions identity but also problematises it. The person who translates as/like a
woman refuses to view her practices in terms of constructing normative, static
meanings. Rather, she seeks to ensure that her style of work embraces the
implications of reading and interpreting the interaction of gender/text, keeping in
</TARGET "god">

42 PILAR GODAYOL NOGU

mind that all texts and all subjects say what they say on the basis of what they
do not say. Both a translated text and feminine subjectivity must be described not
only in terms of the content of their discourse, but also in terms of what is
excluded. The maxims: a text expresses what it expresses, but also expresses
what it does not express and a feminine subjectivity contains what it contains,
but it also contains what it does not contain, should be read not as negations
and contradictions but rather as corrections and (re)armations.
Moreover, translating as/like a woman does not presuppose a materialisation
of the feminine subject translator, but neither does it negate it rather, it
problematises it. The identity of all feminine subjects is found, then, in the
critical reection of the processes of exclusion which form the basis on which
the subject establishes her meanings, that is, what she provisionally excludes
from her representation while recognising that this judgment is neither denitive
nor irrevocable.
If the process of translating as/like a woman avoids postulating the existence
of a feminine subject translator as a xed, immobile, absolute category as the
traditional ontological arguments do (Butler 1990: 134), this means it can never
ascertain the absolute truth of any translation. This in turn implies that feminist
translation can only aim for permanent reection and self-criticism in its
representations, its methods, applications, focalisations, textual processes and
provisional tactical decisions.
Finally, to return to the concept-metaphor of the borderlands with which we
began since, although it may seem that we have left it far behind, it has served
as guide throughout this textual journey. Living on the border is an allegory for
translating as/like a woman. In the end, what is translating as/like a woman if not
situating oneself in an indeterminate space, neither inside nor outside, questioning
and problematising ones own identity? It is an activity that involves making use
not of speciously neutral, so-called objective strategies, but rather dynamic
procedures and tactics, which negotiate and are negotiable, open and contingent,
which never assume the absolute totality of a feminine subjectivity in translation.
Thus the non-absolute and non-categorical representation of the feminine subject
in the practice of translating as/like a woman neither means the collapse of nor
demonstrates the utopian nature of its politics of identity. Rather, it means the
demand for a kind of subjectivity dierent from the conventional, predetermined
and immobilised subjectivity of patriarchal discourses. It is a question of a
feminine subjectivity which demands openness in readings and meanings, which
makes plain that both text and translator say what they say on the basis of what
they do not say, and which implies that all feminine subjects neither entail nor
deny textual truth in translation. This means to assert our claim to dierence, to
the feminine, on the basis of questioning and problematising identity in translation.
This is tantamount to living the possible impossibility of translating on the border.
<TARGET "chu" DOCINFO

AUTHOR "Chu Chi Yu"

TITLE "Translation Theory in Chinese Translations of Buddhist Texts"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 5

Translation Theory in
Chinese Translations of Buddhist Texts

Chu Chi Yu
The Hong Kong Polytechnic University

This paper attempts to summarize an ancient debate on translation paradigms

which ran through the entire history of Chinese translation of Buddhism from the
Eastern Han to the Northern Song dynasties. The debate was on two translation
methods known as simple translation and sophisticated translation. In the
following pages, I will try to dene the terminology, trace the evolution of the
theory in chronological order, and analyze the concept of matching the mean-
ing developed by Buddhist translators alongside the debate on simple vs.
sophisticated translation. (All the English translations from the classical
Chinese are mine.)

Terminology

What are known as literal translation and free translation in the West are called
direct (or straightforward) and sense translation in China. In ancient times,
however, Chinese translators of the Buddhist scriptures referred to the binary
poles as simple (or plain) translation and sophisticated (or sometimes
polished, elegant, or elaborate) translation. The so-called simple
translation and sophisticated translation are largely synonymous with direct
vs. sense translation used in modern times, but they are not exactly identical.
Sophisticated translation is usually understood to be sense translation. For
example, a 6th century commentary states that, Zhi Qians translations are
elegant and sophisticated and procure the holy message indirectly. (quoted in
Shi Huijiao 1992: 15). Here sophistication was equated with indirectness
(sense translation).
However, the term sophisticated may sometimes contain the meaning of
44 CHU CHI YU

beautiful, as in The Preface to Fa ju jing (Dhamapada) Zhi Quian (

223253) wrote, Beautiful words are not faithful, and faithful words are not
beautiful. Conveying the Sanskrit meaning, straightforward rendition is
preferred. (1984: 22). Here the beauty of words appears to be incompatible with
faithfulness and straightforwardness. (Usually the opposite of straightforward
is sophisticated.)
The terms simple and sophisticated were also used in discussion of
exhaustive translation and succinct translation. Sanskrit sutras are verbose
and repetitive, while the ancient Chinese placed a high value on brevity of style.
The unabridged rendition, which adhered strictly and often slavishly to the
original, was termed simple translation; that which left out the repetitive,
superuous passages in order to accommodate the Chinese taste was considered
sophisticated. Perhaps we would not distinguish the two methodologies using
direct and sense translations today. This is another discrepancy between
traditional and contemporary terminologies.
There is yet another procedure opposed to sense translation, the translation
of sound, or transliteration. When Zhi Qian criticized Lokaksins ( 160180)
translations as being overly simple with excessive use of foreign sound, he
likened simple translation to transliteration (Shi Sengyou 1995: 270). In
translating between languages generically unrelated and phonetically remote such
as Sanskrit and Chinese, it is usually a case of either translation or transliteration:
there is no grey area in between. (The cases where both meaning and sound
happen to match the original are rare exceptions.) Today, we usually separate
transliteration from direct translation even though transliteration may be
regarded as the most direct direct translation as, after all, transliteration is
almost always used in translating words (as opposed to sentences). Therefore, to
avoid confusion, transliteration is not treated in this paper.
To put it simply, simple translation allows minimal changes, while
sophisticated translation sanctions improvements. In modern terms, simple
translation reveres the authority of the original text; sophisticated translation
emphasizes the quality of the nished product. It is in this sense that the ancient
binary opposition is comparable to modern phraseology, such as literal vs. free
translation.
The general attitudes of the 4th5th century critics towards the so-called
simple and sophisticated translations were not neutral, as can be seen in their
comments on some translations. For example, they made no negative remarks on
the pure simple translations of Lokaksin, but they criticized the work of the
sophisticated translator Zhi Qian as though beautiful in style, confused in
respect of the central theme. The monastic ideal is distorted by his undue
 CHINESE TRANSLATIONS OF BUDDHIST TEXTS 45

elegance of language, and the true taste attened by the extravagance of style
(quoted in Shi Sengrui 1984a: 36). The highly acclaimed translations by Kumara-
jiva, whom modern critics tend to classify as a sense translator, the ancient
scholars never termed as sophisticated; on the contrary, they spoke of his
versions as simple but not uncouth, brief but without losing the essence
(quoted in Shi Sengzhao 1984: 39). Apparently, the early views were in favour
of simple translation.
To what extent, then, are the simple translations simple? A verse from
Diamond Sutra provides a typical example: twelve hundred and fty bhikkhus
(the Buddhas disciples) was translated as ban shisan biqiu bai or, literally,
half thirteen bhikkhus hundred. It is virtually impossible for any reader without
knowledge of the original to understand that the number should be read as
(0.5+13)x100. It is like translating the Roman numerals IV as one ve
with the expectation that the reader will decipher it to mean minus one plus
ve. Even numbers, which are perhaps the easiest to translate, were rendered
unintelligible by early Chinese translators under the guidance of Indian monks.
This example shows that simple is not always as the name suggests.

Development

Let us turn to the theories. The earliest extant discussion on translation problems
is Zhi Qians Preface to Dhammapada, in which two opposing views of
translation were recorded:
Vighna of India came to Wuchang in the third year of Huanwu [224]. I studied
these ve hundred ghatas under him and asked his colleague Zhu Jiangyan to
translate them into Chinese. Although Zhu was well versed in Sanskrit, he
knew little Chinese. He often dictated his translation in Sanskrit or in translit-
eration. As a result, the language of the product was excessively simple and
straightforward. Initially, I objected to its being too inelegant. Vighna said,
Translation of Buddhas teachings should copy his words unadorned, and
reproduce his canons without decoration. A good translator of the Scriptures
should render the texts comprehensible without loss of meaning. People in the
audience all echoed, Laozi said, Beautiful words are not faithful, and faithful
words not beautiful.. Conveying the Sanskrit meaning, straightforward
rendition is preferred. Therefore, the translation was done following exactly
the literal meaning passed over by the translator without any literary embellish-
ment. (Zhi Qian 1984: 22).
46 CHU CHI YU

In this rst debate, the simple school prevailed. Translators before Zhi Qian were
all foreign monks. With their limited knowledge of the Chinese language and the
lack of theory and experience in translation, they had no alternative but to resort
to word-for-word rendering in order to convey faithfully the sacred message,
with the conviction that simple translation is closer to the source (Shi Huijiao
1992: 24). Earlier versions are generally inclined to direct translation. Some
parts follow the original structure, with repetitions and inverted word order and
thus are incomprehensible at the rst glance. (Lu Cheng 1979: 175). This was
the earliest direct translation. Although Zhi Qian was not Chinese either, he
was born in China. From childhood he read Chinese classics and at thirteen
devoted himself to study foreign language books and was procient in six
languages. His knowledge of the Chinese language and literature was far superior
to that of his precursors. He also studied Buddhist sutras extensively. By the time
he began to translate, Buddhist translation had made some progress. (Even
Vighna, who refuted Zhi Qians idea, claimed that translations should be
comprehensible.) He certainly was discontented with translations with inverted
word order, yet he was unable to theoretically discredit the far-fetched argument
of the people in the audience.
Shi Daoan (314385) was, in a manner of speaking, one of the few
professional theorists, for he was only in charge of the translation bureau and
did not do any translation himself. Daoan did not know any foreign language.
He compared several versions when he studied Buddhist sutras and found many
translation problems. He made the rst serious attempt to discuss translation
problems theoretically. In one of his essays, he quoted his colleague Zhao Zheng,
the imperial secretary, serving as the director of the translation bureau, as regards
attitudes towards translation:
Most translators of the past adapted their translations to their contemporary
style, because they disliked the simplicity of the Sanskrit language. With this
I do not agree. The reason is that the aim of translation is to convey the
meaning of the original which is otherwise unobtainable. Why should simplici-
ty of style matter? The simple style of sutras was ordained by the spirit of the
time when they were composed, and it was for a purpose. But if the translation
is incomplete, the translator is to blame (quoted in Shi Sengyou 1995: 382).

Here simplicity essentially means the opposite of brevity. Daoan entirely

agreed with Zhao Zheng and advocated that translation should follow the source
closely without missing a single word, and change nothing except for inverted
word order (quoted in Shi Sengyou 1995: 382). Judging from these quotations,
we may group him with the simple school of translators. The best known of
Daoans theories is the so-called ve types of losses in translation. That is to
 CHINESE TRANSLATIONS OF BUDDHIST TEXTS 47

say that he allowed these losses in translation. The rst two types are: First, the
original word order may be reversed in order to conform to Chinese syntactical
rules. Second, Sanskrit is simple but the Chinese love polished literary style. To
conform to the taste of a large audience, only an embellished style can achieve
the desired eect (quoted in Shi Sengyou 1995: 290).
Zhi Qians simple theory only touched on comprehensibility and
without decoration, but Daoan stated at the outset that he allowed the word
order of the original to be changed. This is the rst time that the simple school
conceded that translations must obey the structural rules of the target language.
The second type is rather confusing: was he advocating polished translation
which he and Zhao Zheng criticized? The only rationalization might be that his
embellished style was meant for the benet of the reader, not for its own sake.
Many of Daoans discussions addressed the question whether translation should
be exhaustive or succinct i.e. the issue of editing out the repetitive parts.
Indeed, the last three of the ve types of losses addressed this problem. In this
regard he took a rm stand: repetitions should be removed. This, again, would
lead to the incomplete translation that Zhao Zheng abominated. In practice,
Daoan once reduced a text with 482,304 Sanskrit words to 195,250 Chinese
characters (Shi Sengyou 1995: 377). (Usually a Chinese translation from Sanskrit
is longer than the original in terms of the number of characters/words). Daoan,
as well as later critics, might like to think that he was a simplist, he neverthe-
less provided a theoretical ground for the sense translator Kumarajiva (344
413), who was still to come.
Daoan, however, did not indiscriminately cut all translations. A story he
recorded at his translation bureau, similar in a way to the above episode told by
Zhi Qian, will illustrate this point.
In the past, I got a book of disciplinary precepts from Faqian of Wusui. I did
not like its simple style. Thinking it was too long-winded and repetitive, I
asked Huichang to remove the repeated passages. Huichang stood up from his
seat and said, On no account should we do that, sir. The precepts are the
same as the Confucian rites. We have Confucian classics, which are
simple and straightforward in style, but no one dares to change it, because we
understand they are the words of the sage kings and should be followed. For
what reason, when it comes to the Buddhist precepts revered by saints and
worthy men, should we make changes so as to accommodate local prefer-
ence?. In the end, the translation adhered closely to the original; only the
inverted word order was reversed (quoted in Shi Sengyou 1995: 413).

Why didnt Daoan insist on weeding out the superuities and repetitions this
time? The words of Huichang (who acted as the pen-man, or the assistant who
48 CHU CHI YU

took the dictation) provide an answer: because they were translating Buddhist
precepts, similar to todays legal work. Daoan recorded this story because he
was vaguely aware of the relationship between text type and translation strategy,
even though he did not develop what we call translation typology today.
In fact, Daoan did not put too much emphasis on the distinction between
simple and sophisticated translation. He spoke highly of Parthamasiriss (
140) work, saying that he either replaced simplicity with sophistication, or
preserved the simplicity without decoration. Superb was Parthamasiris, who
captured the central theme (quoted in Shi Sengyou 1995: 367). To him, it seems,
as long as the translation captured the central theme, it was not essential
whether its style was simple or sophisticated. Daoans contribution to
Chinese translation theory consisted of objectively freeing the translator from the
yoke of simple translation.
Kumarajiva (350409), one of the four greatest translators in the history of
Chinese tripitaka, began his translation career about twenty years after Daoans
death. His translations are unparalleled either in terms of translation technique
or degree of delity (Lu Cheng 1979: 88). One of his famous remarks is that
translation from Sanskrit into Chinese could never capture the avour of the
original style, even though the main ideas can be more or less conveyed. It is
like feeding a person with food already masticated by another in that not only is
the taste lost, but it might cause nausea (Kumarajiva 1984: 32). Kumarajiva is
generally viewed as a sense translator. A typical example in this respect is also
a frequently quoted story:
[Shi Seng] Rui participated in Kumarajivas translation project. A line from
Lotus Sutra translated by Dharmaraksha reads: Heaven sees man, and man
sees Heaven. When Kumarajiva came to this line, he said, This is the same
as the original, but the language is too simple. Rui asked, Doesnt that mean
Heaven and man are united and see each other? Exactly, exclaimed
Kumarajiva with delight (quoted in Shi Huijiao 1992: 244).

Sengrui (371?-438?) acted as Kumarajivas pen-man. He was sceptical about

Kumarajivas translation method at the beginning, but he soon realized that the
masters translations did not diverge from the central theme even though his
wording might not exactly match that of the original (quoted in Shi Sengrui
1984a: 34). This time the sophisticated school prevailed, but only in practice,
not in theory. (In fact the so-called sophisticated school never developed a
theory of its own.) With all his opposition to excessively simple translation,
Kumarajiva changed some of Zhi Qians sense translation back to translitera-
tion. To him, it seems, certain Buddhist concepts could not be rendered into
Chinese by sense translation. Kumarajiva left little discussion on translation
 CHINESE TRANSLATIONS OF BUDDHIST TEXTS 49

problems. His views are scattered in the writings of his assistants and critics,
among whom the most theoretically perceptive was Daoans student Shi
Huiyuan (344416).
Like his master Daoan, Huiyuan did not translate himself, but he was an
astute critic and editor of translation. In his study of translation methodologies of
the Han and Jin dynasties, he concluded: In their translations, either the
meaning is obscured by the owery style, or the argumentation weakened by the
simplicity of diction (quoted in Shi Sengyou 1995: 391). He argued strongly
against the inexibility of translation methods adopted by both schools because
they disregarded the nature of the text being translated. He said, The saints set
down their directives in dierent styles as the subject matters required. To
translate a simple text using a sophisticated style gives rise to doubts; on the
other hand, to simplify what is sophisticated in the original pleases few (quoted
in Shi Sengyou 1995: 391). Contemporary theorists view this remark as a
compromise after hundreds of years of confrontation between the two schools.
But the real issue is whether the words simple and sophisticated were clear
enough to assist the working translator, and to describe the complicated phenom-
enon of translation.
Scholars tend to think of Kumarajiva as a sense translator, mainly because
his versions are succinct. But when Huiyuan edited Kumarajivas Treatise on
the Great Perfection of Wisdom, he cut the hundred juan [chapter], which had
already been abridged by Kumarajiva, to twenty juan. He gave his reason in the
preface:
In translation from Sanskrit, simplicity [verbosity] of style may cause obscurity
of meaning, and closeness in diction may lead to remoteness in theme. If the
meaning is obscure, the imagery will be lost, and if the theme becomes
remote, the subtleties will be hard to detect. An easy path will benet the
beginner; gradual progress brings enlightenment. Therefore, [Hui] Yuan
eliminated the redundant, cumbersome passages in order to highlight the
central theme. As a result, the translation is well balanced in style without
deviating from the original meaning (quoted in Shi Sengyou 1995: 391).

Huiyuan believed that a long-winded translation would cause unnecessary trouble

for the readers, especially beginners. In translating dicult and intricate
treatises, therefore, the translator should select and bring forward the important
points so that the reader could easily grasp the meaning. He knew the translation
was meant for helping the monks cultivate their spirits, so the easier the transla-
tion was to understand, the better results it would achieve. These treatises were
dierent from the Buddhist precepts that Daoan had helped translate. Putting
together Huiyuans remarks and what Daoan had to say about the translation of
50 CHU CHI YU

precepts, we nd that the Chinese Buddhist translators proposed strategies that

varied according to dierent types of texts and purposes of translation 1,500
years ago. They still used terms like simple and sophisticated, but their
attention had turned to the text, the translator, and the reader.

Matching the meaning

Even though the ancient translators placed considerable emphasis upon direct
(simple) translation, in practice, most of them, particularly the earlier ones,
unwittingly adopted a procedure which was actually sense (sophisticated)
translation. The procedure was called geyi or matching the meaning. In their
lectures, the Buddhist masters often used Daoist ideas as analogies to expound
certain dicult concepts for which there were no Chinese corresponding
expressions. For example, in one of Huiyuans lectures, some guests questioned
him about the theory of reality. Though the discussion continued back and forth
for some time, they became increasingly doubtful and bewildered. Thereupon
Huiyuan quoted ideas from Zhuanzi (a Daoist classic) as an analogy, and in this
way the sceptical came to understand.
This method of matching the meaning, when applied to translation, is
similar to Nidas dynamic equivalence, and it is sense translation in both
name and nature. One typical example is the concept of shouyi, or mind
concentration (used to translated the Sanskrit concept dhyana, a state of serene
contemplation attained by meditation), which is a Daoist method of body and
spirit cultivation. Although the misunderstanding caused by the translation proved
to be mutually benecial to the development of Buddhism and Daoism, it in fact
confused them at the same time. As the method was mostly used in translating
certain concepts, which seldom went beyond the word level, it was mistaken for
simple translation. This is also an indication that the so-called simple
methodology was, in the main, word-for-word translation.
Daoan later found the method to be contrary to reason (quoted in Shi
Huijiao 1992: 195). Shi Sengrui also said that matching the meaning is indirect
and misses the source (1984b: 37). Interestingly enough, it was Kumarajiva, a
sense translator, who corrected this procedure and provided authoritative
interpretations and translations.
Xuanzhuang (600664) of the Tang dynasty, arguably the greatest of all
Buddhist translators, gave an incisive analysis of the problem of matching the
meaning. Ordered by the emperor to translate the Daoist classic Dao de jing into
Sanskrit in order to promote Chinese culture in the West, he called in a group of
 CHINESE TRANSLATIONS OF BUDDHIST TEXTS 51

Daoist priests to clarify some subtleties of the text. To his surprise, the priests
quoted Buddhist sayings to explain Daoist concepts. Xuanzhuang asked, Bud-
dhism and Daoism are as far apart as the heavens and the earth. How can you
use the words of the Buddha to explain Daoist classics? If you trace the words
to their origins, they are totally unrelated. The Daoist priests argued, But
Sengzhao [a Buddhist scholar] explained Buddhist concepts with Daoist ideas in
his treatises. Xuanzhuang replied, When Buddhism was rst introduced to
China, there were too many dicult texts, so Laozis Daoist ideas were
borrowed to explain the words of the Buddha. Sengzhao only used a method of
analogy in his treatises. How can one understand metaphors literally? (quoted
in Shi Daoxuan, 1891:9a). Xuangzhuang thus distinguished between the methods
of translating and original writing.
The Song monk scholar Shi Zanning (9191001) discussed in more detail
the procedure of matching the meaning in his analysis of translation styles. He
set himself squarely against using external (non-Buddhist) concepts in transla-
tion of Buddhist scriptures. His argument was presented in the form of a
dialogue:
Question: Do those who use the language of Chinese writings in translating
from Sanskrit abuse the Confucian classics?
Answer: No. But one can use it only if the language does not involve classical
allusions or the teachings of philosophers and historians. How can one mix the
language of the street into the translation?.
Question: This method of using Chinese classics was prevalent during the Han
and Wei dynasties, and as a result, Chen Shou wrote in The Record of the
State of Liner [Lumbini] of The Annals of the Three Kingdoms: What is said
in the Buddhist scriptures is similar to the ideas of Laozis Dao de jing. This
is because Laozi taught the Indians Buddhism when he went west of the pass
to India.
Answer: This mistake was caused by the misuse of Daoist terms by translators
which led Chen Shou to the belief that Buddhism and Daoism were related.
We learn from this that those who use external ideas in translation deserve to
be criticized. Kumarajivas version of the Lotus Scripture of the Mysterious Law
can be said to have struck a balance [between simple and sophisticated
methods] and retained the natural avour of the Sanskrit language.To use
crude and slangy language is to write like butchers and wine vendors. But if
we mix Chinese classics into the translation, we might as well not translate at
all. On the other hand, it is still better to be crude than to use allusions. But a
translation given over to crude style is seriously wrong. When we take care of
the balance, the language of the canons will take care of itself. This is the
essence of translation (quoted in Shi Zanning 1987: 5556).
52 CHU CHI YU

Zanning had denitions for his terms: elegant referred to the style of the
written language (language of the classics); crude to the spoken form (language
of the street). Translation, according to him, should be elegant. But this principle
must not be taken in terms of absolutes, as he knew that written Chinese was
mainly from Confucian and Daoist literature. Using this kind of language in
translation would inevitably bring Chinese concepts into Buddhist sutras, which
might lead to absurdities like the one found in the Annals of the Three Kingdoms,
and which, more importantly, might confuse the teachings of the Buddha. He
therefore drew the line: to use written Chinese which did not involve allusions;
otherwise, he would rather choose crude style, but not so crude as the language
of butchers and wine vendors. This is what he called balance.
Xuanzhuang separated original writing from translating, and Zanning went
one step further by distinguishing between language problems and cultural
problems in translation. As all allusions (as well as teachings of philosophers and
historians) are by denition tinged with a very strong cultural colour, the
slightest misuse might cause unnecessary confusion or misunderstanding.

Conclusion

Although the debate on simple and sophisticated translation was the central
issue of the theory of Buddhist translation in China, it was not its main achieve-
ment. It only provided a forum for discussion. The theory of simple and
sophisticated translation developed hand in hand with the practice of Buddhist
translators. The success of Parthamasiriss renditions led to the belief that
simple translation was close to the source. Later, ably assisted by the pen-
man at the translation bureau, Daoan discovered the relationship between text
type and translation strategy, and thus freed the translator from blind belief in
simple translation. In cutting redundant parts for his readers, Huiyuan brought
in the idea of the intention of the translator. Translation practice now began to
move towards sophistication, which Kumarajiva and Xuanzhuang brought to
perfection. But just because Kumarajiva and Xuanzhuang produced the authorita-
tive versions of Buddhist scriptures, it does not follow that sophisticated
translation is a preferred or more advanced methodology. Their new versions,
so were they called traditionally, are certainly more accurate and readable than
the old ones, but more importantly, they correct the errors of the old versions,
some of which are excessively simple, others overly sophisticated. Xuan-
zhuang and Zannings critiques of matching the meaning, particularly
Zannings analysis of translation style, are the highest achievements of the theory
</TARGET "chu">

CHINESE TRANSLATIONS OF BUDDHIST TEXTS 53

of Buddhist translation. The Buddhist writings on translation are usually short, or

succinct but they touch upon the issues of translator, reader, text type, and
language and culture, which are also major concerns of modern translation
theory.
<TARGET "tan" DOCINFO

AUTHOR "Helena Tanqueiro"

TITLE "Self-Translation as an extreme Case of the author-Translator-Dialectic"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 6

Self-Translation as an extreme Case

of the AuthorTranslator-Dialectic

Helena Tanqueiro
Universitat Autnoma de Barcelona

1. Introduction

The ideas presented in this article are largely the fruit of personal experience
gained from translating El Cam de Vincennes by Antoni Mar into Portuguese.
This work concerns the friendship of Rousseau and Diderot while the latter was
a prisoner in the prison of Vincennes, and their ideological disagreements. The
work was originally written in Catalan and was then translated by the author
himself into Castilian Spanish. When I contacted the author for help in clarifying
a few diculties that had arisen, he told me that he too had gone through the
same process and suggested that he might send me his translation in the hope
that it would be of some help. Since I know both languages, I was now faced
with two works instead of one; and as a translator, I was in the lucky position to
have access to an especially privileged interpretation and translation of the work
I was myself translating, since it had been done by the author himself.
On the basis of a comparative study of the original work in Catalan and the
authors own translation into Spanish I was able to conrm that the author had
completely assumed his role as translator and that it constituted an extreme case
in the eld of literary translation since a single person was bringing to bear two
sets of skills and carrying out two tasks which are generally performed by two
dierent people. Study of this self-translated work permitted more objective
analysis of the specic role of the translator in comparison with conventional
translations, i.e., translations carried out by a separate translator.
For, no matter how close the author-translator relationship may be in
conventional translations and no matter how much the translator strives to reach
the real intentions of the author, there are always elements coloured by the
subjectivity of that other party to the process: i.e., the translator. It is all these
56 HELENA TANQUEIRO

elements that have always proved an obstacle to more precise and accurate study
of the complex product of the translators work.
In one of his communications on The task of the literary translator, the
prestigious translator of Portuguese literature into English, Giovanni Pontiero,
echoed the view that one of the main diculties faced by translators is We
must never forget that translation and translators are vulnerable to attack from all
quarters (quoted in Pontiero 1977: 62). I feel that study of a work translated by
the author his or herself may help to clarify and dene the role and the compe-
tence of the literary translator, since the third party mediator has been side-
stepped.
This article is based on a conception of the relationship between author and
translator which sees the role of the translator as one of co-authorship of the
literary work being translated, and this is especially so, of course, in cases where
the translator is also the author of the original work. From this basis, I suggest
that the study of self-translations could constitute another tool which might make
a worthwhile contribution to the theory and practice of literary translation, since
the self-translation is essentially free of external noise or secondary inuences
which could have the eect of distorting the results drawn from conventional
translation studies.

The author-translator relationship

There seems to exist a widespread consensus in translation studies that what we

hear in the translation of a literary work should essentially be the voice of
the author rather than that of the translator, although nalistic theories, for
example, would seem to diverge on this point. Christiane Nord claims what is
actually translated is not the senders intention but the translators interpretation
of the senders intention (Nord 1997: 85). In a similar vein Frances Aparicio
points out, Today translation is not simply a product, a derivative second text,
of parasitic value. To translate implies to interpret, to create. The process is
analogous to the process of creation and is innate to the poetisation of reality. It
is a way of seeing and reading our world (1991: 1415).
Although there still are authors who defend the essential impossibility of
literary translation, and others who demand that the translator be totally subordi-
nated to the original text, such as Milan Kundera who holds that the most
beautiful translation is the one which is faithful, today, the link between author
and translator is increasingly close, as a result of the new communication
technologies, and it is becoming a form of interaction an exchange of ideas
 SELF-TRANSLATION 57

between two language professionals. Antonio Tabucchi, for example, is Italian

but also masters Portuguese language and culture and as a bilingual and a
biculturalist, he follows closely the translation of his works into Portuguese and
Italian. He also spends long hours in conversation with the translators of his
works into other languages, providing them with help and support. Fruit of that
contact is the book Conversaciones con Antonio Tabucchi by Carlos Gumbert, his
translator into Spanish. Similarly, Gnter Grass opts for holding seminars and
lectures with his translators, since his works are complex and dicult, but again
the idea is to grant the translators more independence; indeed, he tends to say,
I have done my part. Now the work is yours. This was also the view taken by
Jorge Luis Borges who told Gregory Rabassa, the translator of his work into
English, Dont translate what Ive written, but what I wanted to say (quoted in
Pontiero 1997: 65). For Jos Saramago, the very act of writing is an act of
translating, To write is to translate. It will always be like that. Even when we
are using our own language (). The source text represents only one of the
possible translations of the authors experience of reality, and the translator has
to convert this translate-text into a text-translation which is of necessity
ambivalent () (1997: 85).
These examples, which for reasons of space I cannot expand on here, reect
a clear change, at least in our eld of studies, in the relationship between author
and translator and, by extension, in the role assigned to the literary translator,
which is increasingly one of co-author, in Russian translation studies tvorceski
vossosdat original (Umbreit 1997: 547). Surely, this change is based on our
growing acceptance of the fact that any translation is just one translation among
many possible translations of an original, which itself was only one among
many possible originals, as pointed out by Saramago. Similarly, Jos Lambert
says that the original is never the only model for a translation (1995: 193). To
the original work must be added, amongst other things, the translators subjective
baggage and this is a factor which conditions and increases the complexity of the
translation process: the task of the translator and the function of the translation
can no longer be determined exclusively in terms of faithfulness to the author
or to the original; rather we must consider other inuences in the translation
process, despite the fact that in general, translators seek to enter into close
contact with the author so as to express more accurately his or her voice.
For analysis of literary translation there is however a problem of references:
if the product to be analysed is just one of the possible translations and the
translation scholar only has access to this one single product, which is packed
with multiple subjectivity, then whatever conclusions are drawn will have little
value for extrapolatory purposes since they will be of limited objectivity. This
58 HELENA TANQUEIRO

freedom which is increasingly bestowed on the translator can be exemplied

in a rather extreme case: self-translation. This is precisely why I feel that self-
translation may well cast light on some theoretical questions which as yet remain
to be answered. Study of a work which has been translated by the author himself
may permit us to determine when and how, in other words, in what translation
situation and by means of what strategies, the self-translator sometimes decides
to follow the route which he himself carved out in the original work and
sometimes decides to beat a dierent path, safe in the condence that comes
from the authority of authorship.

The self-translator as translator

Therefore, the dialectic relationship represented by:

Author / work L1 translator / work L2

nds its extreme expression when authors translate their own works, that is,
when they self-translate, as follows:
Author [ / workL1 translator / ] workL2
While it is true that throughout history there have been many writers who wrote
in more than one language, such as for example, Paul Celan, Derek Walcott,
Samuel Beckett, Primo Levi, Jorge Semprn, Antonio Tabucchi, it is nevertheless
interesting to see that only a few, very few indeed, actually translated their own
work, despite the fact that all of these, and other important writers such as
Hlderlin, Ezra Pound, Valry, and so on, devoted much of their lives to
translation. It is interesting to note that Paul Celan, to take just one example, one
of the great literary translators whose name is today linked with the most
prestigious prize in German language literary translation, never translated any of
his own works, perhaps for the reason indicated by Antonio Tabucchi, () and
nally, and who knows whether this is not the main reason, I have always been
afraid, I have never been brave enough to walk along both my linguistic and
aective shorelines at the same time to use psychoanalytical terms. I have
often gone to the other shore, but I could never come back in the same boat
(quoted in Gumpert 1995: 186187).
If we focus on authors who did actually accept the challenge of self-
translation, such as Beckett, Joyce, Nabokov and Kundera, we intuitively sense
that they acted more like translators than authors. Nabokov, for example,
confesses that his aim was to make his already written work known among the
 SELF-TRANSLATION 59

huge body of English language readers. A clear case of how an author-translator

keeps a distance between both roles when self-translating is Milan Kundera, who
signs his own translations of his works under a pseudonym. Another well-known
self-translator was Samuel Beckett, who wrote both in English (his mother
tongue) and French and who later translated his books into French and English
respectively. He clearly accepted his role as translator although he admitted to
being tired of it, Sick and tired I am of translation and what a losing battle it is
always. Wish I had the courage to wash my hands of it all (quoted in Fitch
1988: 9). So good was the critical reception given his translations that Melvin J.
Friedman was moved to comment Ses propres traductions () peuvent presque
toujours passer pour des versions originales sauf pour les beckettiens les plus
purs () (Friedman 1964: 11).
The main reason which justies a view of the self-translator as more a
translator than an author, is that despite the double nature of the activity of
writing/translating, he or she still enjoys a level of freedom, which is rarely
bestowed on translators, and further, despite the privileged access self-translators
have to their own creative processes and therefore to the real intention of the
author-creator, by the time they begin to translate their own work, the process of
creating a ctional universe is well and truly over, just as it is over when
translators in general begin to work.
Of course, it cannot be overemphasised that the self-translator is a privileged
translator, since in terms of subjectivity there will be no gap between the author
and translator; he will never unwittingly misinterpret his own work, and this
undoubtedly confers great authority on any translator; despite not having the
liberty to change the established ctional world of the literary work, he may well
decide to add to the work in some way since he still maintains his status as an
author; he can move more condently in constructing a new linguistic universe
since he will not be conditioned by the linguistic universe of the source language
and he will know with the utmost certainty when he is justied in departing from
the original text and when he is not, since he knows perfectly just how he
originally concretised his thoughts through words, that is, he knows when these
words are the only words which will serve and when they are only one set of
words among many other equally valid sets.
Since authors then, to my understanding, see themselves more as translators
than authors when they translate, study of self-translations could bring up
extremely interesting information for studies of the process of literary translation
and could well cast light on some of the problems at the very centre of our
discipline, many of which arise due to the diculty of reaching any denite
conclusions about a process which involves two dierent people, each with two
60 HELENA TANQUEIRO

dierent subjective views of reality. This potential is also mentioned by Brian

Fitch in his study of Beckett, where he points out that studies of Beckett either
deal exclusively with one language version, or at most compare both versions for
evidence of the authors bilingualism. This leads Fitch to conclude that the aim
should be to seek to grasp the exact nature of the relationship between the two
versions. It goes without saying that the results of such an enterprise ought to
enable us to understand better the activity of the self-translator and the process
involved therein. They would therefore constitute a valuable contribution to
translation (studies) () (1988: 15).
The unquestionable value which analysis of self-translation holds for
translators who, like myself, are entrusted with the translation of the same work
into another, third language is hinted at by the great writer Cabrera Infante, who
says I have tried to have the versions because they are something more than
translations of my books in English serve as a prototype for future translations.

Analysis of self-translation in the theory and practice of literary translation

Let us start by assuming that the author-translator, when involved in the process
of creating a literary work, is absorbed in a complex system in which either
consciously or intuitively he interacts with multiple factors which will play a role
in determining the nature of the nal result. Two of these factors are especially
important for my purposes here: rstly, the construction of the ctional universe,
and secondly, the way in which this universe comes to life, through its transla-
tion into words which occur to the author at a given moment as the best ones
to translate his ideas and thus give form to the artistic creation.
When the author-translator take on the role of translator (and this is much
easier to detect in cases of self-translation, since author and translator are one
and the same person), his creative work is now channelled more directly into
linguistic creation as opposed to the processes of non-purely linguistic ctional
creation that give rise to characters, the relationships between them, the construc-
tion and denition of time and space, the narrators perspective, and plot
coherence.
Similarly, since the author is absorbed in the complicated process of
building this ctional universe, which becomes as real to him as outside reality
and life itself, he may sometimes neglect to some extent certain aspects related
more directly to linguistic expression, style, rhythm, the music of the words,
prosody, cohesion, and so on, or indeed the veracity of some seemingly factual
claims. All of these are aws the author-turned-translator is quite likely to detect
 SELF-TRANSLATION 61

and revise when self-translating. When the translation is done by a translator

other than the author, the same thing occurs but with the dierence that the
translator follows the standard procedures, strategies and techniques of the
experienced translator while the author may perhaps do it more instinctively.
However, it must be pointed out that despite the dual nature of his role, the self-
translator will normally refrain from interfering with the ctional universe
already built into the work and for this reason will see his translation work as
concerning other more linguistic areas, such as the appropriate and creative use
of vocabulary, morphosyntactic structure, eective use of rhetorical devices,
appropriate style, textual cohesion and coherence. Not unlike any other translator,
he will accept that he cannot interfere with the already established denitions of
character, plot, time and space.
From the perspective of literary translation theory, comparative studies of
self-translated works oer, to my understanding, certain advantages over the
study of conventional translations: they could help to illuminate certain inaccessi-
ble recesses for theory, they might also make a contribution to scholarly debate
and serve to conrm results reached via other routes.
I do not propose to bear out here the benets of this approach by examples
drawn from my previously mentioned experience with El Cam de Vincennes,
since to do so in any adequate form would require much more space than
available; this would be especially so since if we were to try to determine the
translation strategies in use we would have to transcribe fairly long passages.
Furthermore, it is a self-translation from Catalan into Spanish, languages which
are closely related, and as a result the strategies employed would not emerge at
surface level but only after a deeper study, and also many of the essential points
of the study would pass undetected by any reader unfamiliar with Spanish and
Catalan (and by extension, Portuguese). I must also mention that the work in
question concerns the history of thought and does not oer much potential for
the clarication of interesting points with respect to translation theory, such as
cultural marks, dierent registers, dierences in readers previous knowledge,
and so on, which are all points which could emerge from the study of other self-
translations.
However, I do wish to highlight one result which I feel is highly signicant
and which encourages me in my belief that the study of self-translation may be
of great use: the free versus literal translation debate does not nd any echoes
here at text level, but dierences can be noted depending on the given moment
in the plot development and the weight of this within the work as a whole.
Antoni Mar, who says that it was instinct or intuition that guided him in
translating his own book, corrected errors, smoothed out incoherence and
62 HELENA TANQUEIRO

cohesive defects, and he generally rendered a relatively freer translation when

evoking images for example when describing a landscape or the hubbub of a
Paris street-scene, than when focussing on the books central concern, which is
transcription, in the form of dialogues, of the philosophical insights of Diderot
and Rousseau. In these passages his translation is much more strictly literal in
approach in fact it runs parallel line-for-line with his original as if to highlight
that the words employed in the original version had to be retained intact.
Returning to the practice of translation, it is I feel beyond doubt that
translating a work which has already been self-translated by its author into a
third language involves a number of advantages for the translator. With regard
to my own experience, I can say that it lent me both extra condence and
freedom when it came to formulating the text in Portuguese, since I was able to
draw support from the authors self-translation during the dierent phases of
analysis, transfer and synthesis throughout the translation process.
Since the author himself had almost instinctively rewritten his text when
translating and had claried dicult passages and rectied what were clearly
errors, this became for me a source of incalculable help in my research and
interpretation of the original text in Catalan, which served as the basis for my
translation. When it came to the transfer phase I beneted from the fact that, in
addition to the standard almost obligatory translation procedures, the self-
translator in the freedom conferred on him as author frequently made use
of such additional resources as omission, expansion, condensation, substitution,
which translators are generally rather more reluctant to use. The use of these
options opened the door to a greater freedom in decision making: I felt free to
make a choice between the original, the self-translation and my own preference.
For the rewriting of the translation I was lucky enough to be able to draw on
areas of the authors self-translation where once again, secure in the freedom that
comes of being author as well as translator, he had added further information to
improve the texts cohesion and coherence, a feature which is unique to self-
translation.

Conclusions

My aim in this short article was to show that self-translation pertains to the
specic area of literary translation and that its study could prove highly fruitful,
and indeed could even constitute an alternative line of study, within literary
translation theory. I have tried to show that the author, when deciding to self-
translate, plays more the role of translator and rather less that of author, mainly
</TARGET "tan">

SELF-TRANSLATION 63

because he is constrained, like any translator, by the existence of a preestablished

ctional universe in the literary work; nevertheless he feels at greater liberty to
modify the linguistic dimension of the work in question, since he is not condi-
tioned by the linguistic universe of the source language. Self-translation then, can
be seen as an extreme case of the author/work translator/work dialectic, in
that a single person carries out two separate tasks normally carried out by two
separate people. For translation studies, this could open up a eld of study and
research which has yet to be explored and which may help us to examine and
dene more objectively such specic aspects of the nature of translation as the
concepts of translation loyalty, faithfulness and freedom and which may
also help us to conrm or question results stemming from other sources.

Translated by Jacqueline Minett

References

Mar, A. 1995a. El cam de Vincennes. Barcelona: Edicions 62.

Mar, A. 1995b. El camino de Vincennes. Barcelona: Tusquets Editores.
Mar, A. 1997a. O Caminho de Vincennes. Translation: Helena Tanqueiro. Lisboa:
Editorial Teorema.
<TARGET "izq" DOCINFO

AUTHOR "Isabel Garca Izquierdo"

TITLE "The Degree of Grammatical Complexity in Literary Texts as a Translation Problem"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 7

The Degree of Grammatical Complexity in Literary

Texts as a Translation Problem

Isabel Garca Izquierdo and Josep Marco Borillo

Universitat Jaume I, Castell

Within the framework of translation-oriented text analysis (cf. Nord 1991), the
grammatical complexity of sentence structure is one element of style to be taken
into account. In fact, when it comes to literary texts, varying degrees of gram-
matical complexity may be said to mark the authors stylistic intentions, thus
providing a direct link to the works basic thematic concerns and the kind of
experience it attempts to convey.
It is a well-known fact that not all languages accept the same degree of
grammatical complexity. In the present article we are therefore not so much
concerned with the contrastive aspect of grammatical complexity as with its
potential as an expressive device. While contrastive studies generally aim to
register the similarities and dierences in the discursive practices of two or more
languages, attempting to describe what is normal or unmarked for each language,
the present article will deal with the creative, and therefore unpredictable use
which literary discourse (as exemplied in English literary texts and their
corresponding translations into Spanish and Catalan) makes of grammatical
complexity.
Before engaging in a detailed analysis of source and target texts, however,
we will do well to determine precisely what aspects of the issue of grammatical
complexity may be of interest to the translation scholar. Once again, with special
regard to literary translation, we believe it is fundamental to take the following
two factors into account:
(1) the degree of grammatical complexity present in a text, if possible
in quantitative terms;
(2) the stylistic function of varying degrees of grammatical complexity,
in accordance with the use made of each in a given text.
66 ISABEL GARCA IZQUIERDO AND JOSEP MARCO BORILLO

Determining the degree of grammatical complexity

In an attempt to determine the degree of grammatical complexity of a sentence,

one essential variable to be taken into account is that of the mutual dependency
of its component clauses. Halliday (1985: 198 .) identies two types of interde-
pendency between clauses, namely parataxis and hypotaxis, corresponding to the
traditional distinction between coordination and subordination.
While discussing the sentence (which he refers to as a clause complex) as
a unit of analysis, Halliday (1985: 201202) makes the following suggestive
statement:
The clause complex is of particular interest in spoken language, because it
represents the dynamic potential of the system the ability to choreograph
very long and intricate patterns of semantic movement while maintaining a
continuous ow of discourse that is coherent without being constructional.

Yet this quotation merely suggests what Halliday has openly expressed else-
where: contrary to what is often said or assumed, oral discourse is characterised
by grammatical complexity, whereas the main feature of written discourse is
lexical density. Halliday substantiates his view by arguing that written discourse
does not require a high degree of grammatical complexity because the fact that
the speaker can plan ahead what s/he wants to say brings about a considerable
compression or condensation of meanings. In other words, in written discourse,
meanings may in fact be conveyed through the use of relatively simple struc-
tures. Oral discourse, on the other hand, due to time constraints and the inability
of the speaker to plan ahead, must resort to the use of more complex grammati-
cal structures in order to compensate for the relative semantic poverty of its
lexical content. Eggins (1994: 6061) oers the following account of the issue:
Halliday points out that [] spoken language has a higher level of grammati-
cal intricacy. Grammatical intricacy relates to the number of clauses per
sentence, and can be calculated by expressing the number of clauses in a text
as a proportion of the number of sentences in the text. Whereas in spoken
language we tend to chain clauses together one after another, to give often
very long sentences, in written language we tend to use relative few clauses
per sentence.

Positing a direct relationship between relative degrees of complexity and modes

of discourse (oral or written) opens up interesting perspectives from which to
study texts. These perspectives will be specied later on. Beyond this, however,
it should be noted that the Eggins passage quoted above also indicates how the
degree of grammatical complexity in a text may be quantitatively determined,
 GRAMMATICAL COMPLEXITY AS A TRANSLATION PROBLEM 67

i.e., by dividing the number of clauses by the number of sentences. Nevertheless,

if we take into account the fact that hypotactically constructed sentences are
perceived as more complex than their paratactic counterparts, the formula may
be rened to assign double value (2) to hypotactic clauses, and single value (1)
to paratactic clauses. In this way, a sentence containing three paratactic clauses
(3 1 = 3) and two hypotactic ones (2 2 = 4) will be assigned the value of 7
(3 + 4 = 7). The formula thus revised may be represented as follows:
Degree of grammatical complexity of a sentence = number of paratactic
clauses + (2 number of hypotactic clauses).

Examples of how this formula may be applied to actual texts are provided later
on in this article.

The stylistic function of the degree of grammatical complexity

For the translation scholar, however, applying the formula described above is of
interest only as a preliminary step toward discovering the stylistic function of
grammatical complexity, and how it contributes to textual meaning. To this end,
a scale may be established, ranging from zero signication (when grammatical
complexity makes no apparent contribution to meaning) to the more or less
iconic value of grammatical structure (when the relationship between meaning
and grammatical form is non-arbitrary or motivated).
The relationship between elements of grammatical structure and aspects of
a texts overall meaning is most signicant when it can be referred to as iconic,
in the sense that Leech and Short (1981) aord to the term. According to these
authors, iconicity occurs mainly, though not exclusively, in literary texts, and
may be dened as follows (1981: 233235):
A code is iconic to the extent that it imitates, in its signals or textual forms,
the meanings that they represent. The code of trac signs is largely iconic: a
crossroads is signalled by a cross, a narrowing road by converging lines, etc.
The maritime ag code, on the other hand, is non-iconic: there is no connec-
tion between the colour and design of a ag and the meaning [] which it is
used to signify. One of the tenets of modern linguistics is that language is
essentially non-iconic, that the form-meaning relationship is arbitrary. []
Against this orthodox view it can be urged, that iconicity is inherent in
language in a way that the mention of odd words like miaow and thunder does
not begin to show. [] Presumably, then, we are conditioned to expect that
language, for all its arbitrariness, is in various ways an iconic mirror of reality.
It is in the nature of literature to exploit these iconic possibilities: to bring out
68 ISABEL GARCA IZQUIERDO AND JOSEP MARCO BORILLO

associations between form and meaning which are ordinarily dormant. [] It

[iconicity] embraces not only onomatopoeia and sound symbolism, but the
miming or enactment of meaning through patterns of rhythm and syntax.

As we will see in short, the iconic function of grammatical complexity is an

important issue in the translation-oriented stylistic analysis of literary texts, as it
posits a very close link between grammar and meaning.

The degree of grammatical complexity in translation: problems and solutions

In the following pages we will be examining several samples of literary texts

whose degree of grammatical complexity may be assigned some type of stylistic
value or (more or less iconic) function. At the same time, we will also consider
ways in which translation problems linked to this linguistic phenomenon are
confronted. As mentioned earlier, the original sample texts under discussion here
are in English, and they are presented together with their Catalan and Spanish
translations.
Our rst text is an extract from Ernest Hemingways novella The Old Man
and the Sea (pp. 2325 of the 1993 edition). Text 1, below, is a representative
paragraph taken from this passage:
Text 1
Before it was really light he had his baits out and was drifting with the
current. One bait was down forty fathoms. The second was at seventy-ve and
the third and fourth were down in the blue water at one hundred and one
hundred and twenty-ve fathoms. Each bait hung head down with the shank of
the hook inside the bait sh, tied and sewed solid, and all the projecting part
of the hook, the curve and the point, was covered with fresh sardines. Each
sardine was hooked through both eyes so that they made a half garland on the
projecting steel. There was no part of the hook that a great sh could feel
which was not sweet-smelling and good-tasting.

An analysis of the degree of grammatical complexity to be observed in the

selected passage and its corresponding Catalan and Spanish translations yields the
results shown in Table 1, below. (Once again, we recall that the results recorded
here are based on a two-page passage which, for lack of space, has not been
fully reproduced in this paper.)
In light of this data, the average rate of grammatical complexity in the three
passages under analysis is 5.25 for the source text, 5.57 for the Catalan transla-
tion, and 4.63 for the Spanish translation. The dierences between the three are
minimal, and are often due to contrastive reasons, rather than being the result of
 GRAMMATICAL COMPLEXITY AS A TRANSLATION PROBLEM 69

Table 1: The Old Man and the Sea: source text (pp. 2325), Catalan translation (pp.
8082), Spanish translation (pp. 3133)
English source text Catalan translation Spanish translation
Sentence Rate Sentence Rate Sentence Rate
01 ( her.) 7 01 ( lestima.) 9 01 ( la mar.) 1
02 ( woman.) 6 02 ( dona.) 6 02 ( la quieren.) 5
03 ( masculine.) 100 03 ( el mar.) 8 03 ( mujer.) 6
04 ( enemy.) 1 04 ( enemic.) 3 04 ( el mar.) 9
05 ( them.) 9 05 ( tarann.) 130 05 ( enemigo.) 1
06 ( thought.) 5 06 ( dona.) 5 06 ( remediarlo.) 9
07 ( current.) 5 07 ( corrents.) 6 07 ( mujer.) 3
08 ( hour.) 8 08 ( semblant.) 100 08 ( corriente.) 5
09 ( thought.) 4 09 ( buit.) 4 09 ( hora.) 100
10 ( them.) 4 10 ( entremig.) 6 10 ( nada.) 4
11 ( current.) 4 11 ( corrent.) 4 11 ( ellos.) 4
12 ( fathoms.) 1 12 ( braces.) 1 12 ( corriente.) 4
13 ( fathoms.) 2 13 ( blava.) 3 13 ( brazas.) 1
14 ( sardines.) 5 14 ( ferm.) 8 14 ( brazas.) 1
15 ( steel.) 3 15 ( fresca.) 1 15 ( amarrado.) 6
16 ( good-tasting.) 5 16 ( lham.) 3 16 ( frescas.) 1
17 ( attractiveness.) 100 17 ( gros.) 3 17 ( saliente.) 3
18 ( line.) 100 18 ( escaients.) 8 18 ( apetecible.) 5
19 ( depths.) 4 19 ( uixa.) 9 19 ( atraccin.) 8
20 ( rise.) 2 20 ( escaient.) 5 20 ( brazas.) 100
21 ( sortir.) 2 21 ( profundidad.) 4
22 ( momento.) 2

the translators decision to deliberately alter the degree of grammatical complexi-

ty in the text. In fact, the sentential segmentation of the text is nearly identical
in all three languages, with 20 sentences in the original English passage, 21 in
Catalan, and 22 in Spanish. Slight variations in the degree of sentence complexi-
ty from one language to another are generally due to local decisions on the
translators part. A case in point is sentence 4, with one verb in the original
English (They spoke of her as a contestant or a place or even an enemy)
(1993: 23), and three in the Catalan translation: En parlen com dun contrari, o
shi refereixen com a un simple lloc, o ns el titllen denemic (1984: 81). In
any case, because the dierences to be observed in this sense are few, we may
conclude that in both the Spanish and Catalan versions the translators general
strategy has been to preserve the same degree of grammatical complexity found
in the original text.
70 ISABEL GARCA IZQUIERDO AND JOSEP MARCO BORILLO

Interestingly, the degree of grammatical complexity observable in this

particular sample is relatively low. Of course, Hemingways style is known for
its grammatical simplicity, a fact which has been associated with the authors
experience as a journalist. Indeed it is true that journalistic writing, especially in
English, favours the use of unsophisticated syntax, yet in the case of The Old
Man and the Sea, grammatical simplicity may also be seen as (iconically)
reecting the simplicity of the central character (a sherman who has been
unlucky with the sh lately).
Ferran de Pol appears to have been aware of this, as he states, quite
explicitly, in the introduction to his Catalan translation (1984: 5657):
Hemingways literary malice his craft clearly tells him that it is at
bottom a matter of style. A means of expression must be found that is naked,
simple, devoid of cultural impedimenta. And that is what Hemingway does or,
more precisely, what he tries to do. From the very beginning, as if he wanted
both the narrative and the old mans expression to go hand in hand, he
explains the facts and makes the sherman speak in a clear, direct, most
simple fashion. Some critics have even blamed Hemingway for so much
simplicity, as it involves the risk of falling into what the French call simplesse.
Hemingway is not afraid of repetitions, of recurrently using the particle and or
the most commonplace words. Sometimes his prose takes on a hammering
tone, no doubt sought after but bordering on fatigue. [our translation]

Obviously, beyond any personal evaluations, Pol identies the feature in question
as deliberate, and assigns it a stylistic value. He therefore attempts to reect this
feature in his translation, but not without a good deal of eort, as he suggests
later on in the acknowledgements (1984: 58): I am glad to publicly show my
gratitude to my wife, Esyllt T. Lawrence, who has revised the translation and
whose advice has helped me grasp the tone of dicult simplicity of the
original [our translation]. The Spanish translation includes no preface or
introduction, so it is impossible to know to what extent the translator (Lino
Novas) was also aware of this stylistic feature and its function, and yet, his
translation decisions are very similar to those of Ferran de Pol.
Moving on now to the novel Moll Flanders by Daniel Defoe, which exhibits
a relatively high degree of grammatical complexity, we may observe that the
corresponding Catalan and Spanish translations are both unlike the translations
of Hemingway just discussed and dissimilar to each other in the solutions they
propose. Pages 7 through 9 of the source text (Defoe 1981) have been selected
for analysis, together with the corresponding passages in Catalan (Defoe
1982: 1516) and Spanish (Defoe 1996: 78). The following is a representative
fragment of the original English passage:
 GRAMMATICAL COMPLEXITY AS A TRANSLATION PROBLEM 71

Text 2
I have been told, that in one of our Neighbour Nations, whether it be in
France, or where else, I know not; they have an Order from the King, that
when any criminal is condemnd, either to Die, or to the Gallies, or to be
Transported, if they leave any Children, as such are generally unprovided for,
by the Poverty or Forfeiture of their Parents; so they are immediately taken
into the Care of the Government, and put into a Hospital calld the House of
Orphans, where they are Bred up, Cloathd, Fed, Taught, and when t to go
out, are placd out to Trades, or to Services, so as to be well able to provide
for themselves by an honest industrious Behaviour. (Defoe 1981: 78)

An analysis of the grammatical complexity of the selected passage (in its full,
two-page version in the three languages concerned) yields the results reected in
Table 2 below.

Table 2: Moll Flanders: source text (pp. 79), Catalan translation (pp. 1516), Spanish
translation (pp. 78)
English source text Catalan translation Spanish translation
Sentence Rate Sentence Rate Sentence Rate
01 ( Work.) 10 01 ( famlia.) 07 01 ( obra.) 06
02 ( Crimes.) 05 02 ( cosa.) 01 02 ( delitos.) 04
03 ( am.) 21 03 ( delictes.) 06 03 ( actualidad.) 21
04 ( Behaviour.) 23 04 ( sc.) 21 04 ( medios.) 14
05 ( Body.) 19 05 ( treballador.) 19 05 ( honrada.) 10
06 ( here.) 01 06 ( cos.) 15 06 ( suerte.) 05
07 ( Cheapside.) 09 07 ( diferent.) 01 07 ( remediarla.) 04
08 ( Account.) 06 08 ( costa.) 05 08 ( alma.) 02
09 ( sure.) 20 09 ( Cheapside.) 03 09 ( Cheapside.) 03
10 ( it.) 21 10 ( justa.) 08 10 ( verdadera.) 08
11 ( recomanables.) 20 11 ( culpable.) 05
12 ( oral.) 05 12 ( meses.) 08
13 ( dir-ho.) 21 13 ( caso.) 02
14 ( manos.) 06
15 ( certeza.) 01
16 ( odas.) 05
17 ( infancia.) 11
18 ( ms.) 13

The average rate of grammatical complexity is 13.5 in the source text, 10.15
in the Catalan translation and 7.52 in the Spanish translation. These dierences
are due to the fact that both translators have broken down the target texts into a
72 ISABEL GARCA IZQUIERDO AND JOSEP MARCO BORILLO

greater number of sentences, thus reducing their rate of complexity. Nevertheless,

the strategies of the two translators may not be considered similar by any means,
as the Catalan translator makes an eort not to depart signicantly from the
grammatical structure of the original, whereas there is no immediate evidence
that Eduardo Rico, responsible for the Spanish translation, has done the same. A
quick check of the number of relatively complex sentences in each version will
serve to demonstrate this idea: there are ve such sentences in the source text
(rates = 21, 23, 19, 20 and 21), ve in the Catalan translation as well (rates = 21, 19,
15, 20 and 21), but only three in the Spanish translation (rates = 21, 14, and 13).
It could be argued that one of the functions of grammatical complexity as
a stylistic feature is to create a sense of oral discourse, which, while obviously
presented to us in printed form, conserves some traces of spoken language. In
this respect, the association drawn by Halliday between spoken language and
grammatical complexity on the one hand, and written language and lexical
density on the other, is relevant here. Other features in the text are also reminis-
cent of oral discourse, such as the fact that the narrator addresses the reader
directly with expressions such as you may give me leave (Defoe 1981: 7) or
you may be sure (Defoe 1981: 8) or the use of colloquial markers. The
deliberately oral mode of the texts inner context of situation and its most
apparent eect, namely that of creating a sense of immediacy or reducing the
distance between narrator and reader, is amply justied by the narrators
predicament. Moll was, as the front page of the original informs the reader
(Defoe 1981: iii), Twelve Year a Whore, ve times a Wife (whereof once to her
own Brother) Twelve Year a Thief, Eight Year a Transported Felon in Virginia,
at last grew Rich, livd Honest, and died a Penitent . The narrator, if she wishes
her story to be read, must render it palatable to a readership which cannot, in
principle, be assumed to feel much sympathy for the character described. In other
words, the narrator seeks to gain the readers sympathy by using a tone of
closeness and comradeship, rather than one of distance and neutrality.
The Catalan translator of the novel was undoubtedly aware of the stylistic
prole of the source text, as his preface to the translation clearly shows (Defoe
1982: 12):
As to the language, it must be said that I have tried to retain, as far as
possible, the slapdash and sometimes muddled style of the original, and the
small dierence in style between narrative and dialogue, which is almost
negligible in the original, has been left unaltered. [our translation]

Both the fact that he refers to the style as slapdash and sometimes muddled
and the small dierence between narrative and dialogue are perfectly consistent
with the sense of orality alluded to above. Spoken language is indeed more
 GRAMMATICAL COMPLEXITY AS A TRANSLATION PROBLEM 73

slapdash than written language, for reasons linked to the amount of time the
speaker has to plan his/her utterances. Furthermore, if the narrative mode here is
deliberately oral, the distance separating the narrative proper from the dialogues
has naturally been reduced or even neutralised. Unfortunately, no direct informa-
tion regarding the Spanish translators stylistic awareness is available to us, as
the Spanish version includes no preface or presentation.

Conclusion

In light of the data gleaned from our analysis, the following (tentative) hypothe-
sis may be put forth: regardless of stylistic function, the more grammatically
complex the English source text is, the more translators tend to alter its gram-
matical structure in Spanish and Catalan. However, two caveats are in order here.
First, saying regardless of stylistic function does not mean that translators pay
no attention to stylistic function; in this sense we have seen how Miquel Desclot,
in the case of Moll Flanders, and Ferran de Pol, in the Catalan translation of The
Old Man and the Sea, were fully aware of it, as indicated in their own prefaces.
Secondly, the corpus used for this study is, needless to say, far too limited to
allow for any sort of denitive conclusion. If our hypothesis is to be elevated to
the rank of a law (in Tourys sense, e.g. 1995) or a general tendency, it must
rst be validated by studies of other translations, both of the same texts consid-
ered here and of other, alternate source texts.
Should our hypothesis prove to be valid, the tendency to simplify the
grammatical structure of complex source texts may be seen as totally consistent
with a more general tendency noted by Baker (1992: 212), namely that translators
often choose to reduce the processing eort required of the reader by making
explicit what the original left implicit, or by simplifying what was complex in
the source text, as in the present case. Again, this tendency must be understood
only in terms of statistical validity, for it will not always match actual translation
practice, with all its richness of detail and particularity.
In sum, if one of the ultimate goals of Descriptive Translation Studies, as
Toury (1995) claims, is to set up general laws of translation behaviour, such laws
must be well grounded in a careful study of the facts, in this case, of source texts
and their corresponding translations. It has been the aim of this article to draw
attention to a tendency observed, but the need for further study in this regard can
hardly be overstressed.
</TARGET "izq">

74 ISABEL GARCA IZQUIERDO AND JOSEP MARCO BORILLO

References

Defoe, D. 1981. Moll Flanders. Oxford: Oxford University Press.

Defoe, D. 1982. Moll Flanders. Barcelona: Edicions 62.
Defoe, D. 1996. Moll Flanders. Barcelona: Thasslia.
Hemingway, E. 1984. El vell i la mar. Barcelona: Proa.
Hemingway, E. 1993. The Old Man and the Sea. London: Arrow Books.
Hemingway, E. 1997. El viejo y el mar. Barcelona: Planeta.
<TARGET "p2"
</TARGET "p2">DOCINFO

AUTHOR ""

TITLE "Investigating the Translation Process"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

S II

Investigating the Translation Process

<TARGET "gil" DOCINFO

AUTHOR "Daniel Gile"

TITLE "Opportunities in Conference Interpreting Research"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 8

Opportunities in Conference Interpreting Research

Daniel Gile
Universit Lumire Lyon 2

1. Introduction: Conference interpreting research, an underdeveloped

field

As is often recalled in the literature, conference interpreting research (CIR) is

less than 40 years old. It is practiced by a very small community of active
researchers, at most a few dozen worldwide (see Pchhacker 1995). CIR has no
infrastructural basis such as research centers of its own or academic interpreting
departments; interpreting is taught mostly in translation and interpretation
schools, which are generally considered professional training courses rather than
academic institutions, hence a marginal role for research though this is
slowly changing. Neither does it benet from research funding, and understand-
ably, it has little research work to boast of.
During its short history, its production has nevertheless evolved from less
than 20 scholarly texts a year in the sixties to about 100 to 150 texts in the early
nineties, and has stabilized at this level. Out of these texts, a few dozen at most
each year, mostly graduation theses by student interpreters, report empirical
studies (see Gambier et al. 1997, The IR(TI)N Bulletin issues 1 to 20, and Gran
and Viezzi 1995). Since their authors are raw beginners in research, who are
interested in becoming professional interpreters rather than academics, and since
most of their supervisors are either strangers to conference interpreting or
interpreters with no research background themselves (with some exceptions to
this rule, for instance in Trieste see Gran and Taylor 1990; Gran and Viezzi
1995; Fabbro and Gran 1997), many of them suer from rather severe method-
ological weaknesses. In research on written translation (the term Translation
Studies or TS will be used here), the situation is better. There is a larger
number of scholarly studies, including many by trained and experienced scholars,
in particular those coming from the literary disciplines and conducted in universi-
78 DANIEL GILE

ties in numerous countries in the relevant departments of literature, modern

languages, etc.
An important point is that this state of underdevelopment of CIR is not
essentially due to nancial problems: while some work in the neurophysiological
and cognitive paradigms requires laboratory equipment, much ground can be
explored at virtually no cost, and almost all of the research done so far was done
without any institutional funding. Neither is the present underdeveloped state of
CIR due to a wall of formidable methodological obstacles. This paper argues that
though serious obstacles of this kind exist, especially in the cognitive paradigm,
much headway can still be made without hitting them.
At this point, general weakness as regards motivation and research training
is the main factor which holds back CIR development. Taking the glass-half-full
mentality, one can argue that there are very good opportunities for those
individuals and institutions which have the necessary motivation and receive or
provide (respectively) proper training. This paper discusses the main issues
involved and oers some suggestions.

2. Phenomenological and social limitations and constraints

Before presenting the opportunities and options, it may be appropriate to discuss

briey the main objective hurdles and some social obstacles that stand in the
way of CIR:

2.1 The complexity of interpreting

Linguistics, psycholinguistics and neurolinguistics have been struggling with the

complexity of speech production and speech perception for decades. A look at
textbooks in these disciplines should be enough to convince readers that after
thousands of experiments and clinical observations, the phenomena involved still
puzzle the researchers, inter alia because of the intricacy of the processes and
interactions between the relevant variables. In interpreting, the picture is even
more complex:
There is simultaneous processing of two languages, one in reception mode,
and one in production mode, meaning that not only are the subjects bilin-
gual with possibly specic cerebral organizational patterns for languages
(Fabbro and Gran 1995: 12), but interpreting must involve some language-
switching mechanisms that are not necessary in monolinguistic activity.
As long as translation was viewed as a translinguistic unit-by-unit trans-
 OPPORTUNITIES IN CONFERENCE INTERPRETING RESEARCH 79

coding operation, that is, as long as it was investigated by linguists only, it

was assumed that contrastive linguistics could account for translational
phenomena to a large extent. When more and more practicing translators
and translation teachers made their voices heard, it became clear that human
translation is a strategy-oriented activity, with strategies aiming not only at
producing correct target-language structures, but also at maximizing
communication eciency, minimizing time and cost, fullling social
requirements, satisfying psychological and sociological requirements (Toury
1995) etc. The investigation of translational behavior therefore also has to
take into account these variables.
In interpreting, both simultaneous and consecutive, the on-line, immediate
nature of the processes imposes a severe mental-load constraint (Gile
1995a,b), which also has to be taken on board if interpreting behavior is to
be accounted for.

2.2 Interpreting diversity

Interpreting phenomena are highly diversied, because they depend on many

factors, each of which can take on many values: source language/target language
pairs, speech content, source-speech delivery parameters, the interpreters
experience, cognitive skills, knowledge of the languages involved, state of mind,
previous knowledge of the eld, motivation, training, ethics, to quote just a few.

2.3 Dicult access to data

To scientic investigators, studying complex and diversied phenomena implies

requirements for data from large samples (Shipman 1988: 59) and/or much
replication work, since without a large enough sample of data points, it is
dicult to make reliable inferences on patterns, trends and interactions between
variables. Conference interpreting oers limited opportunities for data collection,
due mainly to three factors:
The total number of conference interpreters worldwide is very small, around
a few thousand, with only a few hundred interpreters in the largest geo-
graphical centers such as Brussels, Tokyo or Paris.
Many interpreted meetings are condential.
Interpreters are generally reluctant to have their work recorded and studied
thoroughly, and may not be available or willing to volunteer as subjects for
CIR studies. Like translators (among numerous references, see Komissarov
1985: 208, Baker 1992: 3, or Danaher 1992: 15, who claims he has never
80 DANIEL GILE

found anything in theory which has been the least use in [his] practice, or
even had relevance whatever to [his] work) , most of them are indierent
at best, and even hostile to research work which they perceive as useless,
and often dangerous (see Stenzl 1983; Shlesinger 1989; Gile 1995b).

3. Opportunities

As mentioned above, the paucity of CIR so far means that much ground remains
to explore. Moreover, the diversity of situations means there is still much
potential added value in descriptive work to document it (see Gile 1999 on
variability in delity perception, and see the recent doctoral thesis by Lamberger-
Felber (1998) for data on variability in several interpreting parameters). Method-
ologically speaking, scientic exploration generally starts with gross observa-
tions, theories and tests, and gradually goes into ner and ner observations with
more sophisticated tools and methods which are developed as a function of
ndings, of obstacles encountered, of new questions and of newly developed
tools (see Kourgano 1958). The scarcity of descriptive work in CIR so far
suggests that there is still much to discover with simple methods.

3.1 CIR opportunities with simple methods

The following are a few examples of CIR which require no highly sophisticated
tools and methods. This does not mean that research is therefore easy and can be
conducted carelessly. Even in the simplest research techniques, there are many
pitfalls associated with the lack of careful planning, inappropriate sampling
methods, loose inferences, excessive reliance on the face value of utterances by
interviewees, etc. (see for example Gile 1995d), but technically speaking, no
complicated tools and methods need to be used for studies such as suggested here.

a. Quality perception
A solid comprehension of quality-perception components which determine the
response of various actors directly and indirectly involved in the interpreting
process, including the speaking interpreters, non-active interpreters in the booth,
employers and end-users of interpreting are an essential foundation for training
and for professional rules of behavior. And yet, traditionally, both in interpreting
and in translation, the literature has been presenting prescriptive views, but very
little descriptive and analytical research data. The situation is slowly changing,
but the amount of work to be done at a very fundamental level is still consider-
 OPPORTUNITIES IN CONFERENCE INTERPRETING RESEARCH 81

able (see for instance Kurz 1996). At this stage, general questionnaires and eld
interviews are still required to collect raw data, though a few studies (Gile
1995c; Collados Ais 1996; Gile 1999) have already started to zero in on specic
variables.

b. Language performance
Another relevant aspect of interpreting performance is the linguistic quality of
the target speech. In particular, naturalistic error analysis of linguistic output in
both A and B languages in students and professionals is likely to reveal frequent
language-combination-specic weaknesses (typical errors of native French
speakers in English, of native Spanish speakers in German, etc.) and to suggest
remedial action through appropriate exercises. At the simplest level of investiga-
tion, such research involves listening to interpreted speeches, identifying
linguistic weaknesses, measuring their relative frequencies in specic language
combinations and classifying them appropriately. So far, no studies devoted to
the exploration of linguistic weaknesses in interpreting are known in mainstream
interpreting literature, except for Kopczynski 1980 and Gile 1987, both of which
only deal with student interpreters. Part of the problem is due to historic reasons:
throughout the seventies and up to the mid-eighties, the prevailing paradigm in
the interpreting community forcefully claimed that by denition, professional
interpreters had achieved thorough mastery of their working languages, and that
any investigation of linguistic issues in the eld of interpreting was out-of-place,
as it could not pertain to competent interpreters (see the analysis in Gile 1995a).
Attitudes have changed, and there is ample room for simple descriptive studies
of the interpreters linguistic output.

c. Interpreting strategies
In recent years, the concept of strategies in translatorial behaviour has become
popular. The reference here is to both deliberate problem-solving strategies and
to unconscious behavioral patterns before and during the translation act (prepara-
tion for a conference, behavior when encountering diculties, etc.). An increas-
ing number of texts in the literature have been using this term, but a much
smaller number of studies have attempted to describe such strategies through
observation and interviews (see however Ahrens 1994; Gile 1995a,b; Kalina 1996).

d. Lexicometric properties of source speeches

The ocial requirements from AIIC, the International Association of Conference
Interpreters, call for a near-native prociency of its members in linguistic skills
across the board. It is clear, however, that some sociolects are frequently
82 DANIEL GILE

encountered in conferences, and others rarely if ever, depending on the specic

market. It therefore makes sense to study linguistically the types of speeches
most frequently encountered so as to help interpreters strengthen their skills in
their most relevant components as rapidly as possible. Such descriptive studies
on the syntactic and stylistic aspects of the speeches require some knowledge of
linguistics. Lexicological exploration, however, can be conducted even by
beginners, and the exploration of rather large corpora is now possible using
modern computer technology.

3.2 CIR opportunities with more advanced methods

The following are a few examples of descriptive studies which can be conducted
with a rather modest set of skills and tools:

a. Sociological studies of the interpreting environment

There have been a few studies on job satisfaction among conference interpreters
(Kurz 1983; Rojas 1987). Many other questions could be of interest to sociolo-
gists and ethnologists, and could be taken up by young investigators through
observation, interviews and questionnaires, provided they have some knowledge
about existing theories and are supported by some methodological guidance from
experts (such guidance is necessary because of the numerous traps associated
with what may appear as a straightforward set of methodological tools at rst
glance see for example Frankfort-Nachmias and Nachmias 1992; Shipman 1988).

b. Language performance in interpreters: syntactic, phonological and text-

linguistic aspects.
Here, the analysis of language performance refers to aspects which require some
solid knowledge of linguistic concepts. Good examples in this category are two
studies by Shlesinger. In the rst (Shlesinger 1989), she used the features of
written and oral speeches as derived from linguistics to show that oral-like
speeches tend to be interpreted in a more written-like manner and vice-versa.
In the second (Shlesinger 1995), she used tools from text linguistics to study
changes in cohesion that occur when speeches are interpreted.
In these two examples, the method consisted in comparing original and
interpreted speeches and in counting and analyzing occurrences of specic
features. The technical part consisted in identifying the relevant features in the
literature, but there were no complicated experimental set-ups, no particular tools
to develop, no severe methodological diculties.
 OPPORTUNITIES IN CONFERENCE INTERPRETING RESEARCH 83

3.3 CIR opportunities with advanced methods

The most frequently found studies in this category are interdisciplinary and deal
with testing and/or implementing linguistic, psychological and neurophysiological
theories and methods. Besides the studies in the sixties and seventies in the
cognitive psychology paradigm reported regularly in recent dissertations (see also
Lambert and Moser-Mercer 1994), a few dozen examples can be found in Gran
and Taylor 1990 and Tommola 1995. Dillinger 1989 is a typical example of the
complexity of such a study, in which even the speeches to be interpreted were
written specically for the purpose of complying with specic linguistic require-
ments. Opportunities with advanced methods are numerous, but they require
either a solid foundation in the relevant discipline, or at least the availability of
an expert for team work.

4. Risks for beginners

Before embarking upon new research projects, beginners would gain from careful
planning taking into account the following diculties, for fear of nding
themselves stuck for months or years anywhere along the path and eventually
giving up, as has happened dozens of times in past years:

4.1 Dicult access to subjects

This problem, explained in Section 2.3, implies that those studies which require
a large number of subjects, for instance experimental studies under more than
two conditions, may simply not be feasible, depending on the local environment.
One easy way out is to use students: this is legitimate for studies of the students
environment (for instance in research on training and on the evolution of the
students skills see Kurz 1996), but questionable in research on interpreting
per se (see Gile 1995a).

4.2 Advanced knowledge and know-how requirements

Experienced members of the CIR community often receive subject proposals

from beginners who wish to conduct a study on cognitive or neurophysiological
aspects of interpreting. It is important to understand that reading and following
the rationale in a paper reporting such a study as carried out by an expert is one
thing; planning and implementing one without the years of study and specialized
84 DANIEL GILE

guidance that experts in the relevant discipline benet from is quite another.
Depending on whether such expert supervisors are available to guide the
beginner, the project may not be feasible at all, or be feasible, but at the expense
of disproportionate eorts. There is also a risk that theories and methods from
cognate disciplines may be misperceived due to insucient contact with the
eld. Alan Baddeley, whose model of working memory (see for example
Baddeley and Hitch 1974) has developed into a central concept and investigation
area in cognitive psychology and is also used extensively in CIR theorizing,
pointed out in a private conversation that a statement on memory in note-taking
in consecutive interpreting in a paper by one (well known) CIR researcher was
based on an incorrect interpretation of the model.

4.3 Dicult access to the literature and to experts

In some universities, young interpreting scholars may have access to experts in

the cognitive elds, in linguistics, in sociology etc., and through them to a vast
body of relevant literature. In many other centers, access to the literature is
extremely dicult, be it for economic reasons (in many countries, the budget of
libraries may not allow the purchase of enough relevant publications, and
librarians may consider that literature from adjacent disciplines is not relevant in
an institution which essentially trains professional interpreters), for institutional
reasons having to do with the contacts (or lack thereof) between the interpreting
institution and the libraries of the cognate departments, or for linguistic reasons.
Even when students do have access to a good library, it may be dicult for
them to nd their way into the relevant publications without guidance from an
expert.

4.4 Direct importing of theories and methods from cognate disciplines

Throughout the seventies, the prevailing paradigm called for research on

interpreting by interpreters only, and those voices which called for interdisci-
plinarity were not heard (see Gile 1995a). The prevailing attitude has changed
and interdisciplinarity has become fashionable. However, theories and methods
evolve within certain disciplines and certain contexts, and cannot necessarily be
imported directly into other disciplines and contexts. For instance, experiments
using inferential statistics can be very useful when samples are large enough,
when appropriate sampling procedures are feasible, and when enough is known
about variable distribution from previous studies to choose the appropriate
statistical tests. In interpreting, these conditions are generally not met, and such
 OPPORTUNITIES IN CONFERENCE INTERPRETING RESEARCH 85

experiments are not always the best investigation method in the CIR environ-
ment. Yet another point is that interpretation is highly strategy-oriented (as is
translation). Analyzing it in the sole framework of linguistic or cognitive theories
amounts to neglecting relevant and sometimes very important variables. A case
in point is propositional analysis, which is being explored by Tommola in
particular (see for example Tommola & Lakso 1997). The idea is to break-up the
source speech and the target speech into small meaning units (propositions) and
to count the proportion of such units which correspond in the source and target
texts, vs. those which do not, presumably indicating a problem in the inter-
preters performance. While the metric is appropriate much of the time, some-
times, the interpreter feels that a better interpretation would result from some
additions and omissions in the target speech (see for example Jones 1998), but
does not nd the necessary cognitive resources to make the changes. In such
cases, a target-text formulation which mirrors the corresponding source-text
proposition may be a sign of diculty rather than a sign of optimal interpreting.
Propositional analysis may therefore produce misleading results.

4.5 Developing new theories

Young scholars are often attracted by the idea of developing a new theory which
will explain a central phenomenon in translation or interpretation. The fact that
existing theories contain visible gaps, lack supporting evidence or do not
incorporate all state-of-the-art developments in the relevant disciplines (especially
in the cognitive sciences) tends to encourage them to go ahead in that endeavor.
Unfortunately, most of the time, not only will it take the young scholar too much
time and eort to read and understand all the literature forming the basis for
existing theories, but coming up with a stronger, new theory is a dicult task,
requiring long-winded work at PhD level and beyond, to appreciate the
amount of work and diculty involved in coming up with an original theoretical
contribution, see for example Setton (1997).
In many cases, students who are unaware of these diculties start with an
overly ambitious project and give up after several months or years without
having ever taken o, or else waste much time trying to achieve the initial
research objectives and change direction at the last moment, when they realize
these objectives are beyond their reach, ending up with an unsatisfactory result.
86 DANIEL GILE

5. Suggestions: Research policy

As long as CIR is only an individual option for interpreters and interpretation

instructors, who have no obligation to conduct research, most of them are likely
to continue giving priority to competing activities, such as professional interpret-
ing and teaching, which are more lucrative. This will do nothing to improve the
situation. A close look at centers which have been rather productive in the past
(see Gile 1995a; Pchhacker 1995) suggests that relying on the charismatic
personality of one or two leaders is not enough to ensure their vitality in the long
term, probably because in the absence of institutional incentives, dedicated
individuals with enough leadership to take over when a generation of leaders
leaves the scene are too rare. Institutionalizing research by making it an integral
part of the interpreting schools activity, with implications on the nature of
graduation theses and on research requirements from the teaching sta, is bound
to improve the situation. A more institutionalized framework for research could
also foster replication, methodological discussion and teamwork beyond what is
likely to occur in a purely inter-individual context. Institutional action could
focus on a number of policy components:

5.1 Research training

As pointed out earlier, many young student-interpreters who must write a thesis
for graduation have had no training in research methods and work with supervi-
sors who are in the same situation. They often proceed by imitating authors of
previous theses and other publications, just as one would prepare a contract in
business after having read other contracts but without having studied law. Some
of the results are surprisingly good in spite of this handicap, but in other cases,
the texts appear acceptable at rst sight, since the appropriate structure of the
thesis or paper, the language and the bibliographical references comply with the
standards, but a careful reader will soon encounter methodological weaknesses,
some of them very serious, beneath the surface. In particular, previous ndings
are not taken into account, the methods used are not described explicitly enough
for readers to be able to assess the results, and there are many problems with the
rationale (see Gile 1995d). Such weaknesses reect not a lack of intelligence or
dedication, but a lack of training, and it is important that authors receive some
training before they start and solid guidance throughout their work.
Such basic training, which is most easily organized by academic institutions
(in this respect it is encouraging to see the doctoral courses being set up in
Spanish universities), could include input on the nature of scientic investigation
 OPPORTUNITIES IN CONFERENCE INTERPRETING RESEARCH 87

especially as regards its strict requirements in terms of cartesian logic,

exercises in critical thinking and exercises in critical reading of research texts.
Teaching specic theories may be less fundamental during this initial introduc-
tion to research, though something about specic theories can be taught indirect-
ly, when these are used as material for exercises in critical thinking and critical
reading.

5.2 Replication

Replication is a critical part of any empirical science. Multiplying naturalistic

observations of similar phenomena and conducting repeated tests on a specic
issue is a matter of routine in most disciplines and makes it possible not only to
corroborate or detect problems with previous results, but also to ne-tune
hypotheses and to formulate new hypotheses, for instance if a small variation in
the experimental set-up or the observational environment leads systematically to
dierent results. Typically, replication is part of research work done on a regular
basis by professional research teams, i.e. by people whose regular job it is to
conduct research. Academics holding teaching jobs do much less replication
work, and M.A. and PhD candidates virtually none, since their thesis or disserta-
tion must meet the originality criterion. In CIR, there is no professional
researcher (save, until a short while ago, Franco Fabbro of Trieste, in the
neurophysiological paradigm) and no research team, and almost all authors are
also active conference interpreters who share their time between teaching, some
research and professional conference interpreting. It is therefore not surprising
that so little replication is done spontaneously.
Whenever possible, institutions should promote replication. One way would
be to have research students replicate systematically simple experiments and
observations as part of their training. Another would be to accept replication
studies as graduation theses in schools which require such theses from interpret-
ing students. A good replication operation with the necessary adaptation and
improvements over the original study requires good comprehension of the latter,
makes it possible for students to save much time in nding a subject and
provides them with an alternative to preparing a terminological or other type of
thesis they are not interested in. It is thus protable for them and for the CIR
community as a whole. It goes without saying that those students interested in
and capable of conducting an original study should be allowed to do so, but from
experience, there are few such students.
88 DANIEL GILE

5.3 Giving methodology high priority

In many well-established disciplines, there are methodological traditions which

new researchers acquire both during formal training courses and when working
with their elders in research teams ( apprenticeship is the almost invariable
rule a young hopeful attaches himself as a graduate student to some senior
scientist and hopes to learn his trade Medawar 1981: 12). In CIR, there is
no such tradition, and there are few masters who could guide beginners.
Moreover, the problem is compounded by the interdisciplinary character of the
work done, as explained in Section 4.4.
So far, the CIR communitys interest has focused on theories and models,
and has given relatively little attention to actual empirical studies, and even less
to methodological issues. Institutions, especially translation and interpretation
schools, could help improve the situation by assigning higher priority to method-
ological discussion and development in their regular publications and meetings,
thus heightening the communitys awareness of such issues and improving the
methodological quality of CIR. In this respect, it is encouraging to note a plenary
by Miriam Shlesinger in a recent conference (Anovadores de nos, anosadores de
vos, organized in Vigo in March 1998), which was devoted to a methodological
problem encountered in an experimental study.

5.4 Inter-center cooperation

Competition between schools has been a constant feature of the interpretation

training scene from its very inception, partly because of the elitist ambitions of
some leaders, and partly because of the relatively large number of such schools
for a modest market. At this time, there does not seem to be similar competition
as regards research. From the mid seventies to the mid-eighties, ESIT in Paris
was by far the most productive center, but it has now virtually disappeared from
the interpreting research scene (see Gile 1995a). In the early nineties, the
SSLMIT at Trieste produced a far larger number of studies than all other centers
(see Pchhacker 1995; Gran and Viezzi 1995), but maintained a non-competitive
open attitude with other schools and individual researchers. While the situation
may change in the future, for the time being, cooperating with other schools is
likely to be protable to all insofar as it will help achieve the critical mass of
research without which a discipline will not be acknowledged by the scientic
establishment. Such cooperation would be particularly welcome in operations
such as setting up joint research training programmes, especially for institutions
which are geographically close to each other, as in Finland, in Spain and in
Belgium.
</TARGET "gil">

OPPORTUNITIES IN CONFERENCE INTERPRETING RESEARCH 89

6. Conclusion

The conclusions of this analysis can be summed up under four points:

a. Opportunities such as described in this paper seem real enough and could
attract researchers if these are motivated enough and if acceptability criteria for
publications, theses and dissertations are made stricter. As long as publications
are too easily accepted, as is the case now, there is little prospect for change, and
individuals can move up from raw beginner to distinguished professor without
having carried out any proper research.
b. While there is no ruling out a spontaneous evolution arising from inter-
individual competition or the arrival onto the CIR scene of a sucient number
of well-trained researchers from other disciplines, such a trend has yet to
materialize and it is not clear why and how this should happen. Hence the
importance of an external, institutional guiding hand which translation and
interpreting training institutions are probably in the best position to provide.
c. Because of the lack of empirical research so far, there is still much scope for
simple methods which will lead to signicant discoveries, especially if replica-
tion and inter-center cooperation are fostered.
d. It is important that individual researchers be made aware of the importance of
reliable albeit simple methods at this stage of the development of CIR. The use
of such reliable methods should be encouraged institutionally. Aiming too high,
in particular by importing indiscriminately models and methods from cognate
disciplines, may jeopardize the potential added value of research endeavors.
Against this background, it appears that at this point, even single institutional
units, such as academic translation and interpreting departments, can have an
important role to play in the development of CIR and become leaders in the eld
very rapidly at relatively little expense in terms of funding and labor, provided
the legal and regulatory framework allow it and there is the appropriate political
will in the institution. Spain has been very active over the past few years in the
eld of translation studies, with new doctoral courses, completed PhDs on and
around conference interpreting (Padilla 1995; Collados Ais 1997), conferences
and seminars. Will Spain show the way and lead at the turn of the next century,
just as ESIT in Paris did in the early seventies and Triestes SSLMIT in the late
eighties?
<TARGET "neu2" DOCINFO

AUTHOR "Wilhelm Neunzig"

TITLE "The Computer in Empirical Studies for the Didactics of Translation"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 9

The Computer in Empirical Studies for

the Didactics of Translation

Wilhelm Neunzig
Universitat Autnoma de Barcelona

Introduction

In order to obtain data or knowledge concerning a physiological, psychological

or social phenomenon, science uses methods for data collection and scientic
research which allow certain objective truths to be sought and found. The
methodology applied for this purpose may begin with non-standard methods
as is very often the case in the eld of translation studies such as introspec-
tion, hermeneutics, qualitative analysis, etc., as well as from standard methods
which, from the point of view of the positivist theory of science, appear more
scientic. In translation studies, we see the application of such methods as the
questionnaire, the interview, the test and the experiment, the latter being the one
which concerns us here.

On the use of the experiment in translation research

In an experiment, the main aim of which is to investigate the causes of a

phenomenon, certain conditions are articially manipulated in order to observe
the eects; that is to say, the relationship between the variables is identied: the
inuence of a variable (the dependent variable, which can be manipulated, for
example the pedagogical input) on another variable (the independent variable,
which can be measured, for example the result of a translation).
Scientic theory has dened certain criteria which must be met by experi-
mental design in order to ensure the validity of the results. These are: objectivity,
reliability, replicability, validity, extrapolability, quantiability and ecological or
environmental validity (we prefer to use the term environmental validity to
92 WILHELM NEUNZIG

refer to the criteria that the experiment should reect a normal situation, as
unarticial as possible; the denition of ecological validity is still ambiguous).
In addition, the experimental design should include, in our opinion, certain
criteria which are derived directly from the experimental context in which
experiments in the didactics of translation are normally carried out and in which
at least some of the subjects are students of the teacher/researcher. Those criteria
are fairness (teaching ethics requires that all the students should have exactly the
same opportunity to learn the contents and/or skills being taught), and experi-
mental practicability and economy (the experiment must be conducted in normal
classroom situations so that students are prevented from spending too long on the
experiment; otherwise, their motivation would be aected and the results
distorted.
Regarding experimental design, translation studies research most often
conceives eld experiments, in which measurements are taken in a natural
environment to determine the inuence of a given variable, but which pose the
considerable problem of making it dicult or even impossible to control many
extraneous variables. Their great advantage is that the subjects act in a natural
context, their reactions therefore being spontaneous, a fact which has a positive
inuence on the environmental validity of the experiments.
Recently, there has been a growing tendency to design laboratory experi-
ments, in which experimental conditions are controlled and which allow extrane-
ous variables to be eliminated or systematically controlled; moreover, they permit
more exact measurements. The major problem with this kind of experiment is its
environmental validity; in other words, the articial nature of the situation in
which the data are obtained.

On experimental design and the measuring instruments

Of crucial importance when it comes to designing an experiment are the

appropriate measuring instruments which will ensure its validity; that is to say,
which will measure what the experiment is really designed to measure. At this
point, we shall take a brief critical look at the best-known experimental instru-
ments in the eld of translation studies research, and particularly at those used
in the didactics of translation: the test, the interview and the famous TAPs or
think aloud protocols and, more recently, psychological and physiological tests.
The test method, involving the analysis of translations or translation skills
tests carried out under given conditions, which is the most traditional instru-
ment in translation studies research, would appear to be suitable for use in eld
 THE DIDACTICS OF TRANSLATION 93

studies in order to obtain a large volume of data on translation activity (for

example, in order to gather data on the translation errors frequently made by a
given group of subjects and to establish pedagogical goals). The test method also
has its place in eld and laboratory experiments designed to collect data
concerning the product (the translated text). Within the framework of the
didactics of translation, for example, there are experiments which use translation
tests to measure the inuence of a certain pedagogical input (the translation
class) in modifying translator practice and, by extension, the acquisition of
translator competence. In my opinion, however, such experiments pose problems
of experimental design: apart from the question of the legitimacy of drawing
conclusions about processes (such as the learning of certain translation tech-
niques or strategies) on the basis of a product the translation before and after
the pedagogical input this kind of design raises problems of objectivity,
replicability and reliability: it is dicult to standardize the teacher/researchers
input and thereby ensure that the results obtained by the experiment are compara-
ble to those obtained with other teachers and students. Moreover, it is unlikely
that the extraneous variables which distort the results during the experiment
could be controlled, in addition to the well-known problems attached to assessing
student translations.
To compensate for these diculties, exploratory interviews are included in
the experimental design in order to discover in a systematic way what the subject
thought during the translation process. These interviews have the advantage of
not exerting a direct inuence at the time when the data are collected (as the
translation is carried out), but they too pose serious problems of validity and
objectivity, since these ex post facto methods may measure something that the
experiment does not aim to measure, such as the students memory or ability to
adapt to the teachers expectations.
The think-aloud method (TAPs), which have become so fashionable and
which involves allowing the subject to translate aloud and recording on video
or on audio-cassette what he or she says, is appropriate to inductive research in
clarifying matters relating to the translation process: large quantities of data are
obtained concerning inuence factors, underlying translation strategies, decision-
taking, possible regularities in tackling a problem, etc., which help us to formu-
late or rene hypotheses, as has been demonstrated in works by Krings,
Lrscher, Kussmaul, Tirkkonen-Condit, Jskelinen and others. TAPs are, in my
opinion, particularly useful as a pedagogical method, since they provide a kind
of auto-feedback to correct the erroneous strategies adopted by a student.
However, as a method for validating hypotheses they pose serious problems of
environmental validity and extrapolability. Although, according to Krings (1986),
94 WILHELM NEUNZIG

there is a special anity between translating aloud and thinking aloud, given
that the verbal expression of everything (thoughts, previous knowledge, memo-
ries, procedures and the search for solutions) that is entailed in the nal written
product depends on cognitive processes, thus validating the result of the analysis
of the protocols, the method is clearly articial: how often does a translator
actually translate before a video recorder while explaining what is going through
his or her mind? However, what I consider to be the most convincing argument
is to be found in Hansen et al. (1998: 62): One of the problems in relation to
TAPs is whether it is possible to engage in two complicated actions of a similar
nature (namely translating and thinking aloud) simultaneously, and whether one
inuences the other. Having to think aloud during the translation process may
change the process, which obviously aects the quality of the data. (For a
consideration of the advantages and disadvantages of using TAPs, see Schmid 1994).
In our opinion, another instrument is needed in order to ensure, at least from
a theoretical point of view, the objectivity and environmental validity of the
experiment, since these criteria are central to empirical research in the eld of
translation studies in general, and the didactics of translation in particular.

The computer as an instrument of research in translation studies

It is an accepted fact that the computer is a means that can assist the teacher in
developing aspects of the competencies that dene the translator, such as
linguistic, documentational, cultural, terminological competence, etc., but of
immediate concern to us in the present study is its application in supporting
research in translation studies, particularly in the eld of the didactics of
translation. In my opinion, the characteristics of this medium make it more
suitable than other instruments (only in certain experimental designs, of course)
for the following reasons:
1. Translating directly onto the computer is common practice among professional
translators and students. That is to say, a student using a computer, for whatever
reason, in the computer room at the Faculty is not in an articial situation. This
ensures, at least in part, environmental validity.
2. Students are used to working with self-teaching computer programmes in their
free time and accept this medium as a virtual teacher, as has been conrmed
in various studies (see Ensinger 1997). Learning by means of computer-assisted
teaching programmes can be considered a normal pedagogical situation among
students these days, a fact which further conrms the environmental validity of
the medium.
 THE DIDACTICS OF TRANSLATION 95

3. In didactics of translation experiments, there is usually a particular interest in

the inuence of pedagogical input, and in this aspect the computer makes it
possible to standardise and manipulate this independent variable. This ability to
standardise the variable has a positive repercussion on the internal validity and
reliability of the experiment.
4. The response of an appropriately programmed computer to a given action by
the subject will remain constant, thus contributing to ensure experimental
objectivity and replicability.
5. The computer is increasingly useful in providing access to computerised
documentation systems, dictionaries, encyclopaedias, parallel texts, etc., as well
as recording the time taken to perform a translation, thus allowing the inuence
of all the extraneous variables which may distort the results of a eld experiment
to be controlled; the results obtained will therefore be relatively reliable.
6. The computer can be programmed in such a way that, unseen by the subject,
all the steps taken by the translator are recorded in a separate le, including
corrections, consultation of documents, modications, etc. Thus, data are
obtained in a completely natural environment, so supporting the internal as well
as the environmental validity of the experiment.

The computer in experimental practice

To highlight the importance of the computer as a tool for translation research in

general, I will now describe some of its main applications in the areas of
recording or registering data and as a means of standardizing information.

A record of the written translation process

Until some years ago, it was common practice for translators to rst do a
handwritten draft of their translation, which was only typed after a process of
revision. In the case of a text intended for publication, the translator (especially
the literary translator) latter received the galley proofs which again required nal
checking before denitive publication of the text. Study of all these versions
would have enabled researchers to observe the modications eected and thus
gain indirect access to the translation process, surprisingly however, as far as I
can tell, translation studies has never paid much attention to this potential source
of data. Nowadays however, translators tend to work directly onto the computer
and corrections and modications made are not recorded, and so we have no
96 WILHELM NEUNZIG

direct evidence of the actual translation process.

What I propose is that the entire translation process should be recorded so
as to enable later study of each individual step of the process. Systematic
collections of the translation process data, which we could term Translation
Protocols (TP) would provide us with simple access to the part of the translation
process, which we presently try to accede to by means of TAPs, with all the
validity problems which their use in translation studies involves. Use of Transla-
tion Protocols would be much simpler and more exhaustive than the study of
galley proofs and would also be free of distortion since the data would be
collected in the natural work setting of the translator.
The Handelshojskolen University of Copenhagen has developed a program
Translog which permits the process to be recorded letter by letter, and
incorporates a number of other features of great interest, however, I see it as
rather unrealistic since it cannot disguise the fact that the translation is
embedded within an experiment, and the programme is written in MS-DOS, and
thus does not provide access to or record on-line Internet or CD-Rom help
programmes. To overcome these deciencies we are presently working on the
Proxy programme as part of the PACTE research project. Proxy does allow us
to operate in the natural working environment of the translator (Word, WordPer-
fect), thereby ensuring ecological or situational validity, since the subjects need
not even know that their work is part of an experiment, and it also allows for
recording of all steps and use of on-line resources. In addition, these records, like
TAPs, have teaching applications: they can be consulted by students with a view
to learning from their mistakes or imitating procedures which lead to satisfactory
solutions.

A record of the documentation process

An eective documentation strategy is without any doubt an integral element of

translator competence and this area has been one of the main concerns of
empirical research in translation studies. A step by step record of the translators
use of on-line documentation sources could determine whether there are specic
procedures which give rise to better results.

On-line feedback from the teacher

The author is currently carrying out an experiment which aims to determine

student response to ongoing feedback received from the teacher concerning their
translation. The computer (Proxy programme) is the only means capable of
 THE DIDACTICS OF TRANSLATION 97

analyzing what the student does immediately after receiving the teachers
feedback. When the subject pauses (for example, a full-stop), the teacher, who is
located in an adjoining room and whose computer is connected to the students,
sends feedback in the form of a comment on the translation, however this
feedback appears to be generated by the computer itself.
This experiment is part of a line of research being pursued by the author for
a number of years and represents a modication of the approach taken in a wider
study the results of which are available in Neunzig (1997a, 1997b, 1998). An
exercise was designed for computer simulation of a translation class involving a
text in Spanish which was to be translated into German and which included two
feedback messages for the same foreseen student input: in one exercise the
feedback is designed to inuence the translation process and we could term it
cognitive feedback and, in the other, the responses provide information on the
product, through what we could term punishment /reward feedback. The experi-
ment aimed to determine the dierences between the reactions of the students to
each of the two types of feedback.
This on-line feedback can also be used by the teacher for individual and
interactive teaching in the computer room, since he or she can access each
computer individually and provide a response to the translation of a single
student, in a way similar to what occurs in the interpreting room.

Standardization of teaching feedback

In the experiment described above, the computer helps to ensure objectivity since
it guarantees that all subjects receive the same feedback for the same translation,
which, I believe, cannot be guaranteed when using human feedback, no matter
how great an eort is made to always provide the same response to the students
input.

Standardization of the information and teaching input

In recent months, we have been carrying out experiments in Barcelona which

aim to assess the inuence of the input or general information provided by the
translation text and for this purpose it is vital that all subjects receive the same
information to ensure the objectivity and replicability of the experiment. This is
very dicult to achieve with human teachers but much easier when using a
virtual teacher such as the computer.
In the experiment, two parallel groups of students complete a translation
</TARGET "neu2">

98 WILHELM NEUNZIG

from German into Spanish, one group gathers in-depth information on the subject
matter (of a markedly culture-centered nature) prior to commencing translation
work through a computer exercise, while the other group (the control group) does
not. Once again the entire translation process is recorded to allow analysis of the
dierent choices of translation strategy.

Conclusions and future perspectives

Use of the computer as a research tool in translation studies constitutes a new

approach for analysis of the translation process and the rst experiments carried
out in Copenhagen and Barcelona would seem to indicate that it is a valid means
of accessing knowledge and that this path should be followed. The computer as
a tool for data collection and translation research contributes to guarantee
scientic rigor in fullling many of the criteria which govern empirical research:
it always provides the same feedback, independently of the researcher (objectivi-
ty), it enables the pedagogical input to be standardized (reliability), it can repeat
the same situation with any number of subjects (replicability), it can control the
consistency of the design and allows the data obtained to be recorded (internal
validity), it does not discriminate against any of the subjects (fairness) and it
ensures practicability, since it does not aect the normal running of the transla-
tion class and requires no special eort on the part of the students. The most
outstanding feature of the computer as an instrument of research in translation
studies is, without doubt, its ability to obtain data on the translation process in a
laboratory experiment undertaken in a normal working environment, thus
ensuring the environmental validity of the experiment. It is therefore an instru-
ment which allows us to design laboratory experiments which simulate eld
experiments.

Translated by Carl Mac Gabham

<TARGET "pac" DOCINFO

AUTHOR "PACTE"

TITLE "Acquiring Translation Competence"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 10

Acquiring Translation Competence

Hypotheses and Methodological Problems
of a Research Project*

PACTE
Universitat Autnoma de Barcelona

Introduction

The empirical-experimental approach is relatively new in translation studies and

this lack of tradition makes it very dicult to set up an empirical research
project. The natural sciences, and more recently, the social sciences, can rely on
established theoretical and methodological criteria, whereas in translation studies
we cannot. There is little consensus on how to formulate working hypotheses,
design experiments, choose subjects, dene experimental and control groups,
control random variables, and so on. Perhaps the most dicult problem is how
to create reliable instruments to measure the parameters we are interested in and
to provide data for analysis.
Despite the diculties, some empirical studies have been carried out and
some isolated hypotheses have been conrmed. In our opinion, these studies are
important for three main reasons. They have opened up new perspectives to
understanding translation competence and performance, the area of translation
studies that is most closely linked to psychology. They have tested some
instruments to measure this competence. They have conrmed some earlier
intuitions about it, e.g., the translation process is not lineal but recursive,
strategies play a decisive role in the process, procedures vary signicantly
according to the individual, the text, or directionality.
In our project we are building on this pioneer research, working in three

* The PACTE Group includes A. Beeby, L. Berenguer, D. Ensinger, O. Fox, A. Hurtado Albir, N.
Martnez Melis, W. Neunzig, M. Orozco, M. Presas and F. Vega.
100 PACTE

related areas: (1) translation competence, (2) the acquisition of translation

competence, and (3) teaching proposals to develop and evaluate translation
competence. This article summarises our theoretical and working hypotheses and
suggests some of the methodological problems involved in studying such a
highly complex phenomenon.

A holistic and dynamic model of translation competence

Translation competence can be dened as the underlying system of knowledge

and skills needed to be able to translate. This competence is actualised in
dierent ways in dierent situations. The concept itself, borrowed from the idea
of linguistic competence, has become increasingly important in translation studies
as behaviourist paradigms have been replaced by cognitive paradigms. The three
main issues are (1) the components of translation competence, (2) its nature, and
(3) how it is acquired.

The Components of Translation Competence

Just as increasingly detailed models have been developed for linguistic compe-
tence, so componential models have been proposed for translation competence.
As they are all based on observation of the translators behaviour, there is
agreement about some basic components, such as, bilingual competence, transfer
competence, world or subject knowledge (Bell 1991; Wilss 1992; Beeby 1996;
Hurtado 1996a; Presas 1996; Shreve 1997, among others). Furthermore, it seems
obvious that if the translation process is building a target text (TT) from a source
text (ST), translators will always be detecting and solving new problems of a
linguistic nature (Krings 1987; Blanger 1992; Demers 1992; Mondahl and
Jensen 1992). Depending on the type of text they are working with, translators
will use special subject knowledge (Schner 1993; Dancette 1994) and
dierent tools, such as dictionaries or databases (Fraser 1994). Translators are
able to do all this, eciently, without losing sight of their objectives.
However, various studies seem to show the existence of two further types
of sub-competencies. The rst is related to specic problem-solving strategies
(Krings 1986; Lrscher 1991, 1992a, 1993; Kussmaul 1991, 1995; Wilss 1992;
Kiraly 1995). The second is related to certain psychological qualities and skills
that translators draw on, such as creativity (Kussmaul 1995) or attention span
(Jskelinen 1993). Therefore our model of translation competence includes a
problem solving or strategic component and a psycho-physiological component,
as can be seen in Figure 1.
 ACQUIRING TRANSLATION COMPETENCE 101

TRANSLATION COMPETENCE

COMMUNICATIVE EXTRA-LINGUISTIC
COMPETENCE IN THE TWO COMPETENCE
LANGUAGES

TRANSFER COMPETENCE

PROFESSIONAL
PSYCHO-PHYSIOLOGICAL
INSTRUMENTAL
COMPETENCE
COMPETENCE

STRATEGIC COMPETENCE

Figure 1: The Sub-components of Translation Competence

1. Communicative Competence in two languages can be dened in general

terms as the system of underlying knowledge and skills necessary for linguistic
communication. Following Canale (1983), we distinguish linguistic, discourse
and socio-linguistic components. Of course, for translators, this competence
should be separated into understanding in the SL and production in the TL.
2. Extra-linguistic Competence is composed of general world knowledge and
specialist knowledge that can be activated according to the needs of each
translation situation. The sub-components may include explicit or implicit
knowledge about translation, bicultural, encyclopaedic and subject knowledge.
3. Instrumental-Professional Competence is composed of knowledge and skills
related both to the tools of the trade and the profession. The sub-components
may be very diverse: knowledge and use of all kinds of documentation sources and
new technologies, knowledge of the work market (translation briefs, etc.) and how to
behave as a professional translator, especially in relation to professional ethics.
102 PACTE

4. Psycho-physiological Competence can be dened as the ability to use all

kinds of psychomotor, cognitive and attitudinal resources. The most important of
these may be psychomotor skills for reading and writing; cognitive skills (e.g.
memory, attention span, creativity and logical reasoning); psychological attitudes
(e.g. intellectual curiosity, perseverance, rigour, a critical spirit, and self-con-
dence).
5. Transfer Competence is the central competence that integrates all the others.
It is the ability to complete the transfer process from the ST to the TT, i.e. to
understand the ST and re-express it in the TL, taking into account the trans-
lations function and the characteristics of the receptor. The sub-components
include (1) comprehension competence (the ability to analyse, synthesise and
activate extra-linguistic knowledge so as to capture the sense of a text), (2) the
ability to deverbalise and to maintain the SL and the TL in separate compart-
ments (i.e. to control interference), (3) re-expression competence (textual
organisation, creativity in the TL), (4) competence in carrying out the translation
project (the choice of the most adequate method).
6. Strategic Competence includes all the individual procedures, conscious and
unconscious, verbal and non-verbal, used to solve the problems found during the
translation process. The problem-solving process can be described as a series of
acts or recursive, complex acts that lead from an initial state to an objective.
There are several stages in this process, the rst of which is recognising there is
a problem (Sternberg 1996). Examples of strategies are: distinguishing between
main and secondary ideas, establishing conceptual relationships, searching for
information, paraphrasing, back translating, translating out loud, establishing an
order for documentation, etc.

The Nature of Translation Competence

The nature of translation competence is at the heart of another current debate in

translation studies. Is it eminently theoretical or practical, conscious or automatic,
declarative or procedural? Anderson (1983) denes declarative knowledge as
knowing what. It is easy to verbalise, you have it or you do not have it, it is
explicitly dened and its processing is essentially controlled. On the other hand,
he denes procedural knowledge as knowing how. It is dicult to verbalise, it
can be partially possessed, it is acquired gradually, through practice and its
processing is essentially automatic. One type of knowledge does not exclude the
other, they may co-exist or, in a learning situation, as novice becomes expert,
declarative knowledge may give way to procedural knowledge. This idea was
 ACQUIRING TRANSLATION COMPETENCE 103

taken up by Dreyfus and Dreyfus (1986), who characterised expert knowledge as

non-reective (idem: 36) and claimed that if rational thought took place at all, it
would be a critical reection about ones own intuitions (idem: 32).
In the eld of introspective translation studies, the automatic or non-
reective nature of expert, operative knowledge is illustrated by the diculties
expert translators have in verbalising their mental processes (Kiraly 1995). This
has led us to consider translation competence as essentially procedural know-
ledge, in which the strategic component is primordial.

Acquiring Translation Competence

If translation competence is made up of inter-related sub-competencies, then its

acquisition is a dynamic process of building new knowledge on the basis of the
old. What may be most important in the process is the restructuring of existing
knowledge, rather than the addition of new information (Pozo 1996). Thus, the
novice stage in the development of translation competence could be dened as
the stage when the sub-competencies have been acquired, at least partially, but
they do not interact with each other. Therefore, the development from novice to
expert is not only a question of acquiring the missing sub-competencies, but also
of re-structuring the existing sub-competencies to put them at the service of the
transfer competence.
In order to acquire translation competence, the student also has to develop
a learning competence that includes the specic learning strategies that make this
and any learning process possible. These strategies are the system of plans and
operations used by someone who learns something to obtain, store, recover and
use information (Manchn 1994). Figure 2 below shows the main characteristics
of this process.

Hypotheses

In order to investigate translation competence and its acquisition using the above
models, we have formulated the following theoretical and working hypotheses.
The theoretical hypotheses related to translation competence are that
translation competence is a system of sub-competencies that are (1) interrelated,
(2) hierarchical, (3) used in every translation act and (4) whose interrelation and
hierarchy are subject to variations. The working hypotheses are as follows: (1)
the sub-competencies of translation competence are communicative, extra-
linguistic, professional-instrumental, transfer, strategic and psycho-physiological,
(2) transfer competence plays a central role in the hierarchy of sub-competencies,
104 PACTE

pre-translation competence

A
C
Q
U
I
development and integration S learning strategies
of sub-competencies I
T
I
O
N

translation competence

Figure 2: Acquiring Translation Competence

(3) the interaction between the sub-competencies is controlled by strategic

competence, (4) the interaction and hierarchy of the sub-competencies vary
according to directionality (direct or inverse translation), language combinations,
specialisation (literary, legal, technical translation, etc.), the translators experi-
ence or the translation context (translation brief, time available, nancial
conditions, etc.).
The theoretical hypotheses related to acquiring translation competence are
that it takes place through a process of restructuring and developing sub-compe-
tencies (1) which do not develop in parallel, (2) are interrelated, (3) specic
learning strategies are required and (4) the process is subject to variations. The
working hypotheses are as follows: (1) the development and re-structuring of
transfer competence is central to the acquisition of translation competence, (2)
the acquisition and the development of strategic competence are essential to
develop and re-structure the other sub-competencies, (3) the development of
specic learning strategies is necessary to develop and re-structure the other sub-
competencies, (4) the development and re-structuring of the other sub-competen-
cies varies with certain factors: directionality (direct or inverse translation),
language combinations, specialisation (literary, legal, technical translation, etc.),
 ACQUIRING TRANSLATION COMPETENCE 105

or the learning context (guided learning, self learning, etc.).

The next stage in our research is to reformulate these working hypotheses
as empirical ones so that they can be applied to observable behaviour and
measured. Only by doing this can we obtain data that will allow us to validate or
refute the above hypotheses.

The Problems of Empirical-Experimental Research

Empirical-experimental research in translation studies has to follow the same

stages as empirical-experimental research in any other eld. These stages are
shown in Figure 3 below.
Given that all sciences follow practically the same criteria at the conceptual
and analytic stages, here we will briey mention some of the methodological
problems we have in translation studies.1
The rst problem is how to dene the extraordinarily complex dependent,
independent and random variables that aect the reliability and internal and
external validity of our study. The variable we want to measure in one experi-
ment may be a random variable in another experiment. For example, extra-
linguistic competence may act as a random variable when analyzing instrumental
competence.
The second problem is related to measuring instruments. Certainly, there is
very little experience to build on. So far, the most common instruments have
been tests (problems of reliability, repetition, objectivity and control of random
variables), questionnaires and interviews (problems of objectivity and validity:
these post facto methods do not assure one is measuring what one set out to
measure) and TAPs (problems of situational validity). Furthermore, measuring
the acquisition of the dierent sub-competencies is complicated by the very
diverse learning methods used (learning by doing, imitation, cognitive processes,
etc.). Therefore, one of our main tasks is to develop a whole battery of new
instruments to take precise, reliable measurements.
The third problem is how to analyse the data. Here we have to elaborate a
system that will allow us to transform into quantitative data information that is,
by its very nature, qualitative (bicultural, encyclopaedic, professional knowledge,
etc.) This means we have to make a dicult choice. Either we force our data
into a numerical straitjacket to apply inferential methods and risk that the results
obtained do not represent reality, or, we limit ourselves to descriptive statistics
and end up with results that lack the explicative power we need.
</TARGET "pac">

106 PACTE

Conceptual
8. Generalize results
level 1. Delimit the object to be studied
or modify hypotheses

2. Formulate theoretical hypotheses

based on theories or models

3. Formulate empirical hypotheses

Methodological
level 4. Research design

5. Systematic collection of data

Analytical
level 6. Analysis of empirical data

7. Compare data with hypotheses

Figure 3: Research Stages (adapted from Arnau 1995)

Notes

1. For a further discussion of the criteria required in translation research, see W. Neunzigs
contribution in this volume.

Translated by Allison Beeby.

<TARGET "sco" DOCINFO

AUTHOR "Christopher Scott-Tennent"

TITLE "Translation Strategies and Translation Solutions"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 11

Translation Strategies and Translation Solutions

Design of a Teaching Prototype
and Empirical Study of its Results*

Christopher Scott-Tennent
Universitat Rovira i Virgili, Tarragona

Mara Gonzlez Davies

Universitat de Vic

Fernanda Rodrguez Torras

Universitat Rovira i Virgili, Tarragona

Introduction

This research project is an empirical study carried out to observe the eects of
a specically designed training programme on the students application of
translation strategies and solutions. The project has been carried out through
three clearly dierentiated stages. First, the existing literature on translation
strategies was studied, three types of translation problems were chosen and a
theoretically optimal methodology was designed in order to teach the strategies
to solve those problems.1 Then, a pilot study was conducted with two groups
(experimental and control) of undergraduate translation students. The subjects in
the experimental group were trained in the selected strategies, whereas the
control group did not receive this specic training. Finally, data analysis and
interpretation were carried out, and conclusions were drawn in order to optimise
preparation of a second study and decide on our future approach to this particular
area of research.

* Research funded by the Universitat Rovira i Virgili (Tarragona, Spain). Grant: Ajuts a la recerca
9678C.
108 SCOTT-TENNENT, GONZLEZ DAVIES AND RODRGUEZ TORRAS

Review of the literature

Up to now psycholinguistic research on the translation process has been centred

on discovering what happens in the translators mind when switching from one
language to another. According to Lrscher (1992b: 116117), research is needed
on the evaluation of translation strategies, as is the design of a methodology for
their teaching. Zabalbeascoa (1993: 50) emphasises the need to make trainee
translators aware of the dierent factors involved in the translation process and
then teach strategies which may help to improve production.
Our approach stems from a psycholinguistic view of translation as a process
of communication involving information-processing which takes place under a set
of constraints which make it dierent from other processes of communication.
Danks (1991) has pointed out that these constraints can be classied according
to the following categories: (1) task, or the activity which the translator is
required to carry out and the context in which this activity takes place, and (2)
text, or the linguistic and discoursal structure of the ST (Source Text). Transla-
tors operate under these constraints using their linguistic and non-linguistic
knowledge and skills.

The translation process: problems, strategies and solutions

The psycholinguistic approach to translation has investigated the processes

followed by translators to achieve their production (Bell 1988: 9195; Danks
1991: 94; Krings,1986: 87; Lrscher 1991: 92; Sguinot 1989: 91). As shown by
Lrscher (1992b), the process is partly automatic, partly non-automatic. This
means that the translator transfers some segments of information without having
to reect consciously on them and, from time to time, this automatic information
processing is interrupted by a problem encountered either in the comprehension
of the ST, in the production of the TT (Target Text), or in the transfer from one
to the other. Thus, a translation strategy can be understood as the means to
produce translation solutions for a range of translation problems which may be
anticipated either at the micro- or macro-levels. Bearing in mind the pedagogical
focus of this study, the main aim was that the students should develop the ability
to recognise and solve these translation-specic problems: problem-recognition
is a pre-requisite for problem-solving. Therefore, in view of the varying positions
in the literature and considering the pedagogical focus of this study, we dened
translation strategy as the steps, selected from a consciously known range of
potential procedures, taken to solve a translation problem which has been con-
sciously detected and resulting in a consciously applied solution. This does not
exclude the possibility that the application of such strategies may become
automatic with practice but, for the purposes of this study, it does exclude from
 TRANSLATION STRATEGIES AND TRANSLATION SOLUTIONS 109

the category: (a) solutions found without considering a range of potential

solutions, and (b) solutions found without any conscious reasoning at all, i.e.
intuitively.

Training in the translation process

For training in problem-solving, the less restricted to unique instances strategies

or potential solutions are, the more useful they can be. A potentially high
frequency of application is also desirable. On the other hand, the conscious
nature of their application means that the training should clearly go beyond a
collection of helpful hints: consciousness raising techniques become an appropri-
ate training tool in this context.
The rst of these conditions suggests a need for a taxonomy of translation
strategies which, in a systematic and categorical way, identies frequently
occurring translation-specic problems and suggests potential eective solutions.
Such a list could then, if made pedagogically signicant for undergraduate
students, form the basis of syllabus design for strategy training.
The second condition aects methodology rather than syllabus. Conscious-
ness raising can be more eectively induced when the learners not only apply
the target principles, but also evaluate this application themselves: thus, they can
and, in fact, do add self-discovered strategies to existing lists.

Design of the pilot study

A strategy-training course based on the above-mentioned assumptions and

principles was designed. This course was then taught in a naturally occurring
setting, in order to observe, measure, and analyse its eects. This pilot study
was designed to test the following three hypotheses:
1. It is possible in practice to undertake specic strategy training creating a
learning experience clearly perceived as satisfactory by all the participants.
2. As a result of this training, the frequency and/or eectiveness in applying
the strategies and solutions is signicantly enhanced.
3. This enhanced application leads to a signicant improvement in the overall
quality of the target text produced and its eects on the reader, as regards
the following aspects: intelligibility of the TT, and/or appropriateness of TT
to TL norms, and/or ST-TT correspondence of style and register, and/or
ST-TT ideational content correspondence.
110 SCOTT-TENNENT, GONZLEZ DAVIES AND RODRGUEZ TORRAS

The specic training focused on the following three translation problems:

1. No lexical correspondence SL (Source Language)/TL (Target Language)
2. Lexical phrases, proverbs and idioms
3. Cohesion and coherence
The subjects of this experiment were two groups of 12 students in their third
year of undergraduate translation studies at the Facultat de Cincies Humanes,
Traducci i Documentaci at the Universitat de Vic (Spain). Both groups were
taking the course Specialised Translation (60 contact hours) and had not
received any previous specic training in translation strategies. The groups were
naturally occurring within their setting, not created or modied in any way for
purposes of this study. The teacher of the control group was instructed to follow
her usual teaching procedures.
Data-gathering was designed as follows:
1. To test hypothesis 1:
(1) Observation of the experimental group was recorded by the teacher
in a class diary and later analysed
(2) Final semi-structured interview with learners (experimental group)
(3) Final written course evaluation from learners (experimental group)
2. To test hypothesis 2:
(1) Written protocols (in the experimental group) where students con-
stantly recorded their solutions to the translation problems encoun-
tered in texts during training, specifying in each case the problem
encountered, the strategy used to solve it, and the resulting solution
(see Figure 1).

STRATEGIES SOLUTIONS
Read aloud to somebody else Reformulation
Gapping the text for another reader to Leave in SL () in the target text and
complete and discover/discuss other trans- use literal translation or paraphrase either
lation options. in a footnote or in the text itself
Figure 1: Examples of translation strategies/solutions (from the literature and proposed by
the learners).

(2) Pre- and post-test: translation of the same text by both groups at the
beginning and end of the training period.
3. To test hypothesis 3:
(1) Pre- and post-test: translation of the same text by both groups at the
beginning and end of the training period.
 TRANSLATION STRATEGIES AND TRANSLATION SOLUTIONS 111

(2) Reading comprehension test (both groups)

(3) Writing skills test (both groups)
Band scales were used for all three in order to be able to quantify uniform
qualitative judgements so that they could then be processed statistically. Data
from markers who did not correlate acceptably were not included in the study.
Translation strategies and solutions were introduced by creating an awareness of
both notions through a discussion around questions such as: what goes on in our
mind when we translate? How do we spot a translation problem? How do we
solve it? What do we call the solution?
One group of problems was presented every 4 weeks. After the rst presentation
session on translation problems, strategies and solutions, each group of transla-
tion problems was presented through teacher-students and student-student
interaction and, nally, in a brainstorming session, teacher and students drew up
a list of the possible strategies and solutions to be used. Here, solutions observed
in translations by professionals were presented to conrm or complement those
suggested by the students.

Data analysis and results

The translation pre- and post-test

The students in both the experimental and the control groups carried out the
same translation at the beginning and at the end of the training period, with no
previous notice, of a popular science text called Dino DNA?, (Time,
24.11.1994) from English into their L1.
External markers assessed the translation. The overall mark for the transla-
tion test was the sum of the marks obtained for each criterion (see Figure 2).
These criteria had previously been statistically tested for inter-marker reliability.
In order to detect whether an adequate correlation between markers had
been established, a Pearson-moment test was applied and showed that the degree
of interrater correlation for the experimental group is acceptable (correlation 0.8
or above) even for most partial ratings and clearly more than acceptable for total
marks. Therefore, the marks from both raters were averaged out into one set.
A T-test to verify if changes between the results were signicant was then
done in order to observe if there had been a signicant increase in the overall
quality of translation. level, i.e. the level required for a result to be deemed
signicant, was set at the conventional 0.05.
112 SCOTT-TENNENT, GONZLEZ DAVIES AND RODRGUEZ TORRAS

1.A. Intelligibility of TT: To what 2.A. Transfer of ST style and register

extent is TT comprehension hindered
by its linguistic form?
not at all = 5
hardly = 4
slightly = 3
quite a lot = 2
greatly = 1
it is hardly intelligible = 0
1.B. Adequacy of TT to TL norms:
Would a reader detect that the target
text is a translation?
No reader would suspect this = 5
An expert might suspect it = 4
An expert would know for sure = 3
Any reader might suspect it = 2
Any reader would know for sure = 1
The target text is totally unnatural = 0
Figure 2: Band scales for translation assessment
formal correspondence is practically
100% = 3
formal correspondence is very good = 2
formal correspondence is acceptable = 1
clearly signicant formal dierences = 0
2.B. Transfer of ST contents
content correspondence is 100% = 7
dierences are present but almost unde-
tectable = 6
very few/unimportant dierences are pres-
ent = 5
very few signicant dierences are present
=4
several dierences are present = 3
several signicant dierences are present
=2
frequent and signicant content dierences
=1
ST content altered almost beyond recogni-
tion = 0

The results show a clearly signicant increase in the overall quality of the
translation in the experimental group (0.01008859), as well as a signicant
increase in 2 of the 4 criteria involved: intelligibility of TT (0.03253924) and
ST-TT correspondence in style and register (0.02341299). The criterion in which
improvement was least signicant is transfer of ST content (0.14727353).
In the case of the control group, markers did not correlate, so T-tests were
carried out separately. However, the markers verdicts are opposed: for some
improvement was very signicant, and for others it was not signicant. No
conclusions may therefore be reached regarding this group.

Control of the variables reading comprehension in SL and writing skills in TL

The pilot study started with a reading comprehension test in the source language
and 2 writing tasks in the target language in both the experimental and the
control groups at the beginning and end of the training period. Since these
 TRANSLATION STRATEGIES AND TRANSLATION SOLUTIONS 113

elements constitute the linguistic-textual component of translating competence, it

was necessary to control them to detect whether a potential increase in the
quality of the pre- and post-test translations might be due to a development of
these capacities on the part of the students and thus not to training in the
application of translation strategies.
For the reading comprehension test, the text Tribal Behaviour was taken
from a Cambridge Prociency In English exam (1990). The pre- and post-tests
yielded one set of objective scores in each case: a T-test showed no signicant
dierence in either group: in the experimental group the result was 0.21661543
and in the control group, 0.39909084.
Writing skills in the target language were measured by writing a letter
answering an advertisement for a job in translating from a newspaper and an
essay, both in the students L1. Band scales were used for assessment by two
external markers and a high correlation was required before averaging out both
sets of data into one. The T-test was applied to the results. This showed no
signicant increase in the experimental group (0.1086111), while the increase
was clearly signicant in the control group (0.01056797).

Frequency in the application of translation solutions

In order to verify whether there had been an increase in the frequency of

eective application of solutions, the T-test was also applied to the number of
eective solutions in the pre-test and in the post-test. The result was 0.00524944
in the experimental group. This indicates that the increase in the number of
eective solutions is very signicant. The result for the control group was
0.04239666, which also shows a signicant increase, though much less than that
of experimental group. On considering each group of problems/solutions
separately, one signicant increase was found in the experimental group:
0.00984431, to solve Type 1 problems (no lexical correspondence).

Analysis of written protocols: Overall application of strategies

The written protocols that the students in the experimental group lled in on a
weekly basis during the training period reect the percentage of correctly/
incorrectly identied problems and of adequately applied solutions as a result of
a reection on the translation process. They were considered a signicant
element in the experimental training due to their potential consciousness-raising
eect as a result of being a formative self-assessment activity. The categories of
solution production observed in the written protocols were the following:
114 SCOTT-TENNENT, GONZLEZ DAVIES AND RODRGUEZ TORRAS

1. Correctly applied strategy producing an eective solution

2. Incorrectly applied strategy but producing an eective solution
3. Correctly applied strategy but not producing an eective solution
4. Incorrectly applied strategy and not producing an eective solution
For each student in the experimental group and for each translation, a record was
kept of the percentage of strategy application falling within each category. The
group mean was then calculated for each translation, yielding the following data:
1. The percentage of strategies correctly applied and producing an eective
solution uctuates, but increases steadily and constantly throughout the training
period. The percentage of strategies not correctly applied but producing an
eective solution (which seems to suggest the translation problem was solved
intuitively) is already small at the beginning and decreases steadily and
constantly. The percentage of strategies correctly applied but not producing an
eective solution is clearly negligible and sometimes nil throughout, suggesting
that once a student knows what strategy has to be applied to solve the problem,
the result is practically always positive. The percentage of strategies not correctly
applied as well as not producing an eective solution is small from the begin-
ning and does not change signicantly.
2. The percentage of translation problems solved eectively decreases in the
initial stage of training, and subsequently increases steadily and constantly.

Analysis of the teachers class diaries

The class diary written by the teacher of the experimental group was an impor-
tant element of course evaluation. In this diary the teacher wrote down a priori
the aims of each session as well as her observation a posteriori. Data were later
extracted from this observation, and subsequently analysed, under the following
four categories:
Evidence of characteristic methodological aspects of the experimental training.
Evidence of students awareness of the strategies.
Evidence of positive attitudes from the students.
Evidence of negative or questioning attitudes from the students.

Course evaluation by the students

At the end of the training period, the students in the experimental group evaluat-
ed the experimental training through a written commentary and in semi-struc-
tured oral interviews.
 TRANSLATION STRATEGIES AND TRANSLATION SOLUTIONS 115

Conclusions

Regarding the experimental group, all three hypotheses can be said to have been
proved. The rst hypothesis is valid according to the data gathered by observa-
tion in the teachers class diary, as well as through the questionnaires and
interviews. Data from the diary show an initial phase of uncertainty until what
turned out to be the main diculty for the learners was overcome, i.e. distin-
guishing clearly between problem, strategy and solution: clearing up this
particular confusion required a month of training and the positive change in the
students attitude began when this was achieved. The most conclusive data are:
on the one hand, at the end of the experimental training, 83% declared them-
selves completely in favour of specic training in the application of strategies
and solutions, 17% were partially against it but none of them totally against.
Concerning the teacher, the best proof as to her view on this methodology is
that, after ending the study, she spontaneously integrates this type of training in
her subsequent teaching.
The second hypothesis is also valid according to the relevant data. On the
one hand, the increase in the frequency of eective strategy and solution
application in the pre- and post-test is very signicant (T-test = 0.005). On the
other hand, if we also consider what happened during the training according to
the written protocols, the percentage of eectively applied strategies and
solutions uctuates at the beginning of the training period, but it then rises
steadily. Furthermore, the percentage of eectively chosen and applied solutions
was 60% at the beginning and, after two uctuations, it rose steadily to 80% at
the end. Also at the end, data from the class diaries show that some students
were applying strategies and solutions without any explicit elicitation, which
shows that they were already being acquired (Kiraly 1995). An interesting by-
product is that during the experimental training the teacher discovered that
certain strategies and solutions can be applied more often and better in given
types of text. This can explain, at least partially, the fact that the most signicant
increase in the eective application of strategies and solutions was in those of
Type 1. It also poses a question for future research: is it possible to associate
certain strategies and solutions to certain types of text?
As for the third hypothesis, as is shown in the pre- and post-test, the
increase in the overall quality of translation was clearly signicant (T-test =
0.01). This increase can only have been resulted from the experimental training,
since there was no signicant increase in reading competence in the SL or in
writing skills in the TL. The improvement is also signicant in specic aspects
of overall translation quality, such as the clarity of expression in the TT and
</TARGET "sco">

116 SCOTT-TENNENT, GONZLEZ DAVIES AND RODRGUEZ TORRAS

textual correspondence in style and register. The specic aspect in which there
was least improvement is that of content correspondence, and this is logical since
this particular feature depends almost exclusively on reading competence in the
SL, which did not increase signicantly, as has been mentioned. These data thus
suggest that a greater and/or more eective application of strategies and
solutions contributes to an increase in the students translation competence.
Data were gathered from the control group to be contrasted with experimen-
tal group data regarding the second and third hypotheses. Concerning the second
hypothesis, the increase in the eective application of solutions in the translation
pre- and post-test was also signicant in the control group, though not nearly as
much as in the experimental group (T-test = 0.04). This increase may have been
caused by both or either of two factors: (a) the non-experimental teaching in fact
included some elements of implicit or explicit training in the application of
strategies and solutions as a part of the teachers habitual practice, or (b) the
signicant increase in writing skills in the TL that was observed in this group
produces some similar eects to those of a more eective application of
strategies and solutions. The latter possibility is especially interesting. As far as
the third hypothesis is concerned, the results are not conclusive: it can neither be
stated that there was a signicant increase in overall quality of translation, nor
that such an increase did not take place.
Particularly interesting possible directions which have emerged for future
research are: (a) qualitative aspects of the application of strategies and solutions,
(b) a taxonomy of problems, strategies and solutions, (c) studies to explore
whether certain translation strategies are acquired before others, or (d) to discern
to what extent learning or teaching styles, or even translating style, inuence the
acquisition of translation strategies. Directly related to the present pilot study, a
full-blown empirical study of a larger population and a greater control of
variables should throw more light on the results shown here.
Finally, it must be said that the collaboration between specialists in Psycho-
linguistics and in Translation Studies has turned out to be positive and fruitful,
as anticipated by Krawutsche (1989).

Note

1. A detailed description of this methodology will be oered in a forthcoming article.

<TARGET "zab" DOCINFO

AUTHOR "Patrick Zabalbeascoa"

TITLE "From Techniques to Types of Solutions"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 12

From Techniques to Types of Solutions*

Patrick Zabalbeascoa
Universitat Pompeu Fabra, Barcelona

Introduction

The aim of this paper is to present a brief review of translation techniques and
study the development of Vinay and Darbelnets initial proposal (henceforth the
initial proposal). For some people, the study of translation techniques is the
cornerstone of translation methodology and translator training; for others it is a
theoretical anacronysm. Does this mean that some teachers and textbook writers
are not keeping up to date with the latest developments in theoretical studies, or
does it mean that the theorists have chosen to sweep the issue under the carpet
and direct their interests elsewhere? We must be very careful not to oversimplify
the answer to this, although the real question is, put bluntly, in what way is
technique a useful concept? Possible answers might include: (i) to better under-
stand or explain certain phenomena (from the theoretical domain); (ii) as
categories and tools for description (within descriptive studies); (iii) to make the
learning process of trainees more ecient, or to ll in important terminological
gaps for reviewers and critics (within the applied extensions).
The problems posed by the proposal have been pointed out and discussed by
numerous scholars, but that does not seem to have been enough, in many cases,
to break the deadlock. The present situation oers us a host of names such as
procedures, strategies, shifts, methods, replacements and operations, all of which
are synonymous to a greater or lesser degree with techniques, while polysemy is
also a fact of many of these terms due to conceptual distinctions that are drawn

* This study was carried out as a part of Research Project Num PB 950985-C03101 Anlisis
contrastivo de los elementos conectores de la argumentacin y de los elementos temporales en textos
de especialidad (francs/castellano, ingls/castellano) nanced by the Spanish Ministry of Education
and Culture.
118 PATRICK ZABALBEASCOA

by dierent authors, creating much confusion. If we cant always be original

lets at least try to be clear and coherent in our use of terms and concepts to
reduce the degree of confusion. That is the basic aim of this paper.

The translating process

When dealing with the term process, I propose that we distinguish as much as
possible between the concept of a broad context where process covers the
stages of designing a translation project, selecting the source text (ST), hir-
ing/selecting translator(s), getting the ST to the translator, getting the target text
(TT/translation) back from the translator, editing1 of the text and all of the other
stages of publishing (or broadcasting, etc.) and distributing a TT and getting it to
its target users. In this broad context one might even include the process of
keeping the text alive by further stages of critical reviews, reprints and editions,
appearance in bibliographies, quotations, imitations, and the event that the
translation may end up being used as a ST for translation into a third language.
So, the broad context includes aspects of the initial norm (as coined by Toury
1980, referring to translation policies), and the various means of rewriting (as
proposed by Lefevere 1992). The concept of initiator is used to refer to the
person or institution that starts the process in its broad context, i.e. the initiator
sees the need and purpose for the future TT, and acts in order to enable the
existence of a translation process, typically, but not necessarily, by paying for the
whole project (see patronage in Lefevere, op. cit.); the process is completed, in
a way, once it reaches its initially-intended users, whereas the whole life-cycle
of the text is not over until it has ceased to be rewritten and is no longer used
(i.e. read, listened to, or watched). In any case, the process of translation in its
broad context may take weeks, months or years.
Process is also used to refer to the linguistic and/or mental operations
within the narrow context of a translator who is faced with a ST and a
commission to translate it. In this case the process begins when the translator
starts to analyse a text as source-text-to-be-translated and goes on for as
long as s/he tries to nd satisfactory renderings for individual aspects and items
of the ST in the TT and for the text as a whole.
Narrow context studies are primarily interested in linguistic and stylistic
phenomena, and focus on a bottom-up analysis of the process; this is also the
context for psycholinguistic and cognitive notions of process, nurturing the belief
that translation is fundamentally an individual activity, ignoring the possibility of
a task carried out by more than one translator (this would force us to consider
 FROM TECHNIQUES TO TYPES OF SOLUTIONS 119

the convenience of studying such things as communication between co-transla-

tors, comparison of notes or versions as part of the process). The broad context
approach is more interested in the sociocultural (systemic) phenomena and
favours a top-down analysis. This leaves us with a gray area between narrow and
broad for such aspects of translation as teamwork and revision, which are not as
widely studied as others which squarely fall within the (narrow) domain of
psycholinguistics, or (broad) literary systems, for example. In the narrow context,
a long process is more of a conceptual label than a strictly chronological one
(e.g. the duration of hesitations and thought processes is more interesting than
typing speed); from the psycholinguistic point of view a long process is a
complicated one, with lots of stages, operations and research, one where there is
a lot of decision-making, cross-checking, double-checking and backtracking. In
many bottom-up analyses, time is not even used as an operative factor.
A certain vagueness in the notion of process could explain the contradiction
between the use of the term strategy in translation studies and its use in other
elds. When phenomena such as lexical borrowing, or the use of hyperonym in
the TT are referred to as strategies, the implication is that the process is one of
gradually building up a text from its smaller constituents. So, are the smaller
textual constituents part of the process, or part of the product, or both? Where do
we draw the line then between process and product? It probably makes more
sense to regard the segments of a TT as constituent elements of the product.

The need for terminological coherence

Below are related denitions proposed for method, strategy, solution and
technique.
A method is usually dened as way of doing something in accordance with
a predened plan; it is less sensitive to contingencies than a strategy. From this
point of view, the term method would naturally fall in the realm of process rather
than product. However, a translation method is almost always used to refer to the
global characteristics of the product, especially in relation to its source text.
Method is used in product-oriented studies to refer to one or more translational
criteria; it can be distinguished from strategy by introducing the requirement that
it be identiable on the basis of the formal and functional characteristics of a TT
and its ST. Method is the relationship between the source text and its translation
and their respective communicative situations. A good example of a list of
normative methods can be found in Newmark (1988); the range covers the
ground between the long-standing methods of word-for-word and adaptation,
120 PATRICK ZABALBEASCOA

with his own communicative and semantic methods at the centre. Each one
comes with a short recommendation for when it applies (or not, as the case may
be). Another example can be found in the rank-bound v. nonrestricted methods
proposed by Catford (1965). Normative approaches will prescribe certain
methods for certain occasions, although they have also aspired to prescribe a
single, universal method, whereas descriptive studies aim at nding a pattern in
the relationship between each method and other common characteristics of the
TTs that share the same method (e.g. historical period, political regime, status of
translations in the target community). In the paradigm of norm theory and
descriptive studies, a given method can be postulated as the name given to a
combined group of norms for a given TT.
A strategy is a specic pattern of behaviour aimed at solving a problem or
attaining a goal; in translation, the goal is the TT according to its specications.
Strategy is proposed here as any conscious action(s) intended to enhance a
translators performance for a given task, especially in terms of eciency and
eectiveness. Strategies of this kind cannot be discovered by descriptive studies
of the texts alone since the underlying principle is that a given result might be
reached by dierent paths. In a way, this is related to some of the discoveries of
TAPs2 that show how trainees and experienced translators have dierent
behavioural patterns, although this does not necessarily lead to dierent results
(see Kiraly 1995). It seems useful to distinguish behavioural strategies from
mental activity, where the former would include actions that could be observed
directly by the researcher and recorded on video as the translating job is carried
out (consulting a dictionary, writing a draft version, underlining while reading,
taking a break at certain intervals, etc.). Mental activity refers to the thought
processes that can only be detected indirectly by noticing indicators or symptoms
(hesitations, mumblings), or otherwise by means of interviews and think-aloud
protocols, or by special equipment that can track neural activity. If we take an
analogy from chess, mental activity is not included in the strategies. They are
always understood to be the actual movement and arrangement of the chess-
pieces, and also the strategic distribution of time adapted to the players charac-
teristics and the type of moves that are planned or predicted; strategies depend
on the desired result (a win or a draw).
Here are a few examples of areas where various strategies specically
directed at improving translational performance might be put into practice: text
analysis and reading strategies, (TT) writing strategies, research strategies,
information processing (nding, retrieving, ling, exchanging, etc.), revision
strategies, eciency-oriented organizational strategies (systematizing, planning,
distribution of time, layout of materials), fallback strategies (to adapt to new or
 FROM TECHNIQUES TO TYPES OF SOLUTIONS 121

dicult situations), assessment (of price, diculty of task, time given),

selection strategies for identifying and studying problems and for nding
satisfactory solutions. An in-depth study of the complete range of translation
strategies would be very useful for applied studies in translator training, includ-
ing the convenience of dierent strategies for dierent levels of expertise and
at dierent levels of a training programme.
A solution rendering, version is what is reached as a result of a
strategy. If translating is accepted as a decision-making process or a problem-
solving activity, as it often is, then it makes perfect sense to speak of solutions
as the end-result of such a process or activity. The TT is the global solution to
the problem posed by the ST together with the criteria for its translation; the
local solutions are the renderings of identiable segments or features of the ST.
When local solutions cover clearly identiable segments of both the ST and the
TT they may be said to operate within, and determine, the boundaries of a
translation unit. From a prospective point of view, it is a question of parsing the
ST into its constituent parts and then replacing them, one at a time, in order to
build up the TT (e.g. when grammatical structures are taken as units, transposi-
tion is said to occur if the TT solution resorts to a dierent grammatical
category or structure to the one in the replaced segment of the ST). From a
retrospective (descriptive) point of view it is more of a question of rst nding
meaningful bitextual pairs, which means that the length and nature of each
segment is determined by the type of solution, which provides evidence of the
problem as the translator presumably saw it (yielding: unit = [ST] problem +
[TT] solution). Although there is a clear connection between solution and unit of
translation two points still have to be made. First, there may be more than one
type of solution within the boundaries of a given unit. Second, it may be the case
that not all solutions are purely segmental3 in their nature; these solutions are
generally referred to as (dierent forms of) compensation, although there are
segmental forms of compensation, too.
Technique is a concept that is not usually associated to a decision-making
process, but to an acquired skill to be applied according to a prescribed method
or procedure (e.g. a way of playing a musical instrument or of painting). In
Translation Studies, techniques can be kept in the terminology to refer exclusive-
ly to the initial proposal. I do not see the convenience, though, of substituting the
initial proposal with the idea of shifts, which clearly owes much to the idea of
techniques and does little more than perpetuate the metaphor of displacement in
translation. Ian Mason (1994 and 1995) is an example of how an author can pick
up the initial proposal and, by combining it with much more recent developments
in the eld (the insights of text-linguistics and descriptive studies), make it more
122 PATRICK ZABALBEASCOA

useful to our understanding of the nature of translation and its didactics.

The notion of translation technique comes from a prospective (and prescrip-
tive) approach, which involves looking at the ST and deciding what its smallest
constituent parts are for the purpose of translation, i.e. a type of parsing, and
then considering the most convenient way (technique) of rendering each unit.
The basic unit from this angle is the lexical unit, especially content words and
phrases. The diculty, complexity, and even the frequency and desirability of
each technique is more or less implicitly measured against the ideal of literal
translation, following Newmarks motto translate as literally as possible and as
freely as strictly necessary. From a retrospective point of view, translational
units are established on the basis of tracing TT items back to segments of the
ST. Once these bitextual4 units are established, what remains to be done is to
describe the nature of the relationship between the ST segment and the TT
segment, and it is precisely the name given to the relationship between these two
segments of ST and TT that we can call solution-type. A solution-type, then,
is the shared characteristic of a number of dierent solutions. Notice that this
does not lead us automatically to a taxonomy or even an enumeration of
mutually exclusive categories if we admit that within a given STTT bitextual
unit there may be more than one type of relationship. Solution-types may be
conceptually incompatible, i.e. mutually exclusive by denition. However, one
can also use the results of descriptions of a large corpus of translations to
discover conditions that make two solution-types incompatible within the same
unit. For example, addition5 and deletion6 seem to be conceptually incompati-
ble categories, whereas functional equivalence and lexical equivalence might
statistically prove to be highly incompatible, although they may not be so by
denition.7 Descriptive studies could also yield incompatible solution-types not
only within the same unit but also for neighboring units, or for number or
density of solution-types over a whole text. For example, how much borrowing
is allowed in a given text is a dierent question to whether borrowing is or is
not allowed and for which cases.

The training process

From the point of view of translator training the word technique conjures up the
image of the learner acquiring the necessary set of translation techniques in order
to become a qualied professional. This image is somewhat inaccurate, or
overambitious, especially when technique is too closely linked to the idea of
applying certain mechanical procedures or formulae. If our view of translating is
 FROM TECHNIQUES TO TYPES OF SOLUTIONS 123

one of a decision-making process and eective communicating it is especially

important to point out that techniques as formulated by the initial proposal are
really types of solutions, and a translators skills cover a much wider area than
a knowledge of possible low-level correspondences. A trainee needs to develop
an awareness of the operative variables in translation and also needs to be
exposed to a wide range of sample cases (these might be structured according to
a proposal like Newmarks procedures or by showing proposals from a number
of authors). Finally, a trainee needs to be exposed to as many dierent problems
as possible in order to develop problem-solving, decision-making strategies. It is
important to realise that an experts strategies and narrow context translating
process may only be reached gradually, and there may be dierent appropriate
processes for each stage of the learning process, and a dierent set of strategies
for each stage. Hopefully, it is now apparent that a mere presentation of solu-
tion-types is not enough; it has to be integrated into an awareness-raising, skills-
developing, experience-gaining learning process aimed at producing adaptable,
informed and resourceful translators.

Beyond the initial proposal

Certain trends within translation studies are more likely than others to drop or
ignore the initial proposal. The notion of technique relies heavily on the concept
of equivalence, so schools of thought that have argued against the use of
equivalence would understandably make little or no use of the initial proposal.
Descriptivists will frown on its aprioristic and underlying prescriptiveness, while
polysystem-oriented studies tend to be more interested in comparing translations
with target-language originals or other TTs in that language rather than with their
STs. Models based on the concept of deverbalization are not really suitable,
since they are born out of a psycholinguistic approach and the initial proposal is
rooted in comparative stylistics. Equally likely to abandon presentations of
techniques are the didactic models which are more focused on developing
reading, writing, investigative and analytical skills (i.e. strategies) in their
students, and communicative skills in general as well as ways of dealing with
highly specialized situations, rather than getting students to produce, somewhat
mechanically, translations that are as similar as possible their teachers version
by applying certain rules of thumb and transfer procedures.
The alternative to abandoning the initial proposal is to build on it, or aspects
of it, by putting a ner point on some of its concepts, or use some of the
categories for descriptive studies. One can look for norms and frequencies and
124 PATRICK ZABALBEASCOA

carry out comparative studies in order to propose explanatory accounts of some

of the traditional terms (e.g. Masons study of borrowing 1994).

STTT relationships

Here are levels, or discrete planes, on which STTT relationships can be found,
based on verbal/nonverbal, formal/functional and segmental/non-segmental
distinctions. From this basic framework, various types of STTT relationships
can be postulated, such as equivalence and compensation, outlined below.
Formal plane of verbal segments (of any meaningful length, including all
linguistic and textual units)
Formal plane of nonverbal and paralinguistic segments (gestures, pictures,
signs)
Functional plane of abovementioned segments (e.g. cohesive, semantic,
referential, communicative, aesthetic, semiotic functions)
Formal plane of nonsegmental aspects (rhyme, alliteration, intonation,
layout, etc.)
Functional plane of nonsegmental aspects, i.e. the cohesive, etc. functions of
these forms.

Equivalence

Let us consider equivalence as a variable for the lower levels of the text. We
could then use it as a basis for oering a list of types of equivalence as a means
of classifying translational solutions and proposing categories of solution-types.
Equivalence, here, is used to mean: sameness or a degree of similarity that, for
practical purposes, justies being qualied as sameness in dierence (resulting
from two dierent communication acts). Constituent elements might show
equivalence in any one of the following areas:
The functional plane (giving rise to functional equivalence), with a dierent
level for each function, e.g. propositional, interpersonal, aesthetic, intra-
textual, semiotic equivalence
The formal plane (producing formal equivalence), with a dierent level for
each type of form, e.g. morphological, stylistic, structural, and lexical
equivalence
Equivalence of merit, from an evaluative point of view, of such aspects as
quality and originality (of the texts as they are perceived by their respective
users, not necessarily of the translators contribution)
 FROM TECHNIQUES TO TYPES OF SOLUTIONS 125

Others, including equivalence of explicitness, implicitness, deviation from

standard use, vagueness and ambiguity.

Compensation

Compensation tends to imply compensating for the nonavailability of a literal

translation (that is stylistically or pragmatically appropriate) by doing something
rather drastic. From a retrospective point of view compensation, like adaptation,
is a concept that helps to account for high-level solutions, or patterns of solu-
tions. Let us dene compensation as a cause-eect relationship between (one or
more levels of) inequivalence and equivalence in a given solution; the intended
level/s of equivalence is/are the cause (providing a justication for the compen-
satory solution) and inequivalence is the necessary eect or sacrice, which is
justied as the best or only available means of achieving equivalence at
the desired level. There may be additional systemic levels of inequivalence that
also justify the compensatory tactic (systemic dierences between the two
languages or communication situations). For example, if we compensate for not
using rhyme as used in the ST by using alliteration in the TT for the same
purpose or function, which could be mnemonic for instance, then the intended
equivalence is sameness of mnemonic quality, inequivalence is alliteration where
there was rhyme, and systemic inequivalence could be the fact that the typical
mnemonic device of the TT users is alliteration and not rhyme, or the fact that
one language favours rhyme in those situations where the other language uses
alliteration.
Each type of compensatory solution can be and, traditionally tends to be
labelled according to the most obvious type of inequivalence (although
omission or rearrangement can be seen as means of compensation they are also
undeniably aspects of inequivalence). The attempt, for instance, to produce the
same punning eect in a part of the TT that is dierent to the location of its ST
counterpart is often referred to as compensation in place (I think a better term
would be compensation by displacement), rather than punning compensation.
So, apart from types of equivalence for classifying solutions there is also the
possibility of ascribing solutions to compensation by , completing the phrase
for each category with such options as: (i) omission or addition, not only of
segmental units, but also semantic, stylistic and pragmatic features; (ii) substitu-
tion (by omitting something and adding something else as part of the same
solution), including the substitution of verbal for nonverbal signs and vice versa,
or one type of gure of speech for another; (iii) increased explicitation (and,
conversely, increased implicitness); (iv) displacement, as shown above; (v)
126 PATRICK ZABALBEASCOA

rearrangement, the elements are grouped together in a dierent way, or the

order in which they appear is dierent.

Conclusion

In this paper I have argued for a more coherent terminology in the eld and
proposed terms and denitions to that eect. We have a history and a tradition
in Translation Studies, and these cannot and should not be ignored, nor should
they be perpetuated in the theoretical and pedagogical domains. My answer to
this situation is to propose that the term techniques be given a historical dimen-
sion to refer exclusively to Vinay and Darbelnets initial proposal and to similar
proposals by other authors such as Newmark. For the theoretical eld, the
recommendation is to continue working in establishing discrete categories along
the lines of what is proposed above for equivalence and compensation. We need
to change a single list of overlapping categories such as the initial proposal,
probably for several lists of solution-types (one for types of equivalence, another
for types of compensation, etc.) which contain a more coherent set of categories
for research and for better communication among scholars.
The conclusion for translator training is that the initial proposal, if used at
all, should be presented with great caution. The purpose of grouping solutions
into solution-types is to provide meaningful samples of options for the trainee
and illustrations of translator behaviour. Ultimately, the future translator will
have to deal with situations and contingencies that have not been presented in
class or studied in the literature, and it is with this in mind that strategies and
attitudes are to be worked on. Strategies and solution-types are to be presented
as mind-openers, not as a closed set of categories that act as blinkers in the
search for optimal solutions and fully satisfactory translations.

Notes

1. Post-editing of the TT is what usually comes to mind, but pre-editing of the ST must also be
contemplated.
2. Think-aloud protocols, whereby translators are asked to verbalise their thoughts as they go
along.
3. See STTT relationships below.
4. Bitext as proposed by Harris 1988 and dealt with in Toury 1995: 96.
5. Presumably of a TT segment that has no ST equivalent of the same rank.
</TARGET "zab">

FROM TECHNIQUES TO TYPES OF SOLUTIONS 127

6. A ST segment with no equivalent of the same rank in the TT, or the TT segment is empty.
7. In some accounts there seems to be a strong implication that when a solution is at the same
time a lexical equivalent and a functional equivalent it counts as lexical equivalent, so
functional equivalence as a label is actually meant for functional equivalents that cannot be
classied as lexical equivalents, thus making them mutually exclusive by denition.
<TARGET "mun" DOCINFO

AUTHOR "Ricardo Muoz Martn"

TITLE "Translation Strategies"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 13

Translation Strategies
Somewhere Over the Rainbow

Ricardo Muoz Martn

Universidad de Granada

Introduction

In 1982, Hnig and Kussmaul wrote that translation strategies are the main link
between theory and practice, and stated that translation strategies justify transla-
tion theory and should be integrated in the teaching of translation. Ten years
later, Gile (1995b: 13) seemed to think that the teaching of these strategies is
already common in many courses. Translation strategies are also present in most
of the recently published translation manuals, some of them even with exercises
on this topic. The present article aims to propose another way to approach them,
which can be applied to the teaching of translation.

On the concept of strategy

Terminological homogeneity is, paradoxically, not one of the characteristics of

translation theory. Wilss (1988) remarked that translation strategy seems to be
a rather diuse concept which refers to the general transfer perspective or transfer
concept of a particular text. One of the possible starting points is to weed out
the dierent understandings of translation strategy which are not going to be
considered here. Space constraints preclude the discussion of directly related
concepts, such as equivalence, adequacy, function, and translation unit, but some
terminological precisions seem in order. Equivalence and adequacy are taken
here to be value judgements by the translators themselves on the felicity of
(possible) solutions they have created when trying to solve translation problems
related to text segments of various lengths. Function or skopos is assumed to be
a result of the complex mental interaction of various textual and contextual
130 RICARDO MUOZ MARTN

parameters and features, and not an abstract, rationalistic directive. The concept
of text segment is preferred here to that of translation unit, and is dened as any
piece of text separated from the rest to be analyzed as a whole.
From the point of view of the scope they adopt, the attempts to formalize
translation strategies can be grouped in three main lines. The rst group focuses
on text segments, and encompasses the comparative/contrastive proposals of
Vinay and Darbelnet (1958), Vzquez Ayora (1977), Wotjak (1981), and Malone
(1988) on the one hand, and most cognitive approaches, such as Krings (1986),
Lrschers (1991), Sguinots (1991), and Rastalls (1994), to name but a few.
We will come back to these proposals below.
The second group consists of proposals which envision translation strategies
as procedures which aect the processing of the whole text. Reiss (1976) denes
some methods of translation which would apply to the text as a whole, and not
to smaller segments of it. Thelen (1990) also adopts a highly abstract level, and
proposes a two-phase strategy model: analysis and synthesis. Atari (1994),
following Tannen (1980, 1982), uses strategies to refer to the sets of resources
and characteristics associated to either spoken or written language, which he
applies to the teaching of translation. Venuti (1995) uses the term strategy to
mean the attitude or goal with which translators approach their tasks, and
distinguishes between domesticating and foreignizing strategies. Kohn & Kalina
(1996) focus on interpreting (oral translation) strategies, but still from a general
perspective of the activity as a whole. These proposals are either focused on the
translation product or too abstract to be of any use to further students translation
skills, and will not be commented upon.
The third group combines the former scopes in complex proposals which
contain dierent categories. Wilss (1988) makes a dierence between methods,
due to dierent conceptions (hermeneutic and analytical) of what translating is
about (Cap. 7). Then he introduces other categories, such as strategies and
techniques interiorized translation routines (1988: 157), procedures (Cap.
6) and skills (1988: 37) which are synonymous to experience. Newmark (1988)
sees a dierence between translation methods semantic, communicative, etc.
and translation strategies, most of which are drawn from Vinay and Darbel-
nets 1958 spearheading denitions. Hurtado (1996b) distinguishes between types
and methods of translating. In general, all these formulations lack empirical
evidence and the denitions of their categories, let alone the dierences amongst
them, are far from clear and somewhat shaky.
There is still at least one more approach belonging to the third group which
seems to deserve more attention, namely, Kiralys (1990) and Hnigs (1991)
distinction between macro- and microstrategies. Hnig (1991: 8085) postulates
a general strategy that would govern microrules, and which needs to be learned:
Transfer (innate) competence may sometimes produce perfectly acceptable
 TRANSLATION STRATEGIES 131

solutions which, however, are not recognized as such by the translator because of
his lack of translatory (acquired) competence, i.e. a macrostrategy. Hnigs
proposal seems intuitively correct, but no proof can yet be submitted on the
nature, content, and structure of these macrostrategies. Here we want to deal,
more modestly, with what he dubs microstrategies translation strategies from
now on because they are more concrete and seem easier to discern. In other
words, it is theoretically sound to propose the smallest analytical frame and let
the data drive its development. A short review of the proposals of the rst group
follows in the next two sections.

The comparative/contrastive approach

Vinay and Darbelnets well known list of procedures basically consists of

seven categories: Emprunt, calque and traduction litrale are dened as direct
procedures, whereas transposition, modulation, quivalence and adaptation are
qualied as oblique, in a terminology probably inspired in the concept of casus
rectus in classical Latin grammar. The taxonomy is not comprehensive enough
to reect all the possible relationships between two text segments from dierent
languages. In fact, Vinay and Darbelnet themselves list a few more procedures
in the glossary which precedes their book. Although the (complete) list is very
suggestive, it does not seem to be very helpful when learning how to translate.
Vzquez Ayora (1977) basically rewrites Vinay and Darbelnets proposal
from and within a contrastive linguistic paradigm (drawing from both structur-
alist and generative linguistics, according to the author). His main original
contribution is to abandon the original scale of diculty and to substitute it by
another consisting of degrees of closeness to translation proper, in a way that
easily leads him to prescriptivism. Wotjaks (1981) approach oers the most
detailed taxonomy to date, far too long to t in any table, but the most internally
coherent within this line. Malone (1988) oers the most abstract and formalistic
proposal, in a truly generative fashion, but his denitions are so wide that most
categories overlap with, or t in, others. For instance, all categories except for
equation, can be thought of as special cases of substitution. Newmarks (1988)
eclectic approach seems to be based on dierent understandings of strategy,
which may be analytical (as in componential analysis), or production oriented (as
in naturalisation), focusing on the source- or the target language (descriptive or
cultural equivalents), from a linguistic or a stylistic perspective (modulation and
compensation).
In any case, Delisles (1988: 7273) criticism of Vinay and Darbelnets
proposal applies to all of them:
132 RICARDO MUOZ MARTN

Vinay and Darbelnets translation procedures do not help the translator to nd

translation equivalents. A procedure is a method to obtain a result, a way of
doing something, of carrying an activity through to its conclusion. But these
procedures are in fact labels attached to results; the authors describe struc-
tural changes that occur in the translation process, or point out what does not
change. [] the categories of comparative stylistics (and particularly the so-
called translation procedures) cannot really be applied to the analysis and re-
expression of messages, or even the verication of equivalences.

That probably explains why Sguinot (1991) observed that students formally
instructed in the use of these strategies promptly abandoned them when facing
practical translation tasks in class. Mason (1994) points out that the view of
translating underlying the techniques is one of langue-to-langue comparison, i.e.,
an exercise in contrastive linguistics. Thus, the comparative/contrastive approach
might then be of interest when teaching contrastive textual analysis as a prepara-
tion to learn how to translate (Lpez and Minett (1998: 235) but Elena (1994: 61)
does not nd them useful even in retrospective analysis. The comparative/
contrastive approach has made its way into most translation manuals, but their
authors seem unsatised with it and tend to adapt the theoretical proposals in
such a way that no list is identical to any other, as shown for some cases in
Table 2. It is not surprising, then, that other approaches have been tried, such as
the empirical psycholinguistic studies which will be commented below.

The psycholinguistic approaches

Faerch and Kasper (1983: 36) dened communicative strategies as [] poten-

tially conscious plans for solving what to an individual presents itself as a problem
in reaching a particular communicative goal. Krings (1986: 175) seems to think
of translation strategies as a special case of communicative strategies, []
potentiell bewusste Plne eines bersetzers zur Lsung konkreter bersetzungs-
probleme im Rahmen einer konkreten bersetzungsaufgabe, and the same is true
for Lrscher (1991: 79ss.), Sguinot (1991: 79), Rastall (1994: 38), and Muoz
(1995), amongst others. This basic agreement does not prevent the psycho-
linguistic approach from heterogeneous developments. Seguinot (1991: 82) points
out that Strategies is a term which has been used to refer to both conscious and
unconscious procedures, to both overt tactics and mental processes.
As to whether a psycholinguistic approach to translation strategies should
study conscious or unconscious procedures, the studies in the eighties focused on
the translation process as problem-solving (for instance, Kussmaul 1986), but
 TRANSLATION STRATEGIES 133

today intuition seems to be more important. Dreyfus & Dreyfus (1986: 1651)
already pointed out that the decision-taking model (which they seem to consider
synonymous of the problem-solving model) can only apprehend but a small part
of the translation process. Wilss (1996) makes a dierence between original and
automated strategies and claims that the second is a rule-governed behavior, and
Mondahl & Jensen (1996: 102) distinguish spontaneous sequences from problem
sequences.
The psycholinguistic approaches have also dealt dierently with the level
of analysis. While most scholars focus on text segment levels, others, such as
Nord (1993: 290) think that a translation strategy applies to the translation of the
whole text, and that it is rational, although not necessarily conscious:
Anhand des Vergleichs der Ergebnisse von Auftragsanalyse und Ausgangstext-
analyse kann der bersetzer entscheiden, welche bersetzungsstrategie (im
Sinne der Gesamtheit der zu einem bestimmten Ziel fhrenden Verfahrens-
oder Verhaltensweisen) zur Erfllung des bersetzungsauftrags geeignet ist.
Dabei erscheint mir die Feststellung wichtig, dass es sich hier um eine
rationale, vor Beginn der bersetzungsttigkeit unter Bercksichtigung aller
relevanten Faktoren und Gegebenheiten getroene Entscheidung handelt, auch
wenn das routinierte Vorgehen eines professionellen bersetzers die
einzelnen Stufen der Entscheidungsndung nicht (mehr) in jedem Falle
erkennen lsst.

Today we are witnessing an explosion of empirical psycholinguistic studies on

the translation process by means of think-aloud protocols which mainly focus on
translation strategies in one way or other but, apart from the traditional criticisms
of the method (Cf. Toury 1991; Sguinot 1996), there are some aspects which
undermine their validity. First, the information obtained is not always of the
same nature. Thus, while Krings (1996) obtained oral data from sight translations,
other scholars seem to look for a sort of stream of consciousness, something
criticized by Kussmaul, who has modied the method of think-aloud protocols by
studying translators working together and recording their conversation exchanges.
Second, many authors seem to adopt a behaviorist stance in which they tend to
confuse what has been dened as a mental process with its behavioral symptoms.
Thus, we nd psycholinguistic categorizations of translation strategies which
include elements such as repeating the source- or target-language text segment
or the search in monolingual or bilingual dictionaries. It is no wonder, then,
that psycholinguistic approaches are mistrusted and that the community of
translation scholars thinks that translation strategies are still somewhere over the
rainbow.
134 RICARDO MUOZ MARTN

A pedagogical proposal

In spite of all what has been said, some advances may already be applied to the
practical teaching of translation. Sguinot (1991) nds that beginning students
apply inadequate translation strategies whereas excellent students analyze both
the source and the target language texts at various levels, focusing on meaning,
register, structure, cohesion, and so on. Thus, translation strategies can be
thought of as active at various levels, and not necessarily as the binary opposi-
tion between macro- and microstrategies proposed by Hnig (1991) and Kiraly
(1990). The possibility of teaching these strategies independently from the
fact that some people seem to have them from the start is a postulate of
translation theory and will not be further discussed. A second consequence to
draw from Sguinots observation is that some students have or acquire transla-
tion strategies, and that these strategies tend to consider larger segments of text,
or even the text as a whole. The main point here is not whether a translation
strategy applies to the way to translate a given text, but that similar problems
should be given similar solutions throughout the whole text (hence the divergent
denition of translation unit oered by Nord).
Lrscher (1991: 76 passim) correctly states that a translation strategy starts
with the identication of a problem and ends with its solution or with the belief
that it cannot be solved. Note that Lrscher does not necessarily imply that
identifying the problem is a conscious matter. Hnig (1991: 80) thinks that
rational and intuitive elements alternate, follow each other, and combine, in such
a way that the process as a whole is a sort of intuitive-rational hybrid, an opinion
apparently sustained by Kiraly (1990) as well. There are empirical reasons to
believe that intuitive behavior is the rst step of a single, general way of
translating, which would imply supercial processing (basically, searching for
stored potential solutions or models by analogy), that, when unsuccessful, would
lead to a deeper, more eort demanding way of mental processing, oriented to
problem solving (Muoz 1994, 1996; Mayoral and Muoz 1997). In fact, given
that rational strategies may become automated, the dierence between rational
and intuitive seems to blur over time. In a way, learning how to translate is a
process by which many conscious analysis and decision-taking processes become
interiorized.
Beaugrande (1978: 13) stated that translation strategies are the consequence
of combining textual constraints and communicative goals. Rastall (1994: 38f)
also implies that the factors which determine the target-language text play a role
in the choice of the translation strategy:
 TRANSLATION STRATEGIES 135

The selection of communication strategies is determined partly by the experi-

ence to be communicated, partly by the conventions of the language to be used,
partly by the intelligence and disposition of the speaker or writer and partly by
considerations of appropriateness (to the audience, to the situation in which the
communication takes place, to the type of text being created and so on).

From all these considerations, and given the current impossibility of directly
accessing mental processes, a translation strategy could be substituted by a
pedagogical framework consisting of the analysis of the potentially relevant
elements in the source text and in the communicative context postulated for the
target text, an analysis centered in the production of potential solutions. This
framework makes use of published original texts and their translations, since the
study of regularities in the products can shed light on translation norms (Mason
1994: 65). Here is where the concept of space of the problem becomes
relevant. One of the ways of studying complex problem-solving is drafting a
diagram that contains the variables which play a role in a given problem, which
are matched with potential solutions. The goal of this approach is to sensitize
students towards developing and interiorizing their own translation strategies,
which remain unknown, as in the black box model. In other words, students can
be taught to focus on certain textual and contextual constraints, to develop
potentially optimal solutions corresponding to the constraints they have identi-
ed, and to establish an order of precedence in those elements. The result of this
process is a decision tree, the rst attempt of which can be seen in Diagram 1.
The way the decision tree is developed is as follows:
(a) A translation problem is tentatively dened.
(b) A statistically signicant number of published problem instances and their
translations is collected.
(c) The problem denition is rened and statistical analyses are made to
determine the absolute and relative frequency of each identied subtype of
problem and also of their published solutions.
(d) TL communicative contexts are analyzed to establish hypotheses about the
constraints that have determined the choice of one or other solution.
(e) TL communicative contexts constraints are ordered, according to their
precedence.
(f) SL segments and potential solutions are matched, depending on their
frequency in published texts.
(g) The validity of the decision tree arrived at is checked with a statistically
signicant number of new problem instances.
This approach is not free from problems. For instance, it can only be applied to
well dened problems, which is painfully obvious in the early attempt of
136 RICARDO MUOZ MARTN

is the SL segment
irrelevant yes
omission
as a whole?
(a)
no (1)

is there a precedent
for translating yes Established
the SL segment? translation
(b)
no (2)

Does the TL and

culture have means yes functional
to cause the same formulation
effect?
(c)
no (3)

Is the potential
TL solution yes
paraphrase
free of restrictions?
(d)
no (4)

Can the SL segment

Be understood yes
Loan/Calque
In the TL text?
(e)
no (5)

Is it important
to know the SL yes
Combine 4+5
segment anyway?
(f)
no (6)

Create new

(7)

Diagram 1: Mayoral and Muozs (1997) model of translation strategy for culturally
marked text segments
</TARGET "mun">

TRANSLATION STRATEGIES 137

decision tree proposed in Diagram 1. Diagram 2 shows a more solid decision tree
developed by a Granadas Translation School former student, Eva Mara del
guila, for her graduation nal project.
There is only a handful of translation problems so well dened as institu-
tional names and, consequently, the validity of the approach can be questioned.
The same can be said about the binary structure chosen in both models, since
thought is not necessarily binary. But this approach is not trying to pin down a
mental translation strategy. Instead, the diagram is just an attempt to formalize
the variables which have an inuence on the felicity of the matching potential
solutions. In any case, what is really valuable is not the result of the task,
namely, the decision tree, but the perspective gained by the students in their
attempts to formalize such decision trees.
138 RICARDO MUOZ MARTN

Is there an official translation

for the SL segment? yes Established Banco Europeo de Reconstruccin y
translation Desarrollo (BERD)
(a) European Bank for Reconstruction and
No (1) Development (EBRD)

Are there words with similar

meanings in the TL? yes Banco de Espaa
Calque Bank of Spain
(b)
No (2)

Are there TL words that can

formally imitate the SL yes Cognate Cdigo Penal
segment? Criminal Code
(c)
No (3)

Is the SL segment
transparent or yes Banco Bilbao Vizcaya
descriptive? Transcription Banco Bilbao Vizcaya
(d)
No (4)

Is the SL segment an
essential element for yes Transcription Cortes Generales
understanding the TL text? + explanation Cortes Generales, Spains Parliament
(e)
No (5)

Are there any time or space

restrictions for the TL yes Conceptual
segment? formulation
(paraphrase) Estatuto de Autonoma
(f) The basic institutional rule which creates
no (6) each Autonomous Community, Defines its
territory, etc.

Are there concepts in the TL

culture which permit the yes Functional
creation of a TL segment with Comisin Nacional del Mercado de
the same effect as the SL
formulation
Valores
segment? Spanish Securities and Exchange
Commission
(g) 7)
no
omission

(8)

Diagram 2: Del guilas (1997) Model of translation strategy for Spanish English
translation of institutional names of economic and juridic institutions
<TARGET "p3"
</TARGET "p3">DOCINFO

AUTHOR ""

TITLE "Investigating Translation and Ideology"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

S III

Investigating Translation and Ideology

<TARGET "mal" DOCINFO

AUTHOR "Joaquim Mallafr"

TITLE "Language Models and Catalan Translation"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 14

Language Models and Catalan Translation

Joaquim Mallafr
Universitat Rovira i Virgili, Reus-Tarragona

The acquisition of Catalan for translation

The process of how a translator acquires a language appropriate for translation

is interesting because it transcends the individual, taking place, as it does in a
shared cultural framework, whether in an intra-linguistic or an inter-linguistic
community (in both cases the language models with any real inuence are
limited). In Catalonia, where Catalan has long struggled to survive under the
shadow of Spanish, there has been some discussion about language models for
translations into Catalan. Under Francos dictatorial regime, Catalan was banned
in schools and the Catalan cultural tradition was eclipsed by the dominant
culture. Thus, the codication and the style of the language have been transmit-
ted by an intelligentsia that was very dierent from those found in more
established societies. For example, the prestigious Catalan poet and translator,
Josep Carner, was a symbol of a particular cultural movement at the beginning
of the twentieth century, the Noucentisme. This movement belonged to a
historical period that could not evolve normally and had no natural continuity.
Due to this rupture caused by the Civil War and its aftermath, a present-day
reader may nd Carners style strange. The same is true of the style of Carles
Riba, whose translation of The Odyssey is a model for what some would have
wished to be a standard model of cultured Catalan that was brutally frustrated,
according to Parcerisas (1997: 471). Parcerisas includes it amongst those works
that have marked and conditioned how translation has been understood (485).
142 JOAQUIM MALLAFR

Literary canons and linguistic norms

This understanding, the fruit of a literary school and certain political conditions,
has been combined with linguistic norms. Together, they have certainly achieved
undeniably ecient results. However, there are critics that blame both the
literary tradition and the strictly linguistic regulations for creating serious
limitations or imposing a single model, that act as a dead weight round the necks
of creative writers and translators.
This is the position of Pericay and Toutain (1996), who provide arguments
and data, with reference to prose in particular. Their criticism is both skilled and
useful and it makes sense to me as a translator; which is justication enough.
Nevertheless, they do not so much criticise the model itself as models they
consider to be poor imitations. So, despite their contribution, I would question
some over-simplications, their choice of certain examples and partial criticisms
of some translators.
In my opinion, there is a tendency in Pericay and Toutain more exagger-
ated in other less well documented critics to group together both formal
creations and expressions widely used by Catalan speakers as articial creations
belonging to the Noucentisme, and as such to be excluded. This is hardly
justied, any more than would be the exclusion of Joyces use of the Dubliners
language, or his formal expressions based on Tomas de Aquinas. Formal words,
such as, tocom, llur, ensems, de bon antuvi, perfectly legitimate of their kind
should not be classed together with colloquial expressions, such as, abellir,
virosta or enguany, that I, and others, have learned in everyday use, not from
noucentist texts.

TT norms, the individual translator and the ST author

In any case, I think too much weight has been attributed to a single tradition,
even though it has become the principal, ocial norm. All language models,
authors, or translators transcend the ocial norm, which is no more than a basic
guide, some unied rules in the varied and unpredictable game of literary
creation. We should not forget that the rst obligation of the responsible
translator is to the model imposed by the author they are translating, his thought,
language and style.
Therefore, I think it may be of interest to consider my own model of
language, which was, of course, nurtured by a linguistic community. But it was
rst structured by conversations in the family and the society, and only later, by
 LANGUAGE MODELS AND CATALAN TRANSLATION 143

more formal levels of language and a variety of cultural components, some of

which transcended Catalonia. These later inuences, with some additions and
reservations, provide literary and stylistic models and more or less shared norms
that make up a common, coherent language, and, as such, can be transmitted and
exported. These norms and models are superimposed on and enrich the individual
linguistic and literary biography. From this acquired foundation we develop our
own creativity, or we dedicate ourselves to the service of an alien creation when
we decide to translate.
Through experience, we achieve a basic style in accord with our personality
and our epoch. The mechanism is essentially the same in any language where
there are writers with whom we nd points of contact that bring them closer to
us. They become translatable if we have the necessary capacities: reading,
literary criticism and the adequate expression to reproduce them in our language
(Bush: 1997). Only by reading and recreating a work can translator and readers
make it theirs, with the admiration of one who sees that the alien environment,
its style, context and musicality can be expressed in their own language. Perhaps,
this is why the advantages of reading a book in a language one understands
compensates and justies the inevitable disadvantages of translation, as has been
armed by writers from Goethe to Joan Sales and Garca Mrquez.
If I try to outline how I have acquired a language that could be developed
into a literary language, perhaps this will reveal elements that explain both my
model, the choice of certain solutions, and analogies with my contemporaries,
from which has emerged a common historical-linguistic base. Obviously, the
knowledge of some languages and cultures and not others, and personal tastes or
anities explain certain preferences. Beyond personal dierences, it is useful
to contrast the experiences of an epoch that may possibly be systematised for
teaching purposes.
Dierent genres permit dierent translation approaches: the translation of
a novel can keep very close to the original; the translation of a play will depend
on whether it is to be read as a text or performed on the stage; the translation of
poetry tends to adaptation, or the personal poetic projection of the translator.
Even in poetry, however dicult the task, there are some translators who are
determined to re-express the content of the poems, submitting themselves to the
form used by the author. Nevertheless, despite inevitable failures, it is usually the
source text that lays down the rules of the translation: the long sentences of 18th
century prose, the telegraphic density of certain 20th century texts, the registers,
technicalities or allusions of others. I may doubt whether to translate the English
you by vos, vost or tu, or an exact lexical equivalent may not be possible
because it is unknown by the general public. I may be led astray by false friends
144 JOAQUIM MALLAFR

due to insucient knowledge of the source language, or by interference due to

insecurity in the target language. But the identity of the work will not be lost if
we are able to nd literal or equivalent coincidences with the source text.

Catalan at home and Spanish at school

My experience of language began with Catalan, with my family, was widened

through contacts with neighbours, country folk, priests, jokes told in cafs, the
colourful language of the markets, summer holidays. This childish substratum,
was almost erased by schooling in another language, Spanish, with alien models,
that were useful in so far as they represented universals. My Catalan was
reactivated thanks to translation. When translating Fielding and Sterne, Joyce and
Beckett, solutions emerged that I had heard from my elder brother, in the market,
at snooker games and in the slang used by gypsies. This is where I found ready-
made solutions that coincided with the authors expressions: Shut your eyes and
open your mouth Obre la boca i tanca els ulls; Touch the spot Tocar el
voraviu; As uncertain as a childs bottom Insegur com el cul del jaumet;
Both ends meet Els extrems es toquen; Collar the leather, youngun
Enxoma la bimba, baranda.
Apart from textbooks in Spanish, I learnt to read from old Catalan
childrens magazines, where I met the humour of Guillem dOlor, or of Josep
M. Folch i Torres in the Pgines viscudes and in the Biblioteca Patufet. Folch i
Torres was inuenced by Dickens, but he was also inuenced by a literary
apprenticeship of books and plays that was shared by many other children. Folch
i Torres adopted the norms established by Fabra (the rst modern Catalan
grammarian and lexicographer) and this helped to spread a generally accepted
external model. Even before I started to learn Catalan grammar, by reading
Patufets, I knew enough to nd a name for Mr Allfours Mr Quatrepotes;
Lord Walkup of Walkup on Eggs Lord Trepig de Tretjous; or Kissarcius
Besaculis.
Curiously enough the catechism classes of my childhood were in Catalan
and in the Church I was taught a more abstract language. The prayers of the
Catholic liturgy gave me the exact translation of some of the references in
Ulysses, that were more immediately familiar to me than to an English speaking
Protestant reader, for example.
I did not have to go to school to learn, when, outside the classroom, I could
hear popular rhymes, fragments of popular poetry by Catalan writers, such as
Pitarra or Verdaguer. I got used to more sophisticated language through the plays
 LANGUAGE MODELS AND CATALAN TRANSLATION 145

performed on Sundays at our local theatre, lOrfe Reusenc. My literary educa-

tion continued with my reading, so when I had to translate Dear dirty Dublin,
I chose to echo the Catalan poet Salvador Espriu, Pobre, bruta, estimada
Dublin. My translation of Return of Bloom echoed another Catalan poet,
ngel Guimer, Bloom que torna.
In some cases the Catalan translation tradition provided solutions for the
frequent quotations found in English authors. For quotations of Shakespeare, I
could choose the most appropriate version from the translations of Morera i
Galicia, Sagarra and Oliva. When translating Tom Jones, I used Miquel Dols
version of the Eneas when Fielding had used Drydens; and I used Josep M.
Lloveras version of Horace when Fielding used Franciss. In Tristram Shandy,
when Sterne quoted Rabelais, I used the translation by Miguel Angel Snchez
Frriz.
Other problems with Tristram Shandy were solved by a horse breeder, who
conrmed that a horse that was spavined or greazed, tenia esparavanys o
aigerols. Documentation in philosophy, medicine, building or military
fortications conrmed my translations of some of the archaic expressions that
abound in Sterne, such as, homunculus homuncle, radical moisture
humit radical, ravelins revellins and fause-brays falsabragues.
My knowledge of the rules meant that I had at my disposal an ecient,
homogeneous and organised language model, but I rarely felt it was a restriction.
I had no need to reject it in order to feel free to use words that are not in the
dictionary, such as cavallbatallvol for Tristram Shandys hobby-horsical, or
marcavent when searching for an equivalent register for Becketts windgauge.
It is questionable to what extent younger Catalan readers can assimilate my
linguistic experience, and this is a serious problem. However, I console myself
by thinking that English readers today are equally distant from the linguistic
experience of many of their own authors. It is often forgotten that the original
text also ages. Of course, the classics resist the ageing process in so far as they
remain familiar for their public, through the force of tradition they are read again
and again. When I translate a classic I travel to the land of the original text and
I describe it in Catalan to a reader of my language. The success of my transla-
tion will depend on the degree to which a Catalan reader can speak to an English
reader about the same work, the same characters, about a comparable style that
acts as a vehicle for the work, and each one can draw closer, sharing a common
world from the perspectives of their respective languages.
Of course, cultural dierences often mean that exact translations are not
viable, expressions that are clear to the reader of the original text are obscure to
the reader of the translation. But what about passages that are obscure in the
146 JOAQUIM MALLAFR

original text, a text that has aged for readers who have lost touch with their own
tradition? The role of the language authorities is especially delicate in these
circumstances. Perhaps the translator should concentrate on the dicult task of
developing good taste as Alln (1997) suggested. In fact, this is what Carles Riba
suggested in his preface to the second edition of Fabras dictionary (1954) Good
taste: this is a sentiment that is not hereditary; but it is possible to transmit a set
of criteria that will improve the education of each and every one.
If the author can be creative, the translator can only be creative to reect
what the author says. This reection has to be as natural or articial as the
original. Pym (1997: 51) was quite right to suggest we should encourage the idea
that one should translate as one would wish to be translated. This seems to me
to be a fundamental requirement.
If language is to be authoritative it is not enough for it to be real, but it has
to appear real. The artice that is constructed using the language of the commu-
nity and the individual should be plausible, natural for the reader. With time, it
may inuence the readers taste, or, he may even use the language of certain
authors. This was what happened with a famous author of crime novels. On
being told that he seemed very familiar with the underworld that he was able to
reproduce with such realism in his novels, he replied that, in fact, it was the cops
and robbers who were familiar with his novels. The translator also has to
combine the reality and the seeming reality of the original language and repro-
duce it with the tools of the target language, starting from a landscape that is
alien in space, time and culture.

Catalan literary translation in the eighties: A study of two collections

After the Spanish Civil War, translation into Catalan practically disappeared, due
to the political repression. The sixties marked the rst major recovery, with the
successes and failures described by Broch (1991: 189) and Vallverd (1987: 103).
In the eighties there was a second major attempt to publish translations in
Catalan. This was a bigger movement and included initiatives to incorporate
dierent genres, such as the collections of the MOLU (Les Millors Obres de la
Literatura Universal), MOLU XX (Les Millors Obres de la Literatura Universal
segle XX), Poesia del segle XX. Textos loscs, Clssics del pensament modern.
This was possible, in part, due to institutional support and aid from savings
banks. In this period the most important works of Joyce, Proust, Kafka,
Maiakovsky and Henry Miller were also translated and rigorous criteria were
used to translate Boccaccio, Shakespeare, La Fontaine, Sterne and Melville.
 LANGUAGE MODELS AND CATALAN TRANSLATION 147

In this article, two of these collections will be described: MOLU and MOLU
XX, directed by Joaquim Molas and assessed by J.M. Castellet and Pere
Gimferrer. The rst collection included 50 books published between 1981 and
1986. The rst 50 books of the second collection were published between 1986
and 1990. The collections are interesting, not only from the point of view of the
translated texts themselves, but also because of the sociological context in which
they were produced. They provide us not only with the translations, but also with
a whole series of other criteria that can throw light on translation studies. As
Parcerisas (1997) suggested, these criteria are commercial, political, institutional
and historical, and they may all inuence the language model, or models, used.
For example, these two collections cannot be explained without the institutional
aid from the Servei del Llibre of the Catalan Ministry of Culture and the
combined eort of the publishers, Edicions 62, and la Caixa savings bank.
It is very dicult to speak of a single language model for these collections,
given the variety of genres, source text languages, authors and translators, as the
following description shows.
The predominant genre was the novel, taking the place of poetry, which had
been very important in Catalan translation in the past. Of the 100 books studied
in the two collections, 73 were ction (34 and 39 in MOLU and MOLU XX,
respectively), 14 were poetry (9 and 5) and 13 were drama (7 and 6).
The choice of source text languages shows that our vision of universal
literature remained linked to the Western world. There are only 12 languages:
English 32 (13 and 19), French 21 (13 and 8), Italian 15 (6 and 9), German 15
(7 and 8), Russian 9 (7 and 2), Galician and Portuguese 3 (1 and 2). In the rst
collection, MOLU, there are single examples of translations from Latin, Proven-
al, Swedish and Norwegian, and in the second collection, MOLU XX, there are
single examples of translations from Polish and Modern Greek.
The fact that 25% of the translators were also writers proves the dedication
of many Catalan authors to translation. Leaving aside the translators of antholo-
gies that included the work of several translators, many of the translators/authors
are best known for their own work as writers. 13 of the books were translations
by Josep Carner, Manuel de Pedrolo, Josep M. de Sagarra, etc., which had been
published previously, but revised for these collections. 12 were translations by 9
modern authors, Pere Gimferrer, Miguel Mart i Pol, Quim Monz, etc., prepared
especially for these collections. However, it is interesting to note the growing
number of translators who are not creative writers in their own right, some of
them are academics who have specialised in the subject or the author of the book
they have translated. It is signicant that 48 of the translators also wrote the
introductions to the books they translated.
148 JOAQUIM MALLAFR

Translation criteria expressed in introductions and footnotes

On the whole, the translation criteria used in these collections show that the
books are aimed at the general reader rather than the academic. As Josep
Murgades warned when presenting the translation of Goethe, A translation is
not supposed to be a critical edition, (MOLU 41: 11). J.M. Gell did not attempt
to represent Gogols linguistic variations because he considered them to be
more suitable for an edition addressed at specialists and academics than for an
edition addressed at the general public. Despite this non-academic approach, the
collections are full of translators notes, some of which illustrate interesting
theoretical considerations.
There are dierent opinions about translators notes. Some claim they have
no place in an artistic translation, as the translation has to be solved in the text
alone, others use them generously, or keep them down to an absolute minimum,
using them when they are absolutely essential. It would be useful to study the
criteria used and to try to systematise this area where anarchy often reigns, not
only in Catalan translations, but elsewhere as well. The MOLU collections allow
a great deal of freedom for individual initiatives, although there are some
attempts to systematise. In the second collection the translators notes are always
explicit and regularly marked with an asterisk. However, in the rst collection
there are notable irregularities.
In the rst place, the translators notes in the footnotes are not always really
translators notes. Although it is true that the limits are hard to establish, I
believe there are some that are clear. These are: explanations of word play,
borrowing (by the author or the translator), the translation or not of quotations,
verses included in the original language in the text and translated in a footnote
or viceversa, the recognition of earlier translations used by the translator, specic
allusions or references that would not be directly understood by the translation
reader and require translation decisions. There are other notes that do not aect
the translation and are really editorial notes, for example, notes that clarify who
a person is, or what an institution is, or encyclopaedic notes. Secondly, it is not
always clear if the notes are from the original edition or by the translator. It
would be useful to make the distinction. Thirdly, notes that seem to correspond
to specic translation problems are not always marked as being by the translator.
Discussion of translation criteria can also be found in the introduction or in
a preliminary translators note. Many of the theoretical considerations are
concerned with faithfulness, in one or other of the senses of the notion, although
none would defend the use of archaic language to reect the original. As far as
the language is concerned, we have tried to maintain, as far as was possible, the
 LANGUAGE MODELS AND CATALAN TRANSLATION 149

unpolished and sometimes confusing style of the original. We have not altered
the minimal change in style between the narrative and the dialogues, a change
that is hardly noticeable in the original (Desclot, MOLU 11: 7).
Faithfulness to the source text may make the translator consider a literal
translation. I have preferred to run the risk that my translation may sound rigid
because it is so literal, rather than risk the debilitating loss of an overly free
translation, wrote Murgades about his translation of Goethe (MOLU 41: 10).
Joan Casas tended towards a literal respect for the original in his translations,
despite the diculties this involved, so that the deformation is minimal in a
process that is in itself extremely traumatic, a betrayal, the process of passing
from the mould of one language to another. No concessions should be
permitted under the excuse of bringing the original up to date, a concept which
is, after all, no more than a subjective impression (MOLU 42: 11). The resulting
syntax could be complicated, as Joan Casas himself showed with an example of
his translation of a verbal structure by Saint-Simon, es va deixar arrossegar a
gosar fer saber que desitjava (MOLU 37: 10). Casas also translated Bassani,
whose precise language was a sort of spirit that Bassanis own prose recom-
mended to the translator, who soon found himself submerged in the same feeling,
in the certainty that to alter the sense of an adjective, the rhythm of a sentence,
could open an irreparable breach in the soundness of the structure (MOLU XX
33: 10). But when Casas had to translate Cline, he opted for a possible
translation, given the poverty of non-standard urban registers in Catalan. The
lexical limitations were compensated by the use of oral syntax, rhythmical
sentences and set expressions that were not necessarily literal translations
(MOLU XX 19: 1112).
Thus, when a literal translation would be in conict with faithfulness to the
original text, because the literal translation would be incomprehensible, the
translators opt for a certain degree of freedom, for example the translations of
Svevo and Beckett. The same approach is obviously inevitable when translating
poetry. As Miralles remarked, translated poetry has to be read not only as a text
that tries to reect the original correctly, but also as a text with ambitions to
achieve a poetic result in the target language (MOLU XX 25: 16). The transla-
tions of the anthologies vary according to the translators sensitivity and skill and
do not always reect the original poets sensitivity and skill. According to Alain
Verjat, this approximation is perfectly valid because the perfect translation does
not exist (MOLU 44: 18).
The comments made by the translator of Moby Dick, M. A. Oliver, are
particularly interesting in this context. She cites the informants who helped her
to nd Catalan words used in navigation and how to adapt the vocabulary related
150 JOAQUIM MALLAFR

to whales, animals and plants, the set expressions and registers, to identify the
Biblical references or help her to understand certain dicult constructions in
English (MOLU 30: 12).
The question of language models and Catalan translation was also some-
times raised in the introduction or in the translators preliminary note. Dierenc-
es on this point can be seen between the new translations and the 13 books that
were re-editions of earlier translations within the tradition of Noucentisme. In the
earlier translations there are references to the inuence of translation on Catalan
literature and language. For example, in 1921 Carles Riba talked of patriotic
egoism when he saluted the translation of Molire into Catalan by Josep Carner.
Ribas appraisal was reproduced as a foreword to the MOLU edition, This is
our hope, that the Catalan word, following the path laid out by the admirable
translator, may reach its nal and most dicult triumph: to reign supreme in
comedy. We need a language that is independent of Rambouillet Palace, a
language both t for the ear of the Great King, and suited for the good bour-
geois, whether he believes himself to be a gentleman or not (MOLU 2: 6).
However, even a contemporary writer, like Gimferrer, translator of part of
Lducation Sentimentale, acknowledged the inuence of the Catalan style forged
by both translators and writers, and the impact of translation on his own writing.
He stressed his obsession with the prose that was the result of Carners
translations of Dickens, the prose of Josep Vicen Foix and Josep Pla. I gained
another verbal obsession from submerging myself in translating Flaubert and this
prevented me from writing myself for a year and a half (MOLU 20: 7).
However, in the books translated after the 1960s, few translators discussed
language models explicitly. It seems as if there was more condence in the
normative and it was not questioned. These rules in the target language and a
more or less shared cultural and linguistic experience, such as the one described
in the rst part of this paper, provided a language model for translation. Never-
theless, we should not forget the importance of each source text in determining
the translators use of grammar, syntax, register and the personal rhythm of the
work of art.

Conclusions

The analysis of a collection of translations like the one we have described

provides signicant data about translation criteria in our country at the end of an
era. At this time, empirical research can incorporate contributions from the
philological and the linguistic stages, as well as the most recent tendencies in
</TARGET "mal">

LANGUAGE MODELS AND CATALAN TRANSLATION 151

translation studies within the framework of communication theory. The present

panorama with regard to translators has been summed up by Narcis Comadira.
Translation is possible, even though there is no such thing as a polyvalent
translator, that is to say, not everyone can translate everything. There are
translators who are traitors and translators who are incompetent. The former
falsify because of their pride, and the latter because of their ineptitude. However,
there are good translators: these are the ones with the adequate technical
knowledge who are able to silence their own voices so that, in symbiosis with
the other, the voice of the poet they are translating can be heard (MOLU 40:
1112).

Translated by Allison Beeby

<TARGET "iza" DOCINFO

AUTHOR "Natalia Izard"

TITLE "Dubbing for Catalan Television"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 15

Dubbing for Catalan Television

The Acceptable Translation

Natlia Izard
Universitat Pompeu Fabra, Barcelona

Televisi de Catalunya (hereinafter TVC) is a public television station broadcast-

ing in the Catalan language, with almost exclusively dubbed products. Between
1994 and 1996 TVC broadcast the French sitcom Hlne et les garons, which
was dubbed into Catalan with the title Helena, quina canya!. It was a series of
130 half-hour episodes for teenagers, and was shown daily in the late afternoon.
In the Catalan version this show presents two translation characteristics which
are quite unusual and worth analyzing. The rst characteristic is an adaptation of
the cultural referents, from the source culture, to the target culture. The second
characteristic is a linguistic adaptation: a very formal and conservative language
in the French version becomes colloquial and peppered with slang expressions in
Catalan. In this paper we will describe both kinds of adaptations and we will try
to establish a possible explanation for them. We will nally compare this
particular dubbing with other TVC dubbings, with other dubbings in Spain, and
with some dubbings in Europe.
The cultural adaptation is the transformation of most of the references to the
French culture, into references to Catalan culture. Here are some examples:
Names are Catalanized:
Cathy becomes Lali; Bndicte, Magda; Nicolas, Adri, etc.
French people become Catalan people:

(1)
ANGELINA: ben non Il est tout petit
Tout chtif Tout malingre
Finalement Je me suis laisse
avoir parce que ctait le premier
petit frenchie que je recontrais
ANGELINA: Va, dona, s poca
cosa. s petit i esquidet. La
veritat, si mhe deixat enredar s
perqu era el primer catalanet
que trobava.
154 NATALIA IZARD

ANGELINA: No, hes rather ANGELINA: No, hes rather

short, rather ugly. But I liked him short, rather ugly. But I liked him
because he was the rst little because he was the rst little
Frenchy I had met. Catalan I had met.
(Fragment from episode 44)
the last highschool year, le Bac, becomes its equivalent in the Catalan
school system, COU:

(2) HELENE: Bof, a doit pas tre HELENA: Bah, no deu ser ms fumut
plus dur que le Bac quand que aprovar el COU! Has de pencar i
mme! Ya qu travailler, cest prou!
tout
(Fragment from episode 12)
champagne becomes its Catalan equivalent, cava:
(3) JOHANNA: Finalement cest JOANA: Finalment, est b que sigui
bien que a tombe le jour de el dia del meu aniversari. Perqu
mon anniversaire comme a men recordar! I si prenem una mica
je men souviendrai on boit ms de cava?
un peu de champagne !!???
(Fragment from episode 28)
Alfredos, the cafeteria, becomes Can Pere or la granja (the cafeteria).
Other references to French culture are not Catalanized but rather neutralized. In
other words, they loose their Frenchness. In most of the cases the neutraliza-
tion is done by means of hyperonyms:
baguette becomes barreta (roll):
(4) JOHANNA: Moi?! Arrter la JOANA: Ah-ha! Voleu que deixi les
baguette?! Mais a va pas les barretes? No, ni parlar-ne! No,
lles! Pourquoi vous croyez que no, no, no Per qu us penseu que
je suis venue en frane, hein?! he vingut a Frana, pel pa! O sigui
cause du pain! Alors si on que si mel treieu
menlve a
(Fragment from episode 13)
the tour de France becomes the volta ciclista (cycle race)
the university looses its French name: Paris Nord is translated as la
facultat (college):
 DUBBING FOR CATALAN TELEVISION 155

(5) JOHANNA: Des invitations JOANA: Invitacions per la nit de la

pour la nuit de la danse de dansa de la facultat! La Natlia me
Paris Nord!!. Cest Nathalie qui les ha donat!
me les a donnes.
(Fragment from episode 17)
In other cases the neutralization is done by means of an adaptation to a culture
(neither French nor Catalan) that acts as a neutral culture:

JOHANNA: Non, Jean-Pierre JOANA: No, en Harrison Ford i en

Foucault et Patrick Sabatier. Michael Douglas.
(Fragment from episode 28)
Finally, there is a third group of references that remains the same as in the
original. The main one is France, which always remains France:
HELENE: Cest pas mal mais je HELENA : Est b, per em mola
prfre la France! ms Frana!
(Fragment from episode 1)
The cultural adaptation, although obvious and deliberate, is irregular and
inconsistent: France is Frana, but les franais are els catalanets; in one
episode they talk about francs, and in the next one, about pessetas. The currency
is the cultural item which presents the highest level of inconsistency: In some
cases the francs are maintained, in some others they become pessetas, and in yet
some others they are transformed into a neutral exchange value, like in the
expression Je parie 100 francs, which is translated as mhi jugo un sopar (I bet
a dinner that), in episode 55.
The linguistic adaptation consists of informalizing the register. The French
version used a very formal variety of the language, which actually matched the
intended image of the characters: actor-models who dress impeccably, university
students whose only worries are, for women, to please men, and for men, to
please women, in addition to being concerned about playing music. In the
Catalan dubbed version the original standard and formal register is transformed
into a Catalan colloquial register, peppered with expressions of teenage slang: les
garons become els tios (the guys), contrarier is emprenyar (something
like to fuck up, sexual connotation included), vous tes trs mignones is esteu
molt bones (which has a sexual component), mes parents is translated as els
meus vells (my old folks), tu exagres, as thas passat (a very typical
expression of young slang), etc.
156 NATALIA IZARD

In fact, both adaptations have a common objective: to bring the audovisual

product closer to the linguistic and cultural environment of the target audience.
It is an acceptable translation (acceptable being a term created by Gideon
Toury (1995), to describe those translations which come closer to the target
language and culture, as opposed to adequate translations, which are those
which come closer to the source language and culture).
In the case of Helena, quina canya! the acceptability responds to the
reason why TVC was created: to promote the use of the Catalan language. After
centuries of being a minority language, dominated by Spanish, the language of
the central state, Catalan is now being promoted by the Catalan federal govern-
ment. Catalan public television is one of the more important means for this
promotion. According to Marta Trias, the assistant to TVCs director of program-
ming, an acceptable translation would be better accepted by the Catalan audience,
thus furthering the promotion of the language. Hlne et les garons presented a
wonderful opportunity to extend the promotion of the language to teenagers, a
portion of the audience which was still pending.
In a general (not systematic) observation of other TVC dubbings, we also
nd a tendency towards culturally and linguistically acceptable translations. The
following is an example of a cultural adaptation:
In a Dutch documentary, shown in TVC with the title of Conixer Van Gaal
(Meeting Van Gaal) we nd an example of an extremely acceptable transla-
tion. Van Gaal, the Dutch coach of F.C. Barcelona, was explaining to the
interviewers the dierence between a caf con leche and a cortado. In Dutch he
said: Een cortado is een koe met een klein beetje melk. En dat is een caf con
leche. The literal translation would have been: A cortado is a cup of coee
with a little bit of milk. And this is a caf con leche. It would have shown the
attitude of Van Gaal with the interviewers: that of a person in a foreign country,
who explains to his fellow countrymen a cultural feature of his new country. The
Catalan dubbing, instead, said: Jo sempre prenc un tallat. I aix s un caf amb
llet. (I always drink a cortado. And this is a caf con leche). The translation
chosen by TVC eliminates all marks that express the foreign origin of the
discourse.
As in the case of Helena, quina canya!, TVC also achieves acceptability
through the use of the language by producing a variety of the Catalan language
which is very genuinely Catalan. This is an unusual translation norm, as it
reverses a very generalized norm in translation: namely, that translations use a
less genuine and more standardized variety of the language than originals. For
example:
In Retorn al parads (dubbed version of Return to Paradise) the main
 DUBBING FOR CATALAN TELEVISION 157

character, an American soldier hiding in an island in the Pacic, realizes that his
radio is out of order and yells: Punyeta, ja ha tornat a fer ga! Renoi!. In Acci
judicial (dubbed version of Class Action) we can hear expressions like Ja s que
no s peix al cove. In both cases we are dealing with expressions that are
extremely idiomatic, that the users of the language immediately identify with the
Catalan language. They are the kind of idioms the audience would expect in an
original Catalan product, and not in a dubbed product, which tends to use a more
standard level of language. It is as if these translations were not presented as
translations, but as Catalan products. In fact, in the TVC stylebook for dubbing,
we nd a recommendation of this kind: A good translation would be that which
looks like it has been conceived in Catalan (Televisi de Catalunya 1997: 14;
authors translation).
Using this kind of genuine language means moving away from literal
translations. The introduction to a dubbing stylesheet, published internally by
TVC in 1993, says:
We want to contribute to the task of the translators and the dubbing consul-
tants, providing them with alternatives () to literal translations -especially
from English- which distort Catalan as a support of dubbed products ().
Bearing in mind, though, that the adaptation should not loose the original tone
and that a common aim of the professionals of the language is to avoid
expressive poverty through a more open and richer use of the means that are
available to give the dialogs connotation and make them closer and more
credible to the audience who has chosen to watch TVC (Comissi de Norma-
litzaci Lingstica, 1993: 5; authors translation).

This stylesheet was revised and became a stylebook, published by a commercial

company four years later. The introduction to the book insists on the notion of
credibility: If a dubbing is not credible it will be laughable and boring. (Tele-
visi de Catalunya 1997: 15). The stylebook is basically a list of literal transla-
tions, and their recommended equivalents. So, their basic norm for acquiring
credibility is, on the one hand, avoiding literal translations, and using expressive,
genuine Catalan, on the other. For the English Shut your mouth!, they recom-
mend the Catalan Calla! (Televisi de Catalunya 1997: 84) (and not Cierra la
boca! its literal translation, which is a recurrent expression in Spanish dubbings).
Instead of Oh, no! the stylebook recommends No fotis, Carai!, Ostres!,
Du nhi do!, Qu dius?, I ara! o S, home! (Televisi de Catalunya
1997: 53). All of them are idioms of the Catalan language.
Our general observations are conrmed by the literature on TVCs use of
the language. The research group led by Margarida Bassols (Universitat Aut-
noma de Barcelona) states that TVCs policy is to bring the originals cultural
158 NATALIA IZARD

context closer to that of the audience, especially in childrens programmes: The

translated scripts depart from the original in order [] to make the characters
closer to the viewers social context: adapt them to the viewers knowledge.
(Bassols et al. 1995: 411). They also note that translations very often use Catalan
idioms in order to achieve expressivity (Bassols et al. 1995b).
Rosa Agost (1995) compares the Catalan and the Spanish dubbed versions
of the French show Premiers baisers (De qu vas? in Catalan and Primeros besos
in Spanish). Her conclusion about the translation strategies is: [] the general
strategy of the Catalan version is to adapt the French series to the culture and
language of destination. [] If the strategy of the Catalan version is to Catalan-
ize the series, the Spanish version, as I will try to show, could be considered a
more literal version. (1995: 187)
Therefore, we can conclude that Helena, quina canya! is a case of extreme
acceptability, but it is actually not so dierent from the rest of TVC dubbings,
which also tend to acceptability.
We have also compared TVC dubbings with dubbings for television into
Spanish. A general (non systematic) overview leads us to conclude that TVC
dubbings tend to be more acceptable than the others. For example, many TVC
dubbings translate the names of the characters into Catalan, as we have seen in
the case of Helena, quina canya!. This is done especially in programs for
children and teenagers. The popular childrens British show Teletubbies is
dubbed into Catalan with the names of the characters adapted to Catalan. The
names are indeed very Catalan names: Jordi, Pep, Neus, etc. But in the Spanish
dubbed version of the same show, the characters keep their English names:
Sophie, Luke, Eddy, etc. The translator of the Catalan version informed us that
she was instructed to adapt the names, in order to achieve a better identication
of the target children with the program. Generally, in dubbings into Spanish (for
television as well as for the big screen), proper names are not translated.
In Spanish, translations tend to be more literal, not so adapted to the form
and the use of the Spanish language. There are not nearly so many idioms as in
Catalan dubbings. On the contrary, Spanish dubbings are known for using
expressions which are literal translations of the English or the original languages
expressions. It is the case of phrases like Dame la jodida pistola, literal
translation of the English Give me the fucking gun. The general tendency is
not to produce acceptable translations, but, on the contrary, adequate translations.
In other countries, the tendency is similar. In German TV dubbings, for
example, not only the names are kept in the original form, but also the attributes
that precede the names: If the lm is Spanish, it is not Frau Lpez, but
Seora Lpez. This strategy aims at keeping the foreign avour of the
 DUBBING FOR CATALAN TELEVISION 159

production. German dubbings also keep vocative or appellative expressions in the

original language: Sir, Madam, Monsieur, chri, adis, buenas noches, etc. This
is an adequate attitude, exactly the opposite of the acceptable attitude of
TVC, which, in its stylebook, recommends: It makes no sense to call a German
character Herr, or a French one Monsieur. All characters should be ad-
dressed as senyor (Televisi de Catalunya 1997: 29; authors translation).
European researchers conrm our hypothesis. Olivier Goris (1993) analyzes
the French dubbings of some lms from the Netherlands and the USA. He
concludes that the cultural references are neutralized (we should remember that
TVC often Catalanizes). Neutralization is achieved by means of hyperonyms
or explicitations. The characters names or the names of cultural items (for
example Thanksgiving) are not translated into French.
According to Goris, dubbing (like any translation strategy) is culturally
linked. Thus, he identies the neutralizing attitude with a nationalistic attitude:
The various nationalistic governments aim at standardizing imported products
in order to protect the homogeneity of the local system of social values (1993: 171).
Another researcher, Birgit Nedergard-Larsen (1993), notes that in the
subtitles for Danish television:
There are a series of strategies available to the subtitler for solving extra-
linguistic culture-bound translation problems, ranging from verbatim transfer
of source language elements via culturally neutral explicitations or paraphrases
to target language adaptation (1993: 238).
Incidentally there was only one example of cultural adaptation in the four
lms examined (1993: 235).

In other words, the tendency is to produce either adequate or neutralizing

translations, and acceptable ones are hardly ever used.
As a result of our research we can state that Hlne et les garons was
translated following precise translation strategies that made it into an extremely
acceptable dubbing, Helena, quina canya!. Although this is an extreme case, it is
not so dierent from the rest of TVC dubbings. Our nal conclusion is that
TVC dubbings are more acceptable than dubbings into Spanish, on the one hand,
and than some dubbings into other European languages, on the other. Acceptabil-
ity covers linguistic as well as cultural aspects. The purpose of acceptable
dubbings is credibility: TVC believes that the Catalan audience will nd more
credible what is familiar to them, what does not bother them with foreign
elements which might interfere with comprehension. In contrast, other televisions
think that their audience seeks the exotic, foreign component; thus, they produce
adequate dubbings, which oer foreignizing language and cultural elements.
</TARGET "iza">

160 NATALIA IZARD

Notes

1. We must say that no translation is completely adequate nor acceptable. Each one takes a
position of commitment, but a position that is always closer to one of the two poles.
<TARGET "gon" DOCINFO

AUTHOR "Vctor M. Gonzlez Ruiz"

TITLE "La Traduccin del Ttulo Cinematogrco Como Objeto de Autocensura"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 16

La traduccin del ttulo cinematogrfico

como objeto de autocensura
El factor religioso

Vctor M. Gonzlez Ruiz

Universidad de Las Palmas de Gran Canaria

1. Introduccin: objetivos y metodologa

En las siguientes pginas, estudiaremos cmo inuy la ideologa catlica en la

traduccin al espaol de los ttulos de las pelculas extranjeras distribuidas y
exhibidas en Espaa durante la dictadura del general Franco (19391975). Ms
concretamente, nos centraremos en aquellas pelculas americanas e inglesas que
se exhibieron en Espaa durante esos aos. Prestaremos especial atencin a
determinados condicionantes extralingsticos (religiosos, polticos, econmicos),
que, tal como quedar demostrado a continuacin, fueron factores fundamentales
en el proceso de titulacin de estos lmes.
Adoptaremos un punto de vista pragmtico y tolerante hacia el objeto de
estudio (y no prescriptivo y censurador, tal y como se hace con frecuencia al
afrontar un estudio de la traduccin de ttulos cinematogrcos),1 con el n de
poder apreciar en su justa medida la importancia de las distintas circunstancias
histricas del momento como decisivos elementos de inuencia sobre los
tituladores o traductores ocasionales de los ttulos de estas pelculas.

2. Situacin histrica: influencia de la Iglesia catlica

Entre los factores que inuyeron en la sociedad espaola de la poca franquista,

la religin ocup uno de los primeros lugares en importancia. Ya desde el mismo
comienzo de la dictadura del general Franco, la Iglesia catlica tuvo un papel
destacado a la hora de legitimar el golpe de Estado llevado a cabo por ste, as
como todos sus actos a partir de entonces. Los jerarcas catlicos dieron su
162 VCTOR M. GONZLEZ RUIZ

bendicin al inicio de un rgimen que, junto con la espada, utilizara la cruz

como smbolo de la nueva Espaa nacida al trmino de la guerra civil.2
De este modo, la Iglesia catlica y el Estado franquista iniciaron una
estrecha colaboracin que durara hasta los ltimos das de esta etapa poltica. En
virtud de dicha cooperacin, la dictadura lograba, por un lado, el respaldo moral
que necesitaba para dar sentido a su particular cruzada contra el mal, as como
mejoraba levemente su debilitada imagen exterior; por su parte, la Iglesia poda
acceder a plataformas de poder y utilizar vas institucionales con el n de
imponer su particular visin de la moralidad y la decencia, provocando que su
opinin ante estos temas se convirtiese en la posicin ocial del Estado.3 Parece
lgico, por tanto, que, durante los siguientes aos, la incorporacin del nacional-
catolicismo a la vida poltica y cotidiana de Espaa se produjera de manera
progresiva e imparable.
Una vez establecida la conexin entre los jerarcas eclesisticos y la masa de
espaoles gracias a la presin del Estado, los primeros no dudaron un solo
instante a la hora de hacer uso de los instrumentos de control popular que ste
pona a su alcance. La censura (en todos los rdenes de la vida) fue uno de los
principales y aqul que con mayor alborozo recibi la Iglesia catlica, pues
supona una herramienta fundamental por medio de la cual evitar la ltracin de
ciertas ideas y modos de vida indecentes hacia la mayora de ciudadanos. De
este modo, prensa, literatura, radio, televisin y cine fueron las principales
vctimas de las comisiones censoras, las cuales, herederas de las peores virtudes
de la Santa Inquisicin, provocaron con sus decisiones una indiscriminada
sangra que, durante varias dcadas, hiri de muerte el desarrollo intelectual de
la poblacin espaola.

3. Documentos eclesisticos: justificacin de la censura

De cualquier manera, la Iglesia catlica nunca se mostr parca a la hora de

fundamentar toda su actividad censora a travs de documentos del ms variado
origen y rmados por miembros de los distintos estamentos de su jerarqua. Los
ms destacados (por la personalidad de sus autores) son las encclicas que sobre
la radio, la televisin y, en especial, el cine, se publicaron bajo los ponticados
de Po XI y Po XII: Vigilanti Cura (1936) y Miranda Prorsus (1957), respec-
tivamente. En la primera, dirigida fundamentalmente al Episcopado de los
Estados Unidos de Norteamrica, se aconsejaba sobre el recto uso del cine y
sobre la correcta aplicacin de sistemas de clasicacin de pelculas a nivel
nacional (Walsh 1996: 145146). En lo que respecta a la segunda, cuyo campo de
inuencia se extenda a la radio y la televisin (medios que, sobre todo este ltimo,
estaban adquiriendo un gran protagonismo social), se observaba un tono bastante ms
 CENSURA Y TRADUCCIN DE TTULOS CINEMATOGRFICOS 163

positivo y constructivo en la aproximacin a estos espectculos de masas.

No obstante, y a pesar de sus diferencias, ambas encclicas coincidan en su
deseo comn de hacer partcipes a los estamentos ms bajos de la Iglesia en la
lucha contra la indecencia en el cine (medio en el que nos vamos a centrar a
partir de ahora). Si en Vigilanti Cura se exhortaba a los distintos prrocos a
condenar libremente en sus respectivas dicesis lmes que haban recibido un
veredicto demasiado benigno a nivel nacional, en la promulgada por Po XII se
iba an ms all y se involucraba a los mismos eles, a los cuales se haca
responsables de ser mensajero, portador y dispensador de los valores espirituales
en benecio del perfeccionamiento individual y social (Bonet 1957: 23).
Como puede observarse, la jerarqua eclesistica deseaba convertir en tarea
de todos una actividad censora que, ejercida slo desde el poder y a travs de la
imposicin, no contaba con demasiadas posibilidades de xito. Es por esto que
la encclica Miranda Prorsus continuaba destacando que los catlicos deban,
como buenos hijos, desplegar todo esfuerzo para que la Iglesia pueda valerse de
los inventos en orden a la santicacin de las almas (Bonet 1957: 24). Pero, por
medio de estas indicaciones, no se estaba exhortando a los eles slo a seguir al
pie de la letra el dogma catlico; adems, se les animaba a manipular los hechos
con el n de que siempre se pudiera extraer de stos alguna enseanza moral. Es
decir, se haca necesario evitar que el cine, como transmisor de ideas y vehculo
de conocimiento, difundiera el error (Burke 1955: 88). Por tanto, los responsa-
bles de las pelculas (y, entre ellos, los tituladores), en su calidad de informa-
dores, deban tratar de comprender y hacer comprender los fracasos y los
errores cometidos, adems de sugerir el remedio y consiguientemente hacer
obra positiva y constructiva (Bonet 1957: 25).
Por supuesto, los responsables de la industria cinematogrca no estaban
dispuestos a aceptar de buen grado todas estas indicaciones de la Iglesia catlica,
mxime cuando la tan buscada rentabilidad econmica en taquilla pasaba por el
tratamiento de unos temas y el retrato de unos personajes que poco tenan que
ver con lo que se aconsejaba en los documentos eclesisticos. Sin embargo,
ignorar de pleno todos o gran parte de estos consejos hubiese desembocado
probablemente en el rechazo o la mutilacin de la obra flmica. En consecuencia,
se haca necesario comprometerse con lo que dictaban los censores, por lo menos
hasta el mismo lmite de lo permitido, con el n de que stos manipularan lo
menos posible la pelcula.

4. La industria cinematogrfica y la autocensura: el caso de los ttulos

Una de las maneras de hacer esto, principalmente en el caso de los lmes en

otros idiomas que iban a ser distribuidos en Espaa, era a travs de la utilizacin
164 VCTOR M. GONZLEZ RUIZ

de unos ttulos acordes con la corriente religiosa descrita anteriormente. Y es que

el proceso de castellanizacin de la vida espaola (en ocasiones, a travs de
documentos institucionales, el ms notorio de los cuales -la Orden Ministerial de
abril de 1941- impona la obligatoriedad del uso del espaol en todas las
proyecciones cinematogrcas),4 forzaba (no siempre de una manera explcita)
a los distribuidores de pelculas extranjeras a traducir o adaptar libremente todos
los ttulos originales que no estuvieran en espaol. De esta forma, a los respon-
sables de la distribucin se les pona en las manos un arma por medio de la cual
podan manipular el ttulo en otra lengua con el n de ofrecer uno en castellano
que, siguiendo la terminologa catlica, permitiera mejorar, elevar y educar a
los eles y, consecuentemente, adular a los miembros eclesisticos de las
comisiones examinadoras y lograr sus favores.
La autocensura se extenda, as, hasta el campo de la titulacin cinemato-
grca y daba lugar a toda una galera de ttulos que, estudiados desde la
perspectiva actual, se nos muestran reveladores de una poca y un ambiente
marcados por las restricciones moralizantes y unos roles familiares claramente
denidos. Si observamos los ttulos espaoles que se emplearon en la dictadura
franquista para las producciones americanas e inglesas, descubrimos que en
muchos casos se tom el relevo dejado por los jerarcas eclesisticos en el sentido
de participar en la misin evangelizadora de educacin y prevencin de los
eles.
De acuerdo con esta actitud pedaggica de inuencia positiva en favor
de las buenas costumbres y de los valores del espritu (Burke 1955: 89),
numerosos lmes recibieron ttulos cuyo objetivo era el de sealar y prejuzgar de
manera clara los comportamientos indecentes y torcidos del protagonista de la
historia. As, se usaban adjetivos que hacan referencia a pecados y a conductas
tradicionalmente castigadas por la Iglesia catlica; o tambin sustantivos que, por
sus connotaciones tradicionalmente negativas, permitan presagiar y prevenir al
potencial espectador contra los procederes errneos de los personajes.5 Dentro
de este grupo, podemos incluir los siguientes ejemplos:
(1) La tortura de la carne (The Way of All Flesh, 1940)6
(2) La envidiosa (Harriet Craig, 1950)
(3) Sin conciencia (The Enforcer, 1951)
(4) La egosta (Payment on Demand, 1951)
(5) Garras de codicia (Crosswinds, 1951)
(6) Cautivos del mal (The Bad and the Beautiful, 1952)
(7) Amor prohibido (In the Cool of the Day, 1963)
(8) Esclavos del pecado (An American Dream, 1966)
(9) Fiebre de codicia (The Pink Jungle, 1968)
 CENSURA Y TRADUCCIN DE TTULOS CINEMATOGRFICOS 165

Tal como puede observarse, raramente se lleva a cabo una traduccin literal del
ttulo en ingls; en la mayora de los casos, ste se adapta de manera que se
integre en el ambiente de rectitud moralista de la poca. Sustantivos como
tortura, pecado, conciencia o mal, forman parte de la terminologa que los
jerarcas eclesisticos del momento empleaban constantemente para censurar
malos hbitos o costumbres. Por poner tan slo un ejemplo, el arzobispo de
Sevilla dena los bailes en una pastoral de 1946 de la siguiente forma (Abella
1996: 110): El baile es gavilla de demonios, estrago de la inocencia, solemnidad
del inerno, tiniebla de varones, infamia de doncellas, alegra del diablo y
tristeza de los ngeles.
Vistos los ejemplos anteriores, cualquiera de las deniciones del baile
empleadas por el cardenal Segura poda haber sido utilizada por los tituladores
de entonces para sus propias pelculas. No obstante, en ocasiones, se iba an ms
all y se haca referencia directa a los mandamientos (El cuarto mandamiento,
The Magnicent Ambersons, 1942); o incluso a pasajes sagrados, tal como en el
caso del Salmo 23 sobre El buen pastor, una de cuyas partes es empleada en
su doble versin para la traduccin de Dark Passage (La senda tenebrosa,
1947) y The Story of Dr Wassell (Por el valle de las sombras, 1944). Igual-
mente, el constante empleo de ciertas palabras de marcado cariz religioso es otro
de los rasgos caractersticos que presentan los ttulos de las producciones
extranjeras exhibidas durante el rgimen franquista. As, amn de las vistas en el
grupo anterior, ngel, demonio, alma,7 cielo o milagro son un referente constante
para los distribuidores de estos lmes, tal como puede comprobarse en los ttulos
que se exponen a continuacin:
(1) Dos en el cielo (A Guy Named Joe, 1943)
(2) Niebla en las almas (Faces in the Fog, 1944)
(3) El callejn de las almas perdidas (Nightmare Alley, 1947)
(4) El milagro del cuadro (The Light Touch, 1951)
(5) El refugio de los ngeles (The Village, 1954)
(6) El demonio, la carne y el perdn (The Singer, not the Song, 1961)
(7) ngeles sin paraso (A Child Is Waiting, 1963)
(8) Almas de metal (Westworld, 1973)
Sin embargo, el carcter educativo (en el sentido que ya hemos expuesto) de
todas estas traducciones/adaptaciones, podra entenderse tambin ms all de su
funcin de gua positivo hacia todo lo bueno (Burke 1955: 90). Y es que, a
travs del empleo de unos ttulos alusivos a hbitos o conductas calicadas
tradicionalmente de pecaminosas, se estaba lanzando una excitante invitacin a
la morbosa curiosidad de los millones de espaoles que se sentan acosados por
166 VCTOR M. GONZLEZ RUIZ

las constantes llamadas al orden moral desde los plpitos y las instituciones
estatales. De esta forma, ante el reclamo de dos ttulos tan diferentes como El
hombre que vendi su alma (The Devil and Daniel Webster, 1945) y Vocacin
de marino (Youth Takes a Fling, 1938), es probable que el pblico encontrara el
primero mucho ms sugerente y atractivo; hecho del que, con toda seguridad, los
distribuidores eran muy conscientes y que, en consecuencia, explotaban para su
propio benecio.
Parece claro, por tanto, que los encargados de distribuir las pelculas
extranjeras compartan con la jerarqua eclesistica el convencimiento de que el
pblico cinematogrco necesitaba que se le mostrara el camino. Son ml-
tiples las referencias dentro de documentos catlicos a la minora de edad de los
espectadores con respecto a los temas morales. Desde la consideracin de que,
en lo relativo a la sexualidad en el cine, todos son menores (Begoa 1956: 14),
hasta la denicin del sptimo arte como rerum scholae por parte del Papa Po
XI (Bonet 1957: 25), es evidente que, para la Iglesia catlica, la exhibicin
cinematogrca deba ser utilizada como inmejorable medio de propagacin de
sus ideas.
Es por esto que, adems de para juzgar conductas inmorales (tal como
hemos podido comprobar en los ttulos ya presentados), tambin se impona la
utilizacin del cine como forma de expresar el optimismo y el lado positivo de
una cotidianidad que se haca demasiado cuesta arriba para la gran mayora de
espaoles. En consecuencia, se pudieron observar en la pantalla historias y
personajes que demostraban que, tras el sufrimiento, y gracias a una buena dosis
de paciencia cristiana, se encontraba la felicidad completa. Estas ideas se
reejaron tambin en varios ttulos, expresin del gozo y la actitud optimista ante
la vida, tal como puede apreciarse en los siguientes:
(1) El sol sale maana (Our Vines Have Tender Grapes, 1945)
(2) Maana es vivir (Tomorrow Is Forever, 1946)
(3) De ilusin tambin se vive (The Miracle on 34th Street, 1947)
(4) Viva la vida! (You Gotta Stay Happy, 1948)
(5) Vivir es lo que importa (The Young Doctors, 1961)
(6) Una razn para vivir (The Little Shepherd of Kingdom Come, 1961)
(7) La vida vale ms (The Slender Thread, 1965)
(8) Qu hermosa es la vida (Whats So Bad about Feeling Good?, 1968)
Sin embargo, el trabajo de los tituladores no se limit a censurar comporta-
mientos indecentes y a resaltar el lado alegre de una existencia insulsa. La
necesidad de conseguir buenos resultados econmicos llev a los distribuidores
de los lmes a jugar, una vez ms, la carta de la doble moral y la ambigedad
 CENSURA Y TRADUCCIN DE TTULOS CINEMATOGRFICOS 167

mediante la referencia a la mujer y a su rol en la pareja/matrimonio. As que, por

un lado, se exaltaban las virtudes de unas mujeres cuyas tareas estaban bien
denidas desde la Iglesia catlica (esto es, honesta ama de casa, sumisa amante
de su marido y generosa madre de sus hijos), segn se puede comprobar a
continuacin:
(1) Casi un ngel (It Started with Eve, 1941)
(2) Una gran seora (The Great Mans Lady, 1942)
(3) Claudia, esposa moderna (Claudia, 1943)
(4) Heronas annimas (Ladies Courageous, 1944)
(5) Chica para matrimonio (The Marrying Kind, 1952)
(6) Para ella un solo hombre (The Helen Morgan Story, 1957)
(7) Todas las mujeres quieren casarse (Ask Any Girl, 1959)
Y, por el otro lado, se aluda a las relaciones extramatrimoniales del marido, a
la gura maligna de la otra y a la obligada delidad de la esposa, tal como
ilustran los siguientes ejemplos:
(1) No te fes de las tontas (Jane Steps out, 1938)
(2) Ardid femenino (Vivacious Lady, 1938)
(3) Esta mujer es ma (I Take This Woman, 1940)
(4) Preero la secretaria (Moonlight Masquerade, 1942)
(5) Si ella lo supiera (Everybody Does It, 1949)
(6) Entre dos mujeres (Aair with a Stranger, 1953)
(7) Mi marido se divierte (The Tunnel of Love, 1958)
(8) Ellas y las otras (Wives and Lovers, 1963)

5. Conclusin

En varios de estos ejemplos, se advierte de manera clara la existencia de una

estrategia comercial que, a pesar de su simplicidad, se revelaba efectiva. Tal
como se coment anteriormente, la alusin a temas de ndole moral poda, a la
vez que servir de vehculo de evangelizacin para la Iglesia catlica, funcionar
tambin como gancho para un pblico vido de historias pecaminosas y
extremas. Y es que, independientemente del desarrollo posterior del argumento
ya en la propia obra flmica, el ttulo supona y supone el primer punto de
contacto con la gran mayora de los potenciales espectadores, los cuales iban a
identicar esta particular experiencia lingstica con la pelcula determinada. Por
este motivo, es posible que, en muchos casos, este pblico se sintiera frustrado
168 VCTOR M. GONZLEZ RUIZ

tras la proyeccin del lme, puesto que su ttulo haba alentado unas expectativas
que difcilmente podan ser satisfechas.
Todo esto nos devuelve, nalmente, al punto inicial de discusin sobre la
metodologa que se debe emplear a la hora de estudiar los ttulos cinematogr-
cos. Parece claro que no es viable ligar el estudio de stos al de los ttulos
literarios, ya que las caractersticas industriales y masivas de los primeros
raramente pueden aplicarse a los ltimos. Por lo tanto, englobar ambos en un
mismo apartado seguramente llevar a juicios errneos. De la misma manera,
sugerir que el ttulo cinematogrco debe mejorar y facilitar la comunicacin, as
como evitar la comprensin errnea del texto -en este caso, el flmico-
(Jovanovic 1990: 213), tal como se hace al igualarlo a aqul literario, resultara
demasiado ingenuo a la vista de todo lo estudiado en estas pginas.
En el caso de los ttulos cinematogrcos, la armacin de R. Rabadn
(1994: 137) de que los textos traducidos funcionan como una til herramienta
para justicar ciertos valores ya presentes, o que se quieren imponer a una
sociedad, es especialmente cierta. Y esto es as en la medida en que las
versiones espaolas de estos ttulos se ajustan perfectamente a la denicin
cultural que de la traduccin como reescritura dan S. Bassnett y A. Lefevere
(1990: 113), entendiendo sta como manipuladora y cargada de ideologa. Por
lo tanto, muy diversas circunstancias de todo tipo (como, por ejemplo, las
religiosas y econmicas, por citar las ms relevantes en nuestro artculo) han de
tenerse en cuenta si se pretende investigar seriamente en este campo.

Notas

1. En este sentido, otros estudios, como los de I. Pascua Febles (1994) y M. Jovanovic (1990),
prcticamente olvidan la importancia de los factores extralingsticos y se centran en la
manifestacin de errores de traduccin y en la correccin de stos.
2. Ilustracin de este singular matrimonio de fuerzas es el extenso muestrario de discursos, toda
una declaracin de intenciones sin resquicio alguno de ambigedad, realizados tanto por la
jerarqua eclesistica como por el propio Franco, donde se reconocan al dictador mritos casi
divinos. En 1939, por ejemplo, se presentaba al general de esta manera (Abella 1996: 23): El
Caudillo impetra la ayuda de Dios para la forja del Imperio, y es ungido con las palabras
sacramentales de la Iglesia. Igualmente, el nuevo Jefe de Estado asuma sin problemas su
propia tarea como lder espiritual con estas palabras (Abella 1996: 23): Seor Dios, en cuyas
manos est todo derecho y todo poder, prstame tu asistencia para conducir este pueblo a la
plena libertad del Imperio para gloria tuya y de tu Iglesia.
3. Una de las primeras muestras de esta inuencia vendra con la promulgacin, en 1943, de la
Ley de Ordenacin de la Universidad Espaola, en la cual se prescribe que la Universidad
adaptar sus enseanzas al dogma y a la moral catlicas y a las normas del Derecho Cannico
en vigor (Gubern 1981: 54).
</TARGET "gon">

CENSURA Y TRADUCCIN DE TTULOS CINEMATOGRFICOS 169

4. Segn Romn Gubern, esta Orden Ministerial (elaborada a imagen y semejanza de la

mussoliniana Ley de Defensa del Idioma), nunca fue publicada en el B.O.E. (vila
1997: 1214), a pesar de lo cual siempre ha sido considerada como el principal motivador de
la imposicin del doblaje en Espaa. Otros factores, tal como la iniciativa privada o la alta tasa
de analfabetismo de la poca, tambin han de tenerse en cuenta para comprender el estableci-
miento del doblaje en Espaa.
5. A este respecto, un ttulo como No hay crimen impune (Down Three Dark Streets, 1954),
podra introducir y, a su vez, resumir este apartado.
6. Ttulo espaol (ttulo original en ingls, ao de produccin).
7. La palabra alma se utilizaba con frecuencia en los documentos y en las declaraciones de
representantes eclesisticos, en los cuales se haca alusin constante a la salvacin de las
almas. Una muestra de la sintona de ideas entre Iglesia y Estado es la declaracin del catlico
integrista G. Arias Salgado de que, al frente del Ministerio de Informacin y Turismo (recin
creado en 1951), su misin era la de salvar almas para el cielo (Gubern 1981: 122).
<TARGET "mil" DOCINFO

AUTHOR "John Milton"

TITLE "The Translation of Mass Fiction"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 17

The Translation of Mass Fiction

John Milton
Universidade de So Paulo, Brazil

1. High and Low

Despite the enormous amount of mass literature that is translated, little has been
written about these translations. Robyns (1990) looked at the omissions made in
detective novels translated into French, using the belles indles image; Paizis
(1998) found similar omissions in romances translated into French and Greek;
Sohr (1996, 1997) examined pseudotranslations of science ction into Hungari-
an; while Gouvanic (1997) found French translations of science ction novels
more faithful to the originals. This paper will contrast the translation of mass
ction, to which I give the name of factory translation, and whose characteris-
tics I detail at length, with that of literary, philosophical and erudite works, to
which I give the name of aristocratic translation.
Translation studies have traditionally been an integral part of high culture,
dominated by an aristocratic and gentlemanly coterie who have access to foreign
languages and literature. One can think of the Royalist aristocrats in the 17th
century, many of the names Tytler mentions in his Essay on the Principles of
Translation, and the many gentlemanly discussions around translations of Homer.
Many literary translations have been aimed at the learned few, the aristocrat-
ic, gentlemanly or academic coterie. This can clearly be seen, for example, in the
ideas of the German Romantics. More specically, Lawrence Venuti comments
that the kind of translation that Schleiermacher recommends aims to preserve
the linguistic and cultural dierence of the foreign text, but only as it is
perceived in the translation by a limited readership, an educated elite (Venuti
1991: 130). So, paradoxically, a translation, which should or could make a work
available to all the literate members of a new language group, may be directed
to a specic clique or coterie, thereby excluding the great majority of readers.
The rise of democracy with the enfranchisement of the masses in the 19th
172 JOHN MILTON

century produced a resulting cultural reaction in the intellectual elite, worried

about the erosion of their privilege and domination. In La deshumanizacin del
arte (1925), Jos Ortega y Gasset sees modernism as a result of this reaction. He
contrasts romantic art, which he calls popular, a child of democracy, with
contemporary anti-popular art, which many people found dicult to under-
stand. The emotion and human interest central to romantic art contrast with
modern art and literature, where Mallarm, for example, attempted to divest the
poem of all human interest; this is the dehumanization of art. In music
Debussy performed the same dehumanizing function against the heavily personal
and associative themes of Beethoven and Wagner.
Pierre Bourdieu (1986), in Distinction, surveys the cultural tastes of
dierent social classes in France in the 1970s. He found that the intellectual
classes generally believe more in the form of the representation itself, whether it
be the form of the painting, the piece of music or the literary work, than what is
actually represented. There is a withdrawal from the concrete. They are able to
reach Kants level of reection as against the easy pleasure of the senses,
whereas popular taste believes in the things themselves the ethics of the
object are tied in with the aesthetic. In other words, popular taste takes more
interest in what is said rather than the way in which it is said. Thus, it will prefer
more straightforward situations and more simply drawn gures and will wish to
become involved in the lives and moral choices of the characters of the ctional
works.
Ortega, in his essay on translation Miseria y esplendor de la traduccin,
sees translation as a way of escaping from the crowd. It is a mysterious and
impossible task in which one can attempt to bring other languages and great
authors to aect ones own thought and language, and thus to rise above the
cultural levelling of modern society.
I believe that the tradition of the foreignizing translation, with translators
emphasizing the reproduction of the aesthetic qualities of the original in the
translation is, to a great extent, the product of this strong elitist strain in modern-
ism. Here we nd Meschonnics poetic translation, Bermans translation which
is pensante, thique et potique (1985), Augusto and Haroldo de Campos
transliteration they only translate authors who have revolutionized poetic
form. This is also the site of Lawrence Venutis resistant strategies that
foreground the play of the signier by cultivating polysemy, neologism, frag-
mented syntax, discursive heterogeneity (Venuti 1992: 12).
 THE TRANSLATION OF MASS FICTION 173

2. Kitsch and Midcult

Leaving aside highbrow translation to examine what happens when a transla-

tion from high culture is adapted to mass culture, two concepts may be of use.
The rst is kitsch as described by Umberto Eco and the second is the midcult as
described by Dwight MacDonald.
The kitsch is the imposition of a xed eect to be enjoyed by the consum-
er, who does not have to attempt to understand the more complex patterns of the
operations of the artistic work. Yet, by enjoying this eect, the reader or viewer
supposes that he/she is experiencing a privileged aesthetic experience. In other
words, it is an escape from the responsibilities of art. The emotional reaction of
the reader/viewer is all-important, and any kind of reection on the causes of this
reaction is missing (Eco: 7477).
Dwight Macdonald contrasts masscult, inferior literature which has no
pretension of being erudite, with the midcult, which trivializes works of art and,
like the kitsch, deliberately attempts to produce certain eects. He lists the
characteristics of the midcult (Eco: 84): (1) it borrows avantgarde processes and
adapts them to make a message which can be enjoyed and understood by all; (2)
it uses these processes when they have been known, used and are worn out; (3)
it constructs a message as a provocation of eects; (4) it sells them as art; (5)
it tranquilizes consumers, convincing them that they have encountered culture, so
that they wont feel other worries.
Macdonald gives examples of the Midcult: the Revised Standard Version of
the Bible, which destroys the King James version, so as to make the text clear;
book clubs such as the Book of the Month Club; and Our Town by Thornton
Wilder, which uses Brechtian techniques of alienation for consolation and
hypnosis. Thus we can see the complexities of the work of art reduced: mass
culture makes the classics into products to be consumed rather than works to be
contemplated (Arendt in Eco 1993: 41).
Herbert Marcuse regrets this loss of the complexity of classic works:
Plato and Hegel, Shelley and Baudelaire, Marx and Freud [are found] in the
drugstore. [] the classics have left the mausoleum and come to life again,
now that people are just so much more educated. True, but coming to life as
classics, they come to life as other than themselves; they are deprived of their
antagonistic force, of the estrangement which was the very dimension of their
truth. The intent and function of these works have thus fundamentally changed.
If they once stood in contradiction to the status quo, this contradiction is now
attened out (in Bennett 1981: 64).

According to Marcuse, the cheapness and ready availability of the great

174 JOHN MILTON

canonized works of literature lead to a certain blas attitude to these greats.

They are taken down from their pedestals to become the cheapest and most
available of books.
Marcuse and MacDonald look down with traditional aristocratic condescen-
sion towards popular versions. But even if we do not share this snobbery, we
must agree with Marcuse that the strangeness, the literary eect, is lost.
Complexity is reduced. Wuthering Heights and Pride and Prejudice become no
more than the love stories of Catherine and Heathcli, and Elizabeth and
Darcy; Huckleberry Finn loses all socio-political and ethical commentaries to
become an adventure story for children. Moby Dick loses all mythical elements
to become merely the ght between Captain Ahab and the whale. Stylistic
complexity is lost. Only emotions are foregrounded: love, excitement, fulllment,
or struggle. The classic becomes a soap opera. The reader, believing that he has
read the original version, or, at least, a representation of the original version, is
content with this encounter with culture.
These points t in with my studies of adaptations and condensations of
adapted translations in Brazil, especially those of the Clube do Livro book club
(Milton 1995, 1996), in which I found the following to be the main changes in
the book club adaptations:
1. No sub-standard language was accepted; language was homogenized into
standard Portuguese.
2. Likewise, all polyphonic elements were cut. For example, extracts in French,
poems and meta-narratives in Charlotte Bronts The Professor were cut; the
puns, word lists and rhymes go by the board in Gargantua.
3. Sexual references were cut, for example Gulliver playing on the maidens
nipples in the Brobdingnag sequences in Gullivers Travels.
4. Scatological references are also cut, as when Gulliver urinates to put the re
in the palace out.
5. There was a certain religious censorship: extracts of Gargantua which
satirized the Catholic Church, for example suggesting that monks and nuns
should dress nicely and be allowed to get married, were cut.
6. The Clube do Livro thrived during the years of the military dictatorship in
Brazil (196489), and did not challenge the regime in its translation of Hard
Times, sections which suggest the possibility of mass action are considerably
weakened in translation. The Red House in Silas Marner becomes The
Yellow House in the Clube do Livro version!
7. The Clube do Livro was very paternalistic. Footnotes explained classical
 THE TRANSLATION OF MASS FICTION 175

references and dicult words. Footnotes would also warn against the dangers
of drinking alcohol and bad eating habits.

3. Factory Translation

Much of translation theory has worked within the faithful-unfaithful and content-
style frameworks. In recent years we have seen attempts to break out of this
straitjacket with scholars connecting translation with other areas such as psycho-
analysis, deconstruction, philosophy. However, many of these remain within the
area of high culture. Jos Lambert (1994) has written that translation studies
must break the boundaries of high culture, that translation is linked to larger
series and frames of communication. He also writes about the enormous amount
of invisible translation that takes place in every day situations. For example,
when we buy a packet of soap powder, the name, the instructions, the advertis-
ing, the production manuals will almost certainly have involved a large amount
of translation, which is never made obvious.
This study looks at the classic as an artifact, handled and bought by the
masses, used and altered at will by editors, adapters, cartoonists, lm makers, the
record business, abridgers and CD-ROM makers. To use Walter Benjamins
famous metaphor, the translation will certainly have an afterlife, but this afterlife
may take on a very dierent form to the original.
Let us look at some of the characteristics of this kind of translation:
1. Rather than being the work of an individual, the condensed or adapted
translation, or, for that matter, the dubbed or subtitled lm, or the translation
made within industry, will be the work of a team. It is a mere part of the
assembly line. The name of the translator may not appear on the work. If it
does, it may be a pseudonym: a highbrow translator may not wish to have his
name associated with the work, or it may even be an invented name for a team.
Jerusa Pires Ferreira, in her study of O Livro de So Cipriano (1992), shows how
the author of these popular collections of legends, almanacs, spells and
fragments is usually a compiler, a copier, a translator, an updater and an
inventor. In most cases the author was anonymous or used a pseudonym.
Historical parallels can be found in medieval translation, where adaptation,
omissions, retellings, alterations, etc., were a normal part of what we now call
translation.
2. Standardization, or Fordism, is an important factor in the production of
factory novels and translations. Dierent forms of standardization can be
176 JOHN MILTON

found: (a) theme: the work is tailored to suit the tastes of the reader; (b) lan-
guage: sub-standard language and dialect is cut out; (c) style: the work should
not deviate from a strict narrative style; (d) size: after 1960 the Clube do Livro
publications were standardized to 160 pages; (e) weight: a low weight to cut
postal costs will be an important economic factor in many book clubs.
3. Commercial production ignores the so-called sacredness of the author. Walter
Benjamins well-known essay The Work of Art in the Age of Mechanical
Reproduction (1969) emphasizes the fact that the contemporary mechanical
possibilities of reproducing the object will change our relationship to the work of
art and destroy the ritualistic and magical elements surrounding the original.
With lm and photography there is never a single original. The very nature of
both arts is essentially reproductive: lms and photographs cannot be collected
as paintings can. Moreover, the cost of a lm is so high that it must be made
available to the highest number of people possible. Indeed, the nature of lm is
highly democratic as it will make all myths and great authors and artists
available to cinema goers.
The fragmentation of a lm is very dierent to the unity of a painting or
a poem. The director, responsible for the nal product, will depend on the
expertise of many other dierent areas, sound, photography, etc., of which he
will have only a limited knowledge. He or she will be more of a manager of the
nal product. Compare this with the painter or novelist, who will have all
elements of the work of art under his control. The nal result of the lm will
come about as the result of the combination of a number of very dierent
elements. Though Benjamins essay concentrates on the reproducible arts of lm
and photography, his essay also helps us to understand the commercial transla-
tion, such as that of the Clube do Livro, where the editor will coordinate the
work of the translator, sub-editors, illustrator and nance department.
4. Denite commercial strategies will be used. The translation will be directed
to a denite market. There will be markets for dierent segments of society.
Non-condensed, carefully produced translations will be directed towards markets
from a higher social class. (Adapted) translations of Pride and Prejudice and
Wuthering Heights are often directed towards a market of female readers.
Similarly, Moby Dick, Kidnapped, Huckleberry Finn and Gullivers Travels are
usually directed towards the juvenile market.
5. Deadlines are all-important. The product must be released on time even though
it may have a few aws, particularly if it is not a product which is aimed at an
erudite market. Meeting deadlines is much more important than perfectly accurate
copy. The monthly deadline of the Clube do Livro meant a minimum of time for
 THE TRANSLATION OF MASS FICTION 177

proofreading, and errors abound, particularly in the names of authors: Virginia

Wol, Daniel DeFoe, Charlotte Bront, George Elliot, and Kunt (sic) Hansun.
A study by Anik Sohr on translation of mass ction in Hungary shows the
large number of editorial proofreading blunders, including remnants of the
translators own notes and queries remaining in the nal version, which shows
absolutely no revision took place (Sohr 1997: 6869, 71, 8788, 96, etc.). On
one occasion, the translator rkeny Akjay notes that the original, William
Gibsons Count Zero, contains an error, when the author calls the character called
Ramirez Rodriguez. He warns the editor by inserting (!!!error in original!!!).
This English language note ended up printed in the Hungarian published version
as the publishing company, Valhalla Pholy, presumably did not employ a proof-
reader (Sohr 1997: 59).
6. The great novel tradition, above all the 19th century novel, is particularly
reusable, in dierent shapes and guises: condensations, lms, cartoons, etc.
Many of the most popular authors Swift, Dickens, Jane Austen, Scott, George
Eliot, the Bronts, Balzac, Melville, Stevenson, were already read by mass
audiences. The 19th century realist tradition was much closer to popular taste
than modernism. By contrast, the 20th century modernist tradition of the novels
of Joyce, Faulkner, Lawrence and Virginia Woolf hardly lend themselves to
adaptation. The 19th century texts join other standard mythical texts which
have been used and reused in a variety of forms: The Iliad and The Odyssey;
Faust, Shakespeares tragedies, Oedipus, Don Quijote, etc.
7. It may often be cheaper to recycle an already existing translation than to
commission a new translation. The same novel, in the same translation, may even
be directed towards dierent markets. Thus we see the same (or slightly
modied) translation appearing in dierent guises. Ediouro published two almost
identical condensations of Pride and Prejudice in 1970. The small format edition,
translated by Nair Lacerda is slightly longer and its language is slightly more
formal than the large-format illustrated edition translated by Paulo Mendes
Campos and aimed more at the juvenile market.
As Viagens de Gulliver a terras desconhecidas was published in a heavily cut
version by Edies Cultura in 1940, Portuguese translation by Henrique
Marques Junior, scrupulously revised and modernized. Scrupulously revised
is a euphemism for enormous cuts were made. An identical translation, except
for a few uses characteristic of the Portuguese of Portugal which were adapted
for Brazil, was published as Viagens de Gulliver by Jackson in 1957. This time
there was a dierent translator: Cruz Teixeira.
The Clube do Livro often relies on previously published translations. Two
178 JOHN MILTON

translations of Ivanhoe seem to be heavily calqued on the Garnier translation of

1905: the Clube do Livro (1953) here a double edition with smaller print
and the Edies Cultura (1943). Similarly, the Clube do Livro translation of Silas
Marner (1973) is a slightly updated version of the Martins version (1942). Even
the misspelling of the authors name (George Elliot) is copied!
8. Packaging is enormously important. Adaptations of Pride and Prejudice and
Dombey and Son were packaged as Mills & Boon type romances, using photos
of North American 1950s glamour girls.
9. Commercialism is not beyond a few tricks the reader may not be treated
with all that much respect. Translations were generally introduced as special
translations, a euphemism for contains many cuts.
10. A common marketing trick is that of introducing an original work as a
pseudotranslation. The most famous pseudotranslation is Don Quijote, which
Cervantes pretended to be a translation from the Arabic. Cervantes had much
more freedom to satirize the novelas de caballera if he pretended it was a
translation. Macphersons Ossian was one of the greatest of literary hoaxes. The
Scottish poet, James Macpherson, pretended to have discovered and have
translated Celtic poems from the 4th Century AD. For a while Macpherson was
one of the most celebrated gures in the literary world.
Most frequently, the reason for pseudotranslations is commercial. Anik
Sohr documents the mystery of Wayne Mark Chapman, the name given to the
ctitious author of a series of best-selling science ction novels in Hungary
written by a team of authors. In post-communist Eastern Europe, popular ction
from the West has considerable prestige. Science ction written by Hungarian
authors would not be attractive to potential purchasers. Obviously, therefore, it
pays to invent the foreign author, the original work, and even give biographical
information on him. Wayne Chapman lives in Concord, New Hampshire, and his
rst books, Blood Season, and Banners, were published by Pengdragon Publish-
ing Co. Inc., London. Sohr, after a considerable amount of literary detective
work, managed to discover that these novels were in fact pseudotranslations and
that neither the author or the publishing house existed. In addition, after the
series began to be successful, the editors took fewer pains to insist on the fact
that these books were originally written in English (Sohr 1996: 25).
11. Such translations are ephemeral, throwaway, not catalogued by libraries and
information networks. Jerusa Pires Ferreira had little luck in her search for O
Livro de So Cipriano in national libraries (Ferreira: 1993). Indeed a librarian in
the Bibliothque National in Paris was annoyed and said that it was unthinkable
that this kind of work should be found in the national library of France!
</TARGET "mil">

THE TRANSLATION OF MASS FICTION 179

4. Conclusion

This article has examined the characteristics of translations of mass ction and
has criticised the quality of translations such as those of the Clube do Livro for
their censorship, lack of attention to detail and style and carelessness. In terms
of a mere qualitative analysis, such translations will always lose to more
carefully made academic translations. Yet such criticism does not take into
account the fact that these cheap and accessible translations were able to
introduce classic Brazilian and foreign novels to an audience which did not have
a book-buying habit, and may have resulted in an interest in literature and further
reading, playing a signicant part in improving the level of education and world
knowledge in Brazil, where educational facilities were, and still are, precarious.

Translations cited

Austen, Jane. 1970. Orgulho e Preconceito, retold by Nair Lacerda. Rio de Janeiro: Ouro.
Austen, Jane. 1970. Orgulho e Preconceito, retold by Paulo Mendes Campos. Rio de
Janeiro: Ouro.
Austen, Jane. No date. Orgulho e Preconceito, Faixa Preta Romances. Rio de Janeiro:
Faixa Preta Romances. Gerstum Carneiro.
Balzac, Honor de. 1945. Eugnia Grandet. no translator. So Paulo: Clube do Livro.
Balzac, Honor de. 1972. Eugnia Grandet, retold by Marques Rebelo. Rio de Janeiro:
Ediouro.
Bront, Charlotte. 1958. O Professor, tr. Jos Maria Machado. So Paulo: Clube do Livro.
Bront, Emily. No date. O Morro dos Ventos Uivantes, tr. Vera Pedroso, condensao
Luthero Maynard, Romances de Amor da Nova. So Paulo: Cedibra
Dickens, Charles. 1969. Hard Times, tr. Jos Maria Machado. So Paulo: Clube do Livro.
Dickens, Charles. 1960. Dombey & Filho, tr. Vicente Pedroso. So Paulo: Edies
Paulinas.
Eliot, George. 1973. O Tesouro de Silas Marner, translated and adapted by Aristides
Barbosa and Henrique J. Delm. So Paulo: Clube do Livro, 1973
Rabelais, Franois. 1961. O Gigante Gargntua, tr. Jos Maria Machado. So Paulo:
Clube do Livro, 1961.
Swift, Jonathan. 1940. As Viagens de Gulliver a terras Desconhecidas. Portuguese
translation scrupulously revised and modernized. So Paulo: Cultura.
Swift, Jonathan. 1956. As Viagens de Gulliver, tr.: Jos Maria Machado. So Paulo,
Clube do Livro.
Swift, Jonathan. 1957. Viagens de Gulliver, tr: Cruz Teixeira. So Paulo: Jackson.
<TARGET "cla" DOCINFO

AUTHOR "Ana Maria Clark Peres"

TITLE "La traduction des contes de fes"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 18

La Traduction des contes de fes

Lenfant entre la tradition et lavenir

Ana Maria Clark Peres

Universidade Federal de Minas Gerais (Brazil)

Abstract

This paper aims at reecting upon the translation of fairy tales into Portuguese,
more specically about forty versions of the tale Little Red Riding Hood,
published in Brazil since 1953. Trying to analyze the particularities of the
several translations, each was compared to Perraults French model and to the
Grimm Brothers German model. Such a research led me into the following
conclusion: in place of the traditional tale, children are offered gross adapta-
tions of the classical story, marked, in its vast majority, by the omission of
scenes that are important to the narrative in favor of descriptions that idealize
the childs universe. The emphasis on the moralizing tone and the oversimpli-
cation of the language are also characteristics of these translations, traits that
presuppose a nave reader, feeble and incompetent, very different from todays
children who have gradually been inserted in the computer universe and
exposed to the national and international problems of an era of Globalization.

Ce travail a pour but de rchir la traduction des contes de fes en portugais,

plus particulirement quarante versions du conte Le Petit Chaperon Rouge,
publies au Brsil depuis 1953. A la limite, ce qui nous intresse, cest de
vrier le concept denfant sous-jacent ces productions, ainsi que le rle de la
traduction des contes classiques europens dans le contexte socioculturel actuel
de lenfant brsilien.
Pour que nos objectifs soient plus clairs, je crois que nous devons introduire
quelques rexions dordre gnral sur les productions adresses aux enfants.
A mon avis, crire pour des enfants cest toujours traduire, mme sil sagit
dune traduction intralinguale. En se considrant comme des dtenteurs dun
savoir sur les besoins et les dsirs des lecteurs enfants, de nombreux auteurs de
littrature enfantine cherchent traduire leurs perceptions du monde, leurs
182 ANA MARIA CLARK PERES

inquitudes morales et pdagogiques, bref, leur langage dadulte dans un autre

langage, plus simpli et soi-disant adapt aux divers ges des enfants.
Ce travail de traduction gagne actuellement de plus en plus dadeptes: au
moins, au Brsil, le march de ldition grandit chaque jour, et devient un
domaine qui rsiste nimporte quelle crise conomique. Mais malgr ce
dveloppement, presque ern, du genre, pour beaucoup, la littrature enfantine
se confond encore avec les contes de fes. Une telle association nest pas
fortuite, puisquun des points de repre de linstitution de cette sorte de littrature
est justement la traduction (au sens large du terme) pour le public enfantin
danciens rcits populaires et oraux, des ns trs prcises: imposer lenfant
des valeurs morales. Je fais rfrence ici au travail de Charles Perrault, considr
comme un des fondateurs du genre littrature enfantine (en 1697, on publie, en
France, son recueil de contes, Histoires et contes du temps pass, avec des
moralits).
A cette poque-l, une nouvelle conception de lenfance prdomine dans la
socit franaise: considrs comme des tres innocents, fragiles, faibles,
irrationnels, les enfants commencent tre dcouverts, particulariss et isols du
monde adulte, capable de les corrompre.1
Cest pour ces enfants pour les duquer et les distraire qucrit
Perrault. En sappuyant sur la conception de lenfance de lpoque, il aurait
identi la mentalit populaire la mentalit enfantine, toutes deux peu dvelop-
pes: la premire, cause de la condition sociale, la deuxime, cause de lge.
Catholique convaincu, avocat de la cour du Roi Soleil, il rdige en prose, dans
un langage clair, libre, direct et sagement naf , dotant chaque conte dune
moralit, place la n de lhistoire, moralit qui signale toujours les normes
du comportement qui faciliteraient le succs dune personne auprs des autres ou
bien qui lui viteraient des ennuis (Coelho 1981: 238).
Soriano (1977: XXI) nous informe quen ralit on produisait dj des
textes pour enfants: des textes rudits (comme celui des jsuites) et des contes
davertissements, oraux et populaires (Le Petit Chaperon Rouge en serait un
exemple). En gnral, pourtant, les contes folkloriques taient destins aux
adultes et ce nest qu lpoque de Perrault quils ont t utiliss pour instruire
et amuser les enfants.
Plus de cent ans aprs (en 1812), surgissent en Allemagne les adaptations
de ces contes faites par les Frres Grimm, philologues et folkloristes qui
cherchaient xer des rcits oraux de leur pays. La proccupation pour linno-
cence suppose des enfants fait que, dans leur adaptation, les auteurs suppriment
des contes ou bien des passages des contes recueillis.
A partir de ces deux adaptations, les contes de fes sont diuss et sont
devenus source dinterprtations diverses: on met surtout en valeur leur signi
et/ou leur valeur pdagogique dans la formation de lenfant. Cependant, mon
 LA TRADUCTION DES CONTES DE FES 183

avis, aucune de ces tudes ne pose la question essentielle: comment ces rcits
sont-ils prsents aux enfants, aujourdhui, dans les divers pays? En dautres
termes, quelles sont les caractristiques des diverses traductions (ou adaptations),
leur degr dcart par rapport aux adaptations classiques de Perrault et de
Grimm? A quel enfant ces nouvelles productions seraient-elles adresses?
Comme nous lavons dj dit, an de tenter dclaircir ces points, nous
avons choisi danalyser le conte Le Petit Chaperon Rouge, lun des plus diuss
parmi les nombreux contes de fes.2 En choisissant les textes analyser, nous
avons slectionn dabord trente-trois versions (publies au Brsil entre 1953 et
1985), narrant lhistoire traditionnelle (celle de Perrault ou celle de Grimm), sans
quil y ait, mon avis, une intention dlibre de la parodier.
An de faire apparatre plus clairement les caractristiques des diverses
versions, nous examinerons les dirences principales existant entre ladaptation
de Perrault et celle des Frres Grimm.
En ce qui concerne la caractrisation du Petit Chaperon Rouge, la version
de Perrault ne met en relief que sa beaut. Celle de Grimm nous apprend quelle
est petite et tendre. Dans la premire, la mre ne donne pas de conseils sa lle
quand elle lui demande dapporter une galette et un petit pot de beurre la
grand-mre. Dans la seconde, la petite lle doit apporter sa grand-mre un
gteau et une bouteille de vin, et sa mre lui donne plusieurs conseils pour un
bon comportement. Dans le texte franais, en rencontrant le Petit Chaperon
Rouge dans la fort, le loup essaye de la tromper, en lui suggrant un chemin
plus long jusqu la maison de sa grand-mre, tandis quil dcide daller par un
autre, plus court, ce qui fait quil arrive chez la grand-mre plus tt quelle. Dans
le texte allemand, le loup lui suggre dobserver la nature, et lenfant lui obit et
nit par quitter son chemin, en pntrant trs loin dans la fort. Dans les deux
versions, le loup avale la grand-mre en entrant chez elle et, couch dans le lit,
attend larrive du Petit Chaperon Rouge. Dans la premire, quand elle entre
dans la maison (aprs avoir frapp la porte), la petite lle reoit une invitation
du loup: Viens te coucher avec moi . Elle se dshabille et va se mettre dans
le lit avec le loup. Dans la deuxime version, le Petit Chaperon Rouge entre dans
la maison sans frapper (la porte est ouverte) et stonne de lapparence de la
grand-mre . Dans la version franaise, le dialogue nal contient des rfren-
ces aux bras, aux jambes, aux oreilles et aux dents du loup. Dans la version
allemande, le Petit Chaperon Rouge parle des oreilles, des yeux, des mains et du
museau de lanimal. Dans la premire, le loup mange le Petit Chaperon Rouge
et lhistoire se termine sur cela. Ensuite, on prsente la moralit, dans laquelle on
avertit les jeunes lles du danger que reprsente le fait dcouter les loups
doucereux. Dans la version allemande, le loup mange galement le Petit Chape-
ron Rouge, mais le dnouement est heureux: un chasseur arrive providentielle-
ment et ouvre le ventre du loup endormi, en sauvant la grand-mre et sa petite-
184 ANA MARIA CLARK PERES

lle. Le Petit Chaperon Rouge promet de ne plus jamais dsobir sa mre, en

sortant de la fort.3 Le texte de Grimm prsente galement le retour du Petit
Chaperon Rouge la maison de la grand-mre: elle ne court dsormais plus
aucun risque, car sur la route elle ne fait plus attention un loup qui tente de la
dtourner de son chemin.
Quant aux trente-trois versions du conte Le Petit Chaperon Rouge en langue
portugaise, au Brsil, nous constatons que dix dentre elles se veulent originaires
de Perrault et cinq, de Grimm. La plupart (quatorze) nindique pas leur origine,
mais on peut remarquer une prfrence gnrale (trente textes) pour la n
heureuse de Grimm, encore que plusieurs informations de loriginal allemand
soient modies. Il y a quatre textes qui, quoique dles, dans les grandes
lignes, aux intrigues classiques de Perrault et de Grimm, ne font aucune rf-
rence aux deux crivains, mais citent dautres noms dauteurs originaux, appa-
remment des adaptateurs de luvre. Seules sept ditions indiquent quil sagit
de traductions, huit signalent que ce sont des adaptations et mlangent beaucoup
les informations de Perrault et de Grimm (deux prsentent le surnom et non le
nom de ladaptateur). La plupart des versions ne disent pas sil sagit dadapta-
tions ou de traductions, au sens strict du terme. Leur intention est tout simple-
ment de raconter lhistoire du Petit Chaperon Rouge.
Parmi les sept versions qui se veulent des traductions, quatre citent Perrault
comme auteur original et trois dclarent quelles prennent leur origine dans le
texte allemand. En fait, ces versions sapprochent davantage des adaptations
classiques, mais elles contiennent dj des caractristiques qui marqueront les
autres versions, savoir:
(1) lutilisation ritre de diminutifs qui nexistent pas dans les originaux
en portugais: netinha (petite-enfant mignonne); meninazinha (petite enfant
mignonne); vovozinha (petite grand-mre); orezinhas (petites eurs); cestinho
(petit panier); velhinha (petite vieille dame); cheirinho (petite odeur);4 (2)
lemploi excessif dadjectifs en portugais: chapeuzinho lindo (petit
chaperon joli ); menina bem-educada (enfant bien lev ); menina
apetitosa (enfant aptissant ); grande oresta ( grande fort);
bolinhos gostosos (de savoureux gteaux ); pobre vovozinha ( pauvre
petite grand-mre); querida neta ( chre petit-enfant); horrvel cara
( horrible visage); feroz apetite (apptit froce ); enorme barriga
(ventre norme ), etc; (3) le recours des expressions ou des particules
intensives les plus diverses en portugais: bolinhos muito gostosos (des
gteaux trs savoureux); av muito fraca (une grand-mre trs faible);
lobo muito grande e muito mau (un loup trs grand et trs mau-
vais); tesoura muito grande (des ciseaux trs grands); lobo louco de
vontade de comer Chapeuzinho Vermelho (un loup qui crve denvie de
manger le Petit Chaperon Rouge); lobo cai no cho com toda a fora (le loup
 LA TRADUCTION DES CONTES DE FES 185

tombe par terre de toutes ses forces ); ores lindssimas (des eurs dune
extrme beaut ); a av, a menina e o caador alegram-se imensamente (la
grand-mre, lenfant et le chasseur se rjouissent normment ); Chapeuzinho
segue seu caminho bem direitinho (le Petit Chaperon Rouge suit trs
sagement son chemin), etc.; (4) omission de certains passages du rcit (parmi
les versions qui suivent Perrault, seulement une arme que le Petit Chaperon
Rouge rejoint le loup dans le lit).
Des quatre textes qui se veulent des traductions de Perrault, trois ont un
dnouement malheureux, et un seul contient la moralit la n de lhistoire. Des
trois versions qui se disent originaires de Grimm, deux dentre elles seulement
contiennent le retour du Petit Chaperon Rouge chez sa grand-mre.
Voyons maintenant les principales caractristiques des ditions qui prten-
dent tre des adaptations ou qui nexplicitent pas si elles sont des traductions ou
des adaptations:
(1) Lattribution dun nom ou dun ge au Petit Chaperon Rouge est
frquente, en mme temps que lon omet souvent lexpression Il tait une fois
(ces deux modications violent le principe de lindnition, propre aux contes
traditionnels);5 (2) les caractristiques morales de la petite lle sont presque
toujours renforces (elle devient gentille , bonne , obissante , bien
leve , etc); (3) sont introduits galement des ftes, des habitudes, des prome-
nades et des jeux du Petit Chaperon Rouge avec ses petits amis, les petites btes;
(4) de nouveaux conseils sont donns par la mre au Petit Chaperon Rouge; (5)
les aliments varient galement, un appel net ce qui est soi-disant enfantin: ce
sont des plats savoureux ou bien des tartes dlicieuses , places dans de
jolis petits paniers; (6) en arrivant au bois, lenfant emploie son temps faire
de longues promenades, avant mme de rencontrer le loup: elle saute, elle bondit,
elle chante, elle joue avec de petits lapins, qui lui donnent aussi des conseils; (7)
quand le loup rencontre la petite lle, il lui suggre des promenades et des jeux
les plus divers, et fait mme la course avec elle; (8) lentre de lanimal dans la
maison de la grand-mre (et celle du Petit Chaperon Rouge, un peu plus tard),
qui se passe rapidement dans les textes originaux, se prolonge dans les versions
que nous avons analyses; (9) dans un grand nombre de textes le loup ne mange
pas la grand-mre, puisque celle-ci svanouit, ou fuit, ou se cache dans lar-
moire (ou bien y est enferme par lanimal) ou dans le grenier ou dans la cave;
(10) la deuxime rencontre de la petite lle avec le loup est aussi considrable-
ment retarde, grce lintroduction de dialogues divers; (11) aucune de ces
versions ne signale que le Petit Chaperon Rouge se dshabille et se couche dans
le lit avec le loup; (12) quant au dialogue nal, aucun texte ne parle des jambes
du loup, comme le fait Perrault; (13) dans la plupart des versions, le loup ne fait
que menacer le Petit Chaperon Rouge de la manger: celle-ci crie au secours et
est aide par ses amis, les petites btes; ou bien, elle court, poursuivie par le
186 ANA MARIA CLARK PERES

loup, au milieu dune norme agitation; (14) des chasseurs la sauvent, battent le
loup, tirent sur lui avec un fusil ou le laissent senfuir (ou il est arrt, ou bien
il tombe dans un roncier et devient un bon loup); (15) la grand-mre est
libre et il y a de longues ftes pour commmorer la n heureuse, en plus de
nouvelles leons de morale de tous les adultes prsents et des promesses ritres
du Petit Chaperon Rouge de toujours bien se comporter; (16) seule une version
contient le retour du Petit Chaperon Rouge chez sa grand-mre.
En ce qui concerne le langage de ces versions, nous observons une intensi-
cation des procds dj rencontrs dans les traductions des contes: (1) une
utilisation excessive de diminutifs et aussi de nombreux clichs de langage:
belle comme une eur , des cheveux si blonds, qui paraissent des ls dor ,
des eurs couvrent le sol comme un tapis ; (2) emploi excessif aussi dadjec-
tifs: la petite lle est toujours gaie , contente , heureuse ; cest une petite
innocente (le loup, son tour, apparat, dans quelques versions, comme un
gnie de la fort, et, dans dautres, dguis en ange, il a mme un surnom:
Bonbon , et, de mchant, il passe, parfois gourmand , espigle ou
devient un petit loup).
Aprs ce rsum rapide des caractristiques principales des versions en
langue portugaise, au Brsil, il est possible de faire une synthse des modica-
tions proposes:
1. omission de passages du rcit: il semble que des scnes considres comme
tristes, violentes ou indcentes sont supprimes, pour prserver la soi-disant
innocence et ingnuit des enfants;
2. surcrot dinformations ou substitution dexpressions: en retardant le suspens
du rcit, on introduit beaucoup dactions, des dialogues divers, de longues
descriptions: lunivers enfantin est strotyp et idalis, marqu toujours par la
lgret, la joie constante, lingnuit, lirresponsabilit et linconsquence.
Visant peut-tre une simplication du langage et une intensication de
laectivit, il y a un excs de diminutifs et dadjectifs qui ne se trouvent pas
dans les textes originaux. En prtendant transmettre beaucoup dmotion, on
arrive facilement au mlodrame ou une grande banalit, avec lemploi de
plusieurs clichs de langage;
3. emphase du ton moralisateur: la morale exprime dans de nombreux textes ne
se situe plus hors de lhistoire, comme chez Perrault, ou sous la forme de
recommandations de la mre et dune promesse du Petit Chaperon Rouge,
comme chez Grimm. Dans la plupart des versions tudies, nous la trouvons tout
au long du rcit, sous la forme de conseils varis qui doivent tre assimils par
la protagoniste. (En ralit, ce rappel est adress lenfant lecteur.)
Dans la tentative de vrier si les traductions ou adaptations plus rcentes du
 LA TRADUCTION DES CONTES DE FES 187

conte Le Petit Chaperon Rouge ont subi une modication signicative, jai essay
rcemment de slectionner et danalyser un nouveau corpus.
Pour ne pas restreindre ma slection des librairies et des maisons
ddition dune seule ville, jai choisi de faire des recherches partir des ditions
du conte existantes dans une librairie virtuelle qui reoit des commandes de
livres via Internet et les envoie la plupart des tats brsiliens. Jai pu avoir
accs sept ditions du conte.
De ces sept versions, cinq ne se rfrent pas ladaptateur original (Perrault
ou Grimm), mais toutes contiennent le dnouement heureux de Grimm. Quatre
dentre elles, non seulement ne prcisent pas leur origine, mais ne spcient pas
sil sagit de traductions, au sens strict du mot, ou bien dadaptations, et ne
contiennent aucune autre rfrence, hormis le titre de lhistoire. Trois versions se
disent des adaptations (nous navons pas trouv de traductions des originaux).
De nombreuses caractristiques signales auparavant persistent dans la
plupart de ces ditions actuelles du conte Le Petit Chaperon Rouge, savoir: de
nouveaux conseils et des leons de morale sont donns lenfant par sa mre;
des promenades et des jeux avec ses petits amis les btes marquent le chemin du
Petit Chaperon Rouge, qui se dirige vers la maison de sa grand-mre, joyeuse et
bondissante; le loup ne mange ni la grand-mre ni la petite lle; un excs de
diminutifs est prsent dans la plupart des descriptions et des dialogues.
On se rend compte alors que les transformations de notre socit nont pas
t susantes pour altrer la situation de la traduction (ou de ladaptation) des
contes de fes au Brsil, un pays o le genre littrature enfantine a atteint,
comme nous lavons dit, un dveloppement dmesur, notamment partir des
annes soixante-dix.6
Quelle conclusion tirer de toute cette tude?
Lanalyse de toutes ces traductions et adaptations nous montre quelles prsuppo-
sent un lecteur innocent, ingnu, fragile, infrieur, incapable de regarder certaines
scnes ou de comprendre un langage moins simple et aectif . Si le texte de
Perrault, publi au XVIIe sicle, comme nous lavons vu, prsupposait dj un
lecteur peu form, qui avait besoin davoir une bonne formation, et si le texte de
Grimm, au dbut du XIXe sicle, mettait en valeur des leons de bon comporte-
ment, de telles productions intensient considrablement le ton moralisateur et
ce qui est pire, quand il est question de ralisations littraires banalisent la
trame, en utilisant un langage strotyp et soi-disant motif, qui rduit lunivers
de lenfant, conu, avant tout, comme un tre non seulement naf, mais bte.
Mais comment concevoir lenfant aujourdhui?
Comme il est impossible, mon sens, de dnir lenfant contemporain ,
188 ANA MARIA CLARK PERES

paradigme de tous les enfants, une approche qui les situe entre la tradition et
lavenir est peut-tre viable.
Insrs graduellement dans le monde complexe de lInformatique et ouverts
aux problmes nationaux et internationaux, dans cette re de la mondialisation,
beaucoup dentre eux anticipent un avenir dans lequel les exigences de technolo-
gies nouvelles et imprvisibles les rendront de plus en plus actifs et capables.
Ces enfants sintresseraient-ils encore des rcits forcment simplis et
didactiques? Des enfants qui reoivent quotidiennement des invitations trs
sductrices de jeu-vido sophistiqu et sont capables dutiliser avec dsinvolture
des logiciels parfois diciles mme pour les adultes, seraient-ils sensibles, je le
rpte, des histoires de petits lapins qui sautent, de petites lles trs obissan-
tes, de petits paniers, de petites grand-mres, de loups qui ne peuvent pas tre
mchants, des leons et des leons de morale?7
Les contes europens classiques auraient-ils perdu, alors, leur validit?
Je ne le crois pas. A mon avis, il y a quelque chose dimpossible briser: la
prsence, dans lenfance, de la pense magique, de lanimisme, de lattraction
pour le merveilleux, inhrents aux contes de fes anciens. Si comptent et bien
inform quil soit, lenfant ne cesse dtre attir par le monde magique et
incantatoire des rcits anciens qui, originaires des vieilles lgendes et des mythes,
sduisent lhumanit, depuis toujours.8
Si ce sens du merveilleux persiste, une sorte de tradition, cependant, pourrait
tre rompue: celle du didactisme de certaines productions dites littraires , qui
veulent considrer lenfant comme un tre infrieur et mme idiot, qui a besoin
de certaines traductions ou adaptations, pour pouvoir devenir un adulte savant
et intgre .
Dailleurs, cette distinction mme rigide entre littrature pour enfants et
littrature pour adultes ne serait-elle pas un vestige de la socit du XVIIe sicle
(qui avait besoin de sparer les deux mondes pour prserver la soi-disant
innocence enfantine) prsent dans la socit contemporaine?

Notes

1. Il faut souligner quil nexistait pas de particularisation de lenfance dans la socit mdivale
et dans le dbut de lre moderne. Ds quils pouvaient se passer des soins de la mre, les
enfants taient introduits dans la socit des adultes, et se mlaient eux. Tout le monde
samusait et travaillait ensemble, tant donn que les enfants nallaient pas rgulirement
lcole, destine un petit nombre de clercs.
2. Le Petit Chaperon Rouge, ainsi que dautres contes adapts par Perrault, ne contient pas de fes.
Ce seraient avant tout des contes merveilleux, o il y a lintervention de llment magique (par
 LA TRADUCTION DES CONTES DE FES 189

exemple, un loup qui parle). Cependant, on a lhabitude, en France et ailleurs, dappeler tous
ces rcits contes de fes .
3. Daprs les recherches de Paul Delarue et de Madame Rumpft, le conte des Frres Grimm
drive de celui de Perrault. Ils auraient recueilli leur version chez une conteuse dhistoires
dorigine franaise et insr dans le conte la partie nale dautres rcits, comme Le loup et les
sept chevreaux. (In Soriano 1977: 150.)
4. Le suxe inho , en portugais, correspond peu prs au suxe diminutif -ette ou -on ,
en franais.
5. Selon Cooper, Il tait une fois quivaut au Premier Temps des anciens Egyptiens, au temps
mythologique qui fait partie de toutes les traditions du monde . In Cooper 1986: 112.
6. Il est bien vrai que ce boom de livres pour les enfants a favoris, au Brsil, lmergence
duvres polmiques et stimulantes, appeles enfantines , mais, mon avis, la quantit de
titres publis dans les derniers temps, fait progressivement baisser la qualit des productions.
7. Au Brsil, pays marqu par une diversit socio-conomique considrable, tous les enfants ne
sont pas comme cela. Beaucoup dentre eux, ds leur plus jeune ge, doivent lutter contre la
pauvret et labandon pour subsister. Mais, ces enfants non plus ne peuvent pas tre considrs
comme des tres nafs et innocents, pas plus que fragiles et dbiles.
8. Pourquoi alors ne pas traduire, avec une certaine dlit, les adaptations de Perrault et de
Grimm, qui, ayant une structure narrative cohrente et ne prsentant pas un langage imbcile,
demeurent au cours des temps? Pourquoi ne pas les recrer ou les reconstruire en y
introduisant les exigences de la vie contemporaine? Et encore davantage: si les recherches nous
prsentent dj les versions populaires des contes qui ont servi de modle aux adaptations
enfantines de Perrault, pourquoi ne pas les traduire, permettant ainsi aux enfants davoir accs
dautres versions qui nont pas t crites spcialement pour eux?

Works cited

En allemand
Grimm, B. 1974. Rotkppchen. Kinder und Hausmrchen. Frankfurt: Insel.
En franais
Perrault, C. 1984. Le Petit Chaperon Rouge. Contes de Perrault. 7 me tirage. Paris: Grnd.
En portugais
dAguiar, C. D. 1975. O Chapeuzinho Vermelho. Rio de Janeiro: Tecnoprint.
Bandeira, P. Chapeuzinho Vermelho. Alegria, Suplemento da Revista Cludia, So
Paulo, Abril, 262, 36.
Bilder, P. 1983. Chapeuzinho Vermelho. 3a. ed. So Paulo: Melhoramentos (Srie Ver e
Ouvir).
Chapeuzinho Vermelho. Rio de Janeiro: Brasil-Amrica (Coleo Estrelinha).
1982. Chapeuzinho Vermelho. Rio de Janeiro: Cedibra (Coleo Grandes Contos).
1982. Chapeuzinho Vermelho. Rio de Janeiro: Cedibra (Coleo Contos Alegres).
1982. Chapeuzinho Vermelho. Rio de Janeiro: Cedibra (Coleo Contos Fantsticos).
1982. Chapeuzinho Vermelho. Orient. M. C. Machado. Rio de Janeiro: Cedibra.
190 ANA MARIA CLARK PERES

Chapeuzinho Vermelho. Adap. M. C. Machado. Rio de Janeiro: Cedibra.

Chapeuzinho Vermelho. Adap. A. C. Machado. Rio de Janeiro: Record (Coleo Joaninha).
Chapeuzinho Vermelho. Adap. M. Mazzetti. Rio de Janeiro: Rio de Janeiro: Record
(Coleo Contos da Carochinha).
Chapeuzinho Vermelho. Ao livro Tcnico (Coleo Panorama).
1983. Chapeuzinho Vermelho. So Paulo: Siciliano. (Coleo Contos Tradicionais).
Chapeuzinho Vermelho. In JACKSON, W. M.: Tesouro da Juventude. v. XIII. Rio de
Janeiro: Grca Editora Brasileira. 101103.
Chapeuzinho Vermelho. Contos e Lendas dos Irmos Grimm. Trad. I. M. Bonini. So
Paulo: EDIGRAF. v. V, 714.
1982. Chapeuzinho Vermelho. Rio de Janeiro: Bloch.
1995. Chapeuzinho Vermelho. 3a. ed. Recontado por Joo de Barro (Braguinha). Consul-
toria editorial de Nelly Novaes Coelho. So Paulo: Moderna.
1993. Chapeuzinho Vermelho. Blumenau: Eko (Coleo Fbulas de Ouro).
O Chapeuzinho Vermelho e Os Msicos de Bremen. Trad. R. Alberty. So Paulo: Verbo.
616.
1997. Chapeuzinho Vermelho. 11a. ed. Traduo e adaptao Srgio Luiz de Oliveira
Silva. Caramelo (Coleo Pom-Pom).
1997. Chapeuzinho Vermelho. Blumenau: Eko (Coleo Clssicos Ilustrados).
Chapeuzinho Vermelho. So Paulo: Girassol (Coleo Escreva e Pinte).
Chapeuzinho Vermelho. Blumenau: Todolivro (Coleo Fabulndia).
O Chapeuzinho Vermelho. Redao denitiva de J. Roberto Michelazzo. So Paulo:
Egria (Coleo Paraso Infantil).
Grimm, J. e W. Chapeuzinho Vermelho. Adap. R. Azevedo. So Paulo: Rideel (Srie
Minha Maleta Escolar).
Grimm, I. 1985. Chapeuzinho Vermelho. Trad. V. S. Khle. Porto Alegre: Kuarup
(Coleo Era uma vez).
Grimm, I. Chapeuzinho Vermelho. Narrat. B. S. Hazen. Trad. E. Nascimento. Rio de
Janeiro: Record (Coleo Contos de Fadas).
Grimm, I. 1997. Chapeuzinho Vermelho. 6a. ed. Recontado por Sylvie Rainaud. Traduzido
e adaptado por Irami B. Silva e Erdna Perugine Nahum. So Paulo: Scipione.
Machado, M. C. 1975. Chapeuzinho Vermelho. Contos de Grimm. Rio de Janeiro:
Tecnoprint.
Perrault, C. Chapeuzinho Vermelho. Adap. Naufer. Rio de Janeiro: Brasil-Amrica
(Coleo Samba-Lel).
Perrault, C. Chapeuzinho Vermelho. Adap. Hel. Rio de Janeiro: Brasil-Amrica (Coleo
Xuxuquinha).
Perrault, C. Chapeuzinho Vermelho. Adap. Hel e Naumim. Rio de Janeiro: Brasil-
Amrica (Coleo Xod).
Perrault, C. 1979. Chapeuzinho Vermelho. Miudinho (Edio Especial). Rio de Janeiro,
Brasil-Amrica, 4.
Perrault, C. Chapeuzinho Vermelho. Adap. P. Oliveira. Rio de Janeiro: Brasil-Amrica
(Coleo Abracadabra).
</TARGET "cla">

LA TRADUCTION DES CONTES DE FES 191

Perrault, C. 1982. Chapeuzinho Vermelho. Fbulas encantadas. 2a. ed. So Paulo:

Abril.
Perrault, C. O Chapeuzinho Vermelho. Trad. O. Ribeiro Neto. So Paulo: Editora Brasil
(Coleo Os mais belos contos de fadas).
Perrault, C. 1960. A Capinha Vermelha. Contos de fadas. LOBATO, M. So Paulo:
Brasiliense, 58.
Perrault, C. 1963. Chapeuzinho Vermelho. Contos de Perrault. 3a. ed. Trad. Olvia
Krhenbnle. So Paulo: Cultrix, 1114.
Perrault, C. 1968. Chapeuzinho Vermelho. Traduzido da edio francesa: Le Petit
Chaperon Rouge. So Paulo: Agir (Coleo Contos Tradicionais).
Watts, M. Chapeuzinho Vermelho. Trad. e adap. R. Rocha. Abril (Coleo Beija-Flor).
<TARGET "p4"
</TARGET "p4">DOCINFO

AUTHOR ""

TITLE "Investigating Translation Receivers"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

S IV

Investigating Translation Receivers

<TARGET "nor" DOCINFO

AUTHOR "Christiane Nord"

TITLE "What do We Know About the Target-Text Receiver?"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 19

What Do We Know
About the Target-Text Receiver?

Christiane Nord
Fachhochschule Magdeburg

Preliminary remarks

It is a common experience that somebody who sits down to write a text

whether it is a piece of scholarly prose, or an advertisement, or a love letter
has at least a vague idea of the audience they are addressing, and in text or
discourse linguistics there seems to be hardly any debate on the assumption that
the addressee, or rather: the idea of the addressee the author has in mind, is a
very important (if not the most important) criterion guiding the writers stylistic
or linguistic decisions. If a text is to be functional for a certain person or group
of persons, it has to be tailored to their needs and expectations. An elastic text
intended to t all receivers and all sorts of purposes is bound to be equally unt
for any of them, and a specic purpose is best achieved by a text specically
designed for this occasion.
What is common practice in technical writing and advertising, is not so
unanimously accepted in translation. There are still quite a few scholars who
consider the source text (whatever that may be) to be the yardstick by which
they measure the quality of a translation (cf. Koller 1995), and although many of
them agree to the above-mentioned principles of good writing in one moment,
they criticize functionalist approaches to translation on asking how do transla-
tors know who the target audience will be and what their expectations are.
There is, indeed, good reason to ask this question, and we shall try to nd an
answer or at least a method by which a translator may nd an answer in a
particular translation situation.
196 CHRISTIANE NORD

Receiver vs. addressee

If it is indeed the receivers who decide whether a particular text is acceptable

and serves their purpose, that is, whether it is adequate with regard to the
requirements of the situation (Nord 1997: 34.), we have no means of guarantee-
ing text functionality. There is no way to predict who will eventually receive the
text we cannot prevent anybody from reading the text and nding it inade-
quate to their purpose. But, then, professional text producers do not want to
satisfy just any receiver, they usually address a particular target group, and they
spend a lot of eort (and money) to nd out about the interests, emotions,
knowledge, education, hobbies etc. of this particular group. The same applies to
translation, and thus we have to distinguish between the receivers of a translation
(the individual persons who actually read or listen to it) and the addressees of a
translation (the type or prototype of person to whom it is addressed).
Translators, for example, are real receivers of the source text, but they are
not normally the addressees of a source language text, at least if they are
members of the target culture.
The addressee (or target audience) of any text or translation is not a real
person but a concept, an abstraction gained from the sum total of our communi-
cative experience, that is, from the vast number of characteristics of receivers we
have observed in previous communicative occurences that bear some analogy
with the one we are confronted with in a particular situation. The idea of
addressee we have in mind is determined by a kind of situation typology. In a
type-A situation we expect that a text intended to full functions B and C would
have to be addressed to a type-D audience, as depicted in Figure 1 below. There-
fore, we could say that what we have in mind is a prole of the intended
target-text receiver.

Figure 1: Target-text addressee prole

 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 197

In what way does the addressee profile influence text production?

There are three aspects in text production where the inuence of the addressee
prole is felt most:
how much and which information is presented in the text,
how is this information structured, and
which linguistic and stylistic devices are used to present the selected
information.
The rst aspect refers to the knowledge presuppositions. In order to be accept-
able, there has to be achieved an equilibrium between old, i.e. previously known,
information and new information. It is fairly obvious that a text that contains too
much information already known to the receiver will be considered redundant
and of little interest. But if the amount of new information is too high, the text
may be dicult to understand or even incomprehensible to the reader. We have
to ask, therefore, which amount of information is necessary to make the text
understandable, and which details can be expected to belong to the readers
general or culture-specic world knowledge and need not be verbalized in the
text. What will be an adequate degree of specicity in order to achieve the
intended communicative functions? Let us look at an example. In a package of
Danish chocolates we nd a slip of paper with a guarantee cupon in ve
languages, of which I reproduce the Danish original and the English, German
and Spanish translations, indicating the dierences with regard to knowledge
presuppositions by italics.
198 CHRISTIANE NORD
Example 1: Danish chocolates
Anthon Berg har altid bestrbt sig for
at levere varer af neste kvalitet.
Chokoladen kann dog tage skade ved
tryk eller std, eller den kann
pvirkes af varme, strk kulde eller
fugtighed. Sdanne pvirkner lam
fremcalde et grligt skr p over-
aden, hvilket dog ikke behver at
pvirke smagen og heller ikke
behver at betyde, at varen er
gammel eller beskadiget.
Skulle der imidlertid vre noget, De
ikke er tilfreds med, s send venligst
sken med den beskadigede vare
sammen med denne garantiseddel i
udfyldt stand til Anthon Berg (se
bagsiden).
Anthon Berg, Kopenhagen, seit 1884
berhmte Schokoladenfabrik, hat
ihren Ruf auf Verwendung feinster
Rohstoe, strengster Qualitts-
kontrolle und sorgfltiger Ver-
packung aufgebaut.
Schokolade kann jedoch auf dem
Transport Beschdigungen erleiden
oder unter dem Einuss von
Wrme, Klte oder Feuchtigkeit
einen blulich-grauen Schimmer
bekommen, der indessen Qualitt
und Geschmack in keiner Weise
beeintrchtigt. In allen solchen
Fllen, die sich unserer Kontrolle
entziehen, bitten wir Sie,
umseitigen Garantieschein auszufl-
len und ihn zusammen mit der
Packung und dem Rest ihres Inhalts
an uns zurckzusenden. Wir werden
Ihnen alsdann umgehend Ersatz
schicken.
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 199

Anthon Berg of Copenhagen, Den- Anthon Berg de Copenhague,

mark, famous chocolate makers since
1884, has built its reputation on the
exclusive use of the nest raw mate-
rials available, combined with the
strictest quality control and most
careful packaging.
Should the chocolate nevertheless
reach you in less than perfect condi-
tion, for example because of damage
during shipment or discoloration due
to extreme changes in temperature
both conditions beyond our control
we invite you to request a replace-
ment of the package. Please return
this coupon (outside Europe via air-
mail) and as much of the package as
practical and state where and when
bought, price paid, and your reason
for dissatisfaction (see other side).
The terms of this guarantee do not
aect your statutory rights.
Dinamarca, famosos chocolateros
desde 1884, debe su buena repu-
tacin al uso exclusivo de las ms
nas materias primas, combinado
esto con el ms estricto control de
calidad y un empaquetado suma-
mente cuidadoso.
Sin embargo, el chocolate puede
estropearse durante el transporte o
bien por cambios de temperatura,
que podra crear un tinte grisceo
en la supercie. A pesar de ello
esto es inofensivo mantenindose el
mismo sabor.
En el caso de que, a pesar de ello,
Vd. no estuviese satisfecho con
nuestro producto, le rogamos
devolver el envase con el rema-
nente del chocolate junto con este
bono de garanta debidamente
llenado con su nombre y direccin,
nombre del producto, fecha de la
compra, nombre y ubicacin de la
tienda acs [!] como la causa de su
reclamacn a [the coupon follows
immediately]

For Danish receivers, Anthon Berg needs no explication, neither with respect
to its importance and fame nor with respect to its localization. Obviously, Danish
consumers are expected to have bought the product in Denmark, therefore careful
packaging is not mentioned either, whereas English-speaking receivers are even
asked to return the coupon via airmail. German readers are expected to know
where Copenhagen is, whereas for those readers of the English, French and
Spanish texts living in America, Asia or Africa, the reference to Denmark makes
damaging much more likely and, perhaps, even pardonable.
The second aspect refers to the textual macrostructure, that is, the order in
which information bits are presented, including the distribution of topic and
comment elements. A linear thematic progression makes understanding easier,
200 CHRISTIANE NORD

but may also lead to a lack of tension or suspense. Folk tales or childrens
stories very often have a linear progression, as is shown in Example 2. By
changing the order of the topical elements, we produce a text that lacks cohesion.
The reader then has to invest much more eort on nding the narrative thread.
Example 2: A Spanish folk tale
The rst sentences taken from a Spanish folk tale present an absolutely linear
thematic progression (see Figure 2a):
Era un caballero (T1) que sali por el mundo alante a buscar la vida y las
aventuras (C1). Y caminando, caminando (T2) se encontr con tres [!] animales,
un len, un galgo, un guila y una hormiga (C2) que (T3) estaban pelendose por
una era muerta que haban encontrado en el camino (C3).
[Once upon a time] there was a nobleman who went out into the world to seek
life and adventures. As he was wandering along, he met three [!] animals, a lion,
a greyhound, an eagle and an ant, who were struggling over a carcass they had
found on the road. (T = topic, C = comment)

T1 C1

T2 C2

T3 C3

Figure 2a. Linear thematic progression

Por el mundo alante sali (C1a) un caballero (T1) a buscar la vida y las aventuras
(C1b). Un len, un galgo, un guila y una hormiga (T2) estaban pelendose por
una era muerta (C2a). Caminando, caminando, (C3a) el caballero (T3=T1) se
encontr con los animales (C3b =T2), que (T4 = T2) haban encontrado la era
muerta (C4a = C2a) en el camino (C4b C3a C1a ).
The third aspect refers to the style norms and conventions expected in certain
text types and genres. Reading a text that follows conventional syntactic and
lexical patterns, the reader can concentrate on the contents of the message. On
the contrary, stylistic originality attracts the readers attention to the form in
which the message is presented. Therefore, any text producer has to take into
consideration the specic formal and stylistic patterns the reader may expect in
a text even in those cases where they decide precisely to avoid conventional style.
Looking at the texts reproduced in Example 1, we nd a number of
typical patterns both in the English and the German translations (should the
chocolate reach you in less than prefect condition, conditions beyond our
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 201

C1a T1 C1b

T2 C2a

C3a T3 C3b

T4 C4a C4b

Figure 2b. Inverted thematic progression

control, The terms of this guarantee, bitten wir Sie, umseitigen Garantie-
schein auszufllen, Schokolade kann jedoch), whereas the Spanish version
does not conform to conventional patterning.

Jumping the culture barrier

All the considerations mentioned above referred to intracultural communication.

In intercultural communication, a fourth aspect has to be taken into account. The
source-text producer and the target-text addressee belong to two dierent
cultures. In their role as target-text producers, translators cannot just adopt the
linguistic markers of addressee-orientation found in the source text, reproducing
them in the target language. In any translation task, whether it requires a
documentary or an instrumental translation (cf. Nord 1997: 47.), translators are
bound to jump the culture barrier, and after this jump, they will nd them-
selves in target-culture territory, as is shown in Figure 3.
What interests us most, in this context, is the phenomenon of intertextuality,
as is illustrated by Figure 3. Each text forms part of the text repertoire of, at
least, one particular culture (what we call world literature, for example,
belongs to various repertoires). All the texts belonging to the same repertoire are
interrelated with each other, either forming groups of texts which share the same
or similar communicative functions and, therefore, the same or similar formal
characteristics (i.e., text types or genres) or, in other cases, standing apart from
the other texts or text groups precisely because they dont share any characteristics
with them.
202 CHRISTIANE NORD

SOURCE CULTURE TARGET CULTURE

SOURCE TEXT TARGET TEXT

Figure 3: Jumping the culture barrier

When a translated text enters the intertextual system of the target culture, it
has to nd its proper place, either in a particular genre or text type or in a
special category consisting of all translated texts, perhaps with sub-groups
according to the source languages from which they have been translated or a
particular translator by which they have been translated. Some texts may belong
to the overlapping section of two or even more categories (for example, an
instruction manual translated from English by Mary Jones). And there may be
categories which (at a particular moment in time) do not contain any element
whatsoever (see Figure 4).

X XY Y Y

X X X XY Y Y

X XZ XYZ Y
XZ YZ

Z Z Z Z Z Z

ZZZ

Figure 4: Culture-specic intertextuality

If intertextuality within a particular culture forms such a complicated system, it

is obvious that one individual member of this particular culture cannot be
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 203

familiar with all categories and subcategories of texts. This explains, in part, the
frequent cases of dissent between teachers and students in translation classes
about what is normal in a particular text type or genre. Moreover, our textual
knowledge is, to a great extent, passive. That is, we intuitively recognize
certain formal, stylistic or structural characteristics as acceptable or function-
al (or inacceptable and dysfunctional) in a particular text, but we nd it
very dicult to produce a text that will have precisely these (acceptable,
functional) characteristics intuitively recognized by somebody else.
Therefore, what is needed is an intersubjective reference system which
provides reliable information about
the information structures and
the formal and stylistic characteristics
that are considered acceptable and functional, respectively, in a particular text (or
text category) by a large number or, possibly, the majority of text users. To
obtain this information, we need a corpus of parallel texts, preferably in electron-
ic form, which should be composed of three categories of text:
original texts belonging to a particular text type X, e.g. patents or instruc-
tion manuals or scholarly articles, in culture A,
original texts belonging to the same text type in culture B, and
professional translations from A to B (and/or viceversa) of texts belonging
to this text type.
This latter part, i.e. the translation corpus, is intended to provide information on
whether, in a particular culture, existing professional translations that have been
accepted by revisors and clients
reproduce the characeristics of this text type typical of the source culture, or
conform to the patterns of this text type typical of the target culture, or
show characteristics not found in original texts of this type in either culture,
but shared with other translations based on one particular or various source
cultures. This last aspect makes some scholars believe that translations form
a text type of their own (cf. Dressler 1975: 98 or Bhler 1984: 253).

Analyzing the expectations of the target addressee

The aim of this corpus-based project is not to analyse and compare linguistic
structures, as in Comparative Linguistics, but to nd out (and compare) the form,
distribution and frequency of certain communicative acts in two cultures in order
204 CHRISTIANE NORD

to be able to better predict the expectations of certain groups of addressees when

they are confronted with certain text types or genres in certain communicative
situations.
The concept of culture we use here is a dynamic concept. Culture is a
complex system. It can be subdivided into paraculture (i.e. the norms, rules and
conventions valid for an entire society), diaculture (i.e. the norms, rules and
conventions for a particular group within the society, such as a club, a rm or a
regional entity) and even idioculture (the culture of an individual person as
opposed to other individuals, cf. Ammann 1989: 39f.). A culture cannot simply
be equated with a language area since there may be dierences of (linguistic or
non-linguistic) behaviour within a language area (for example, the dierent use
of titles like doctor and professor in Germans and Austrians), and there may be
similarities of (non-linguistic) behaviour (e.g., value systems) between members
of two dierent language communities. In modern multicultural societies we
cannot even say that a town or a street represents a single homogeneous culture.
But a translator does not translate in abstracto. Translators mediate in concrete
situations between people belonging to rather clear-cut diacultures (e.g., North
American civil engineers and Spanish civil engineers). In such situations, most
of the cultural dierences between the corresponding paracultures will be
irrelevant. What is relevant to translation is only the set of dierences with
regard to the communicative activity in question (for example, exchanging views
on a particular method in road construction). As Michael Agar, a North American
anthropologist who worked as an intercultural practitioner in Mexico, put it,
Culture is not something people have; it is something that lls the spaces
between them. And culture is not an exhaustive description of anything; it
focuses on dierences, dierences that can vary from task to task and group
to group. (Agar 1992: 11)

These dierences, which Agar calls rich points (Agar 1991: 168), will be
analysed in the corpus. The tertium comparationis of such an analysis will be any
communicative function or subfunction represented by behaviour units or
culturemes. According to Vermeer (1983: 8), a cultureme is a social phenome-
non of a culture X that is regarded as relevant by the members of this culture
and, when compared with a corresponding social phenomenon in a culture Y, is
found to be specic to culture X. The relevant parameters of the comparison will
be form, distribution and frequency, that is: in which situation do text producers
prefer (and text receivers, therefore, expect) which form with which frequency.
In the following sections, we will look at a few pilote studies based on
this concept.
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 205

Audience-orientation in English and German student textbooks

One of my students at the University of Hildesheim (Germany) analysed the

markers of audience-orientation in a small corpus consisting of
5 English textbooks for university students on Data-Base Management,
Electronic Measurement Techniques, Industrial Robot Applications and
Electronic Digital Components, published in the United States,
5 German textbooks for university students on similar subjects, published in
Germany, and
the translations of the American textbooks into German, published by
German publishers.
The study was focused on (a) text segments which verbalized audience orienta-
tion (such as: metacommunicative and metalinguistic utterances, explicit referenc-
es to the author and/or readership) and (b) segments that would have allowed
such orientation but did not verbalize it. The results showed that, in this small
corpus, there were a considerable number of divergences with regard to form,
distribution and frequency between the English and German originals. However,
the translations followed the source-culture patterns rather closely, although this
strategy more than once had a rather negative eect on the comprehensibility
and stylistic acceptability of the texts. But evidently, they had been considered
functional by the publisher.

Cultural references in German and Spanish newspaper reports

Another student analysed references to foreign cultures in newspaper reports.

This study, which had a methodological and pedagogical focus, was based on
two parallel corpora, one of German newspaper reports dealing with Spain, and
one of Spanish newspaper reports dealing with Germany. After identifying the
cultural references, the author analysed (1) the linguistic forms by which they
were represented in the other culture (such as loanword, calque, literal transla-
tion, translation couplets etc.), (2) the quality and quantity of additional cultural
information (expansions) oered to the readers, and (3) the linguistic form of
these expansions (e.g., explicit explanations by apposition, relative clause,
parentheses, adjectives or adverbs, or implicit explanations in the immediate
context etc.).
206 CHRISTIANE NORD

Metacommunicative utterances in German, French and Spanish textbooks

The last study I would like to comment on was inspired by the diploma thesis on
audience orientation mentioned above. It was based on a corpus consisting of 30
original introductory textbooks for rst-year university students, published in
Germany, France and Spain (10 of each culture). Supposing that the form,
frequency and distribution of metacommunicative utterances did probably not
correlate strongly with subject elds, I included a random range of elds:
economics, linguistics, terminology, translation studies, information technology.
According to Lauer (1986: 33) and Techtmeier (1984: 133), metacommuni-
cation may be dened as the sum total of verbal and nonverbal means used by
the sender/author to comment on both the conditions and factors of the commu-
nicative situation and the interacting partners (author and audience) and their
relationship. Metacommunication aims at monitoring ones own communicative
activities and the (expected, anticipated) reactions of the audience, which have to
be in line with the intended purpose of the communicative act, and thus repre-
sents an ecient way of controlling the social interaction that develops between
author and audience.
When teaching in the classroom, a positive interaction is vital for the
achievement of the intended pedagogical aim or purpose. Metacommunication
may stabilize the relationship between teacher and learner, which is always
asymmetrical: the teacher has some knowledge which the learners do not have
but have to acquire in order to pass their exams. Baumann (1992: 38) states that
the frequency of metacommunicative utterances correlates with the degree of
symmetry between sender and receiver with regard to knowledge: In asymmetri-
cal communication acts, metacommunicative utterances tend to be much more
frequent than in symmetrical communication acts. The greater the equilibrium,
the fewer metacommunicative utterances. Moreover, it aims at keeping alive the
students interest in the subject and producing a relaxed atmosphere that makes
learning easier and more fun. Textbooks are a kind of simulation of classroom
teaching. Therefore, they are usually rich in metacommunicative utterances, as is
emphasized by Gpferich (1995: 405). Here, metacommunication is a subfunc-
tion of the phatic function (cf. Nord 1997: 44.), which aims at establishing,
maintaining or ending contact between sender and receiver, dening their status
and the relationship holding between them (cf. Jakobson 1960).
Experience shows that, as a general rule, the phatic function relies on the
conventionality of the linguistic, non-linguistic and paralinguistic means used in
a particular situation, such as ritualized small talk about the weather or salutation
formulae in letters (for phatic function in titles see also Nord 1993: 102). In
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 207

these cases, unconventionality of form strikes the eye of the reader and makes
them think the author had a special reason for saying something precisely in that
way. A phatic utterance meant as a mere oer of contact may be interpreted
as referential, expressive or even appellative if its form does not correspond to
the receivers expectation of conventional behaviour.
These considerations can also be applied to metacommunication in student
textbooks. Since the situation is the same in all cases (teacher gives students
necessary information on limited subject in written form), any dierences in
form, frequency and distribution of metacommunicative utterances in our corpus
can be considered as culture-specic.
Example 3: Conventionality of metacommunication
Personal opinion may be expressed by an adverbial phrase meiner Meinung
nach, by a verb in the rst person singular creo que or by a verb in the rst
person plural nous pensons que.
Examples can be introduced by z. B. (= zum Beispiel), veamos un caso
ejemplar or titre dexemple.

From a methodological point of view, metacommunicative utterances can refer to

the factors and conditions of the communicative situation, such as place,
moment or motive of communication,
the participants of the interaction (sender, receiver, perhaps third persons as
observers) or
the text as a tool in communication, whose characteristics include its audi-
ence-directedness, its medial properties, the subject matter, its macro- and
microstructures or various linguistic aspects such as terminology, similes
and metaphores, etc.
These relations are represented in Figure 5.
Following the schematic representation (Figure 5), we may distinguish three
basic types of metacommunication: (1) references to the situation (in this
country, nowadays), (2) references to the participants (I/we/, you/rhe-
torical questions), and (3) references to the text (this book, our study).
References to the situation include elements referring to text-internal deixis,
which is verbalized either by local or by temporal metaphors (e.g., as we have
explained above, we shall come back to this point later).
Example 4: Situational references (deixis)
Nuestra intencin, aqu, es ( local deixis)
On ne peut songer fournir ici ( local deixis)
Dies ist hier irrelevant / interessiert hier nicht ( local deixis)
208 CHRISTIANE NORD

Figure 5: References of metacommunication

Wie wir bereits gesehen haben ( temporal deixis)

Acabamos de aprender ( temporal deixis)
que lon retrouvera plus loin ( local deixis)

Example 5: Personal references

Ich habe mich bemht, verstndlich und verstehbar zu sein (sender: 1st person
sing.)
Querra expresar mi punto de vista (sender: 1st person sing.)
Ce mest une vrai joie de prsenter ce nouvel ouvrage (sender: 1st person
sing.)
Quelle est alors lutilit de la rhtorique aujourdhui? (asking receiver)
Il ne faut pas oublier non plus que (impersonal construction)
Von besonderem Interesse ist hier (impersonal construction)
Conviene subrayar que (impersonal construction)
Nous conseillons au lecteur de lire (sender: 1st pers.plur., receiver: 3rd
pers.sing.)
Que le lecteur nous pardonne nos lacunes (sender: 1st pers. plur., receiver:
3rd pers.sing.)
Dies mge verstndnisvoll verziehen werden (passive construction in
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 209

subjunctive, referring to the receiver)

Le lecteur stonnera / constatera (receiver: 3rd pers. sing.)

Example 6: References to the text

Das vorliegende Buch richtet sich an Lehrende, Lernende und Forschende
(medium: book, addressees: teachers, students, researchers)
Ich schreibe nicht denken (medium: writing, metalanguage)
Zu diesem Zweck sind die Ausfhrungen folgendermaen gegliedert: Zunchst
werden die Grundlagen errtert (Kap. 2). Danach wird XXX analysiert (Kap.
3). Anschlieend wird YYY dargestellt (Kap. 4). Eine kurze Zusammenfassung
und ein Literaturverzeichnis runden die Arbeit ab. (macrostructure, speech acts:
discuss, analyse, represent, abstract, round o)
La traductologie, pour utiliser le terme, nexiste pas (metalinguistic
reference to use of term)
Un libro como el presente, que ya en cualquier momento hubiera sido til
(medium: book)

Some other aspects I have analysed so far include

expressions of modesty (e.g., we may assume, allow me to),
explicit and implicit references to the reader, and
references to speech acts, such as quoting, giving an example, dealing
with a topic, placing emphasis on a particular aspect, introducing,
criticizing, and the like.
Example 7: Expressing modesty
wir drfen annehmen, ich mchte darauf hinweisen, ich habe versucht, ich
habe mir erlaubt (modal verbs: may, would like to, try)
Auf X sei hingewiesen, X sei betont (impersonal construction, in subjunctive)
nous essaierons / tenterons de, nous avons essay de.. (modal verbs: we
shall try to)
nous contenterons-nous de quelques brves remarques, on se bornera ici
indiquer (modal verbs: we shall limit ourselves to)
Bien que nous soyons conscient des limites inhrentes (excuse)
Espero contribuir a (modal verb: I hope)

However, the concept of metacommunication is limited to those utterances

referring to the communication in process, i.e. the communication going on
between sender and addressees. Utterances addressed to third persons, such as
acknowledgements, are excluded from this study, although they also oer very
interesting insights into the conventions of scholarly genres.
210 CHRISTIANE NORD

Conclusions

To conclude, I would like to present some of the ndings. The comparison of the
three corpora shows that the sender-audience relationship is much more empha-
sized in Spanish and French texts than in German. As could be expected, when
referring to their own intentions and actions, German authors prefer impersonal
constructions (passive, ist zu + innitive, impersonal pronoun man),
personication of the book, of problems, theories, and the like, even referring to
themselves in the third person (der Autor). Spanish and French authors prefer
personal constructions. Spanish authors alternate between the rst person singular
and plural, while French authors almost exclusively use the rst person plural,
occasionally on. It is interesting to note, moreover, that the rst person plural
in French and Spanish and the pronoun on in French include the audience, as
can be inferred from the verbs used in these contexts (Example 8).
Example 8: Implicit references to the reader
les notions nous seront utiles dans le prochain chapitre (= will be useful for
you)
considrons lexemple (= look at the following example)
on remarque, on verie (= you will nd)
Consideremos, por ejemplo (= think of)
Reexionemos: existe siempre ese primer medio? (= you might ask your-
selves if)
Man erinnere sich, man beachte, man vergleiche (= remember, consid-
er, compare)
X bleibt zu bedenken (= you should take into account)
Immer wird man zu dem Ergebnis kommen, da (= you will always nd)
Pinsese y se comprender (= if you think of, you will understand)

Apart from these implicit references, the frequency of direct references to the
readers and their expected or anticipated reactions is also much higher in Spanish
and French than in German texts, as is shown in Example 9.
Example 9: Explicit references to the reader
Intentamos proporcionar al lector (3rd person singular, object case)
El lector ms avisado podr advertir (3rd person singular, subject case)
Le lecteur constatera, stonnera (3rd person singular, subject case)
X soll es dem (interessierten) Leser ermglichen (3rd person singular, object
case)

Another stylistic marker pointing to a higher emphasis on sender-receiver

relationship is the use of direct questions, not only rhetorical questions implying
 WHAT DO WE KNOW ABOUT THE TARGET-TEXT RECEIVER? 211

the answer yes or no, but also open questions which serve to introduce a
new topic or sub-topic (topical questions). The latter are very frequent in the
French textbooks, a little less frequent in the Spanish textbooks, while they are
almost nonexistent in the German textbooks. German textbook authors prefer
indirect questions, if any.
Example 10: Asking questions
No nos ocultan las cifras una serie de realidades?
(rhetorical question, implied answer: yes, they do)
Faut-il ajouter que?
(rhetorical question, implied answer: no, it is not necessary)
Quelle est alors lutilit de la rhtorique?
(topical question, next subject: the use of rhetoric)
Que faut-il aujourdhui penser des sophistes?
(topical question, next subject: the sophists)
Wie stellt sich die Situation heute dar?
(topical question, next subject: the situation today)
Hier mssen wir uns fragen, wie die Situation sich heute darstellt.
(indirect topical question, next subject: the situation today)

The stronger emphasis on audience-orientation is sometimes even shown in the

title, as in the following examples. Instead of referring to the text function
(Introducing or Analyzing), some titles take a receiver perspective
(How to) or even address the reader directly.
Example 11: Metacommunication in book titles
Grundlegung der Translationstheorie (Laying the foundations of)
Introduction lments de (Introducing)
Anlisis del tercer Plan de Desarrollo (Analysing)
Comprendre les bases de donnes (Understanding data bases)
Comprendre les statistiques descriptives (Understanding statistics)
Lire le franais dhier (Reading Old French)
Para entender la msica (Understanding music)
Husle baue leicht gemacht! (How to build a house)
Teach yourself Modern Greek

Of course, these results are no proof, they are just an indicator of the culture-
specicity of metacommunicative behaviour in some German, French and
Spanish introductory textbooks for university students. It would be great to be
able, one day, to quantify the dierences, and, of course, to nd out whether
and to what extent they are taken into account by translators. But there is still
</TARGET "nor">

212 CHRISTIANE NORD

quite a long way to go until we know a little more about what the target-text
receivers expect.
<TARGET "mac" DOCINFO

AUTHOR "Rosemary Mackenzie"

TITLE "POSI-tive Thinking About Quality in Translator Training in Finland"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 20

POSI-tive Thinking About Quality in

Translator Training in Finland

Rosemary Mackenzie
University of Turku, Centre for Translation and Interpreting

Introduction

The ISO 9000-based quality systems applied to services are built on the principle
that the ultimate measure of quality is customer assessment of the service. The
POSI project (practical orientation of studies in translation and interpreting (T/I)
arose originally out of a need to improve the quality of the T/I graduates entering
the profession in Germany. This later grew into a pan-European project aimed at
a higher and more uniform quality of T/I training and a better match between the
training oered and the needs of the profession and the global market. The
current stage of the POSI project involved setting up national committees which
were to be as representative of the profession and market as possible, including
members from the training institutions, professionals and users of T/I services.
The committee set up in Finland consisted of ve members representing the
universities that oer specialised training in T/I (a rst degree), three representa-
tives from the Finnish Associations of Translators and Interpreters, Translation
Companies and Professional Translators, but only one member from a user of
translation services Nokia Communications. It was decided that the best way
to gather information about the needs of users was to conduct a survey using a
questionnaire. One of the purposes of the questionnaire was to nd out how the
users of T/I services assessed the quality of these services, how well they
considered the basic T/I training to meet the demands translators/interpreters face
in practical work situations and how they considered the training could be
improved. This information can be compared with the opinions of T/I service
providers on the same issues. This is the aspect of the questionnaire I shall
concentrate on here.
214 ROSEMARY MACKENZIE

The target groups: users and providers of T/I services

The target group of the users of T/I services consisted of 48 major companies in
Finland (from the list of the 500 biggest enterprises in Finland), 53 SMEs from
the Turku area engaged in foreign trade, 11 ministries and the information unit
of the Council of State and the administrative oces of the 10 biggest cities in
Finland. The suppliers of translation/interpretation services were 22 translation
companies (which from the point of view of training are classied as users), 20
of them members of the Finnish Association of Translation Companies, and 212
members of the Finnish Association of Translators and Interpreters. Of transla-
tors/interpreters 110 live in Finland and 102 reside at present in another Europe-
an country, some of them as employees or freelancers working for the EU. The
questionnaire was also sent to 3 people responsible for co-ordinating or supervis-
ing translation activities in the EU with the request that they forward the
questionnaire to translators and interpreters and their superiors. The questionnaire
sent to enterprises and to the public sector was to be answered by persons who
act as superiors of translators/interpreters, who outsource T/I services and/or are
responsible for quality control of translations. The users of T/I services also
received a copy of the translator/interpreter questionnaire to be distributed to the
T/I sta in various departments of the organisations.

Respondents

The number of respondents among users totalled 58, 19 major Finnish compa-
nies, 14 SMEs, 8 T/I companies, 15 ministries, 8 municipalities and 2 from the
EU. Thus 33 respondents were from the private sector and 25 from the public
sector. One interesting question is whether their expectations diered.
Among providers of T/I services the total number of respondents was 149,
122 translators and 16 working mainly as interpreters. Eleven were both transla-
tors and interpreters. Of the total number of respondents 106 were employed, 62
in the private and 44 in the public sector, while 43 were either self-employed or
freelancers.

Skills requirements: users views

In the question I deal with here the service users were asked to rate on a scale
from very important to not important at all the knowledge and skills (in
 POSI-TIVE THINKING ABOUT QUALITY 215

addition to language skills) they felt to be important for successful performance

of T/I tasks in their organisations: the skills proposed are profound knowledge of
ones own and foreign cultures and ability to apply this knowledge in performing
T/I tasks, ability to analyse communicative situations and design appropriate
texts, ability to communicate and negotiate with various specialists in order to
solve problems relating to content, terminology or communication, ability to
search and research information from literature and other sources and to assess
the reliability and applicability of the information for the task in hand, ability to
use modern computer hardware and software and modern data communication
technology, and the basics of terminology work and compilation of glossaries.
Finally, there was an open-ended question aimed at eliciting views on other
knowledge and skills required.
In the private sector the skills valued most highly were related to mastery
of computer and information technology, ability to communicate and negotiate
with subject specialists and ability to search and evaluate information, while in
the public sector the basics of terminology work and ability to search and
evaluate information were given most weight.
In answer to the open-ended question on other skills, respondents from both
the major companies and the SMEs emphasised the importance of understanding
technology in general (knowing the terminology of a eld is not enough).
Respondents from T/I companies found the following issues important: ability to
see the big picture, willingness to take responsibility, leadership, communica-
tion skills and technical knowledge and skills. A sense of style in the mother
tongue was also mentioned. In the public sector other knowledge and skills
mentioned were ability to co-operate and social skills. In addition, general
knowledge and mastery of a special eld, knowledge of society and how it
functions, and knowledge of administration and legislation were considered
important.

Development of translator training and further training

The organisations were asked what basic skills they felt to be lacking in newly
recruited translators. All the organisations pinpointed mastery of terminology as
being inadequate in their translators/interpreters, but this deciency is no doubt
remedied with experience as it is not ranked high in the skills required. The
major companies also pointed to a need for more profound knowledge of the
sector concerned. An ability to communicate eectively and to use the Internet
in nding information were also stressed. Special mention was made of decision-
216 ROSEMARY MACKENZIE

making in matters such as: what should be added to the text in order to ensure
eective communication, i.e. skill in transediting texts. Journalistic skills were
also considered desirable, i.e. the ability to write texts directly in a foreign
language rather than translating them. It was also claimed that translators/
interpreters are not suciently well informed about the functioning and language
of the economy. Further education in business economics, accounting, nance
and business contracts and related terminology were suggested as improvements
Of the SMEs, 2 respondents commented on this point, emphasising know-
ledge of technical terminology and basic understanding of technology. Also
mentioned was the importance of knowing how to tackle a new subject and the
skill to read and write between the lines. It was stated that further training
covering everything is naturally not a realistic target, but some introduction to
various elds could in the respondents opinion be provided.
The public sector respondents found many of the same skills to be lacking
in their new translators: terminology, subject knowledge, social skills and
teamwork. One aspect not mentioned elsewhere was accuracy and attention to
detail. Especially for interpreters self-presentation, knowledge of etiquette and
general knowledge were often felt to be lacking.
A total of 7 respondents representing T/I companies expressed their views
on development of training and further training needed. Creating possibilities for
on-the-job learning was found very important. Therefore a period of practical
training should be included in the basic education of translators/interpreters. In
addition to computer skills and subject knowledge the following aspects were
emphasised: matters of economics, productivity and its meaning for the employ-
er, customer orientation and exibility in adapting to changing demands. Further,
know-how in using existing sources of information and mastery of the means of
expression and the grammar of ones mother tongue were pointed out. In
addition to this, translators/interpreters should have the ability to consult custom-
ers (interpersonal skills) and the courage to question the source text when
necessary. The ability to assess the time required for and price of work and the
ability to meet deadlines were further points mentioned by T/I companies.

Skills: providers views

The questionnaire sent to translators and interpreters contained the same ques-
tions as discussed above asking them to rate the skills they need in their jobs.
The most heavily stressed skills were ability to search and evaluate information,
followed by a profound knowledge of ones own and other cultures, with
 POSI-TIVE THINKING ABOUT QUALITY 217

computer skills in third place. The least importance was attached to basics of
glossary and terminology work.
The translators/interpreters were also asked to mention other skills they
rated as important over and above those mentioned. The translators and interpret-
ers employed by companies and organisations mentioned social skills, ability to
communicate, co-operate and negotiate, PR skills, ability to make decisions
rapidly and to justify decisions, ability to concentrate, logical thinking and
analytical skills, exibility and adaptability to new situations, ability to edit and
revise texts, accuracy and speed, teamwork skills, leadership and management
skills as well as broad general and special knowledge economics and law
were the elds most often mentioned. Freelance and self-employed translators
stressed marketing and interpersonal skills, time management, a positive profes-
sional image, social skills cooperation, customer service and good relations
with experts in various elds and not surprisingly knowledge of business
routines, taxation, bookkeeping, invoicing and pricing.

Satisfaction with training

The translators and interpreters were asked how satised they were with the
training they had received. The majority felt their training had equipped them for
working life either well or mainly. Those who had received a university-
level training in T/I were more satised than those who had received another
training. When the translators and interpreters employed in companies and
organisations (106) were asked what they felt to have been missing in their
training, the most frequently mentioned lack was in the teaching of computer
skills for assisting translation and information searching. This is understandable
in the case of older graduates, but even fairly recent graduates complained of
this. Other areas where the training had been found inadequate were: the teaching
of subject knowledge in special elds, terminology research and information
acquisition, cooperation and teamwork with colleagues, editors, experts, custom-
ers etc., entrepreneurial skills including marketing, pricing, invoicing and market
knowledge in general. One translator mentioned knowledge of quality control and
quality criteria in translation. Many felt that not enough attention had been paid
to the mother tongue in teaching (especially interpreters): knowledge of dierent
text types was mentioned as were ability to write uently and logically. Ability
to revise the texts of others was an aspect that one respondent felt to be lacking
in her training. Speed and ability to work under stress were considered to have
been neglected (especially in interpreter training). In general many felt there had
218 ROSEMARY MACKENZIE

not been enough contact between the training institution and real working life in
their training.
In fact therefore, these translators/interpreters had experienced dissatisfaction
with almost every aspect of their training. It must be mentioned that T/I training
has developed by leaps and bounds in recent years, as many respondents also
mentioned. As might be expected, the freelance and self-employed translators
and interpreters (43) stressed the lack of training about the realities of life as an
entrepreneur matters such as founding a company, pricing, invoicing, taxation,
marketing and how to nd customers, management of real translator/client
relations (tight schedules etc.), pensions and so on. Teaching in computer skills
and information technology were mentioned most frequently as having been
neglected in training.

Development of training

When asked how the basic training of translators/interpreters should be devel-

oped, the most frequently chosen alternatives were as follows: translation courses
and seminar work should include genuine translation projects, cooperation
between training institutions and employers should be enhanced in the curriculum
design. The next most popular alternatives were cooperation with other vocation-
al institutes to improve knowledge in special elds and an obligatory period as
a trainee in a translation company.
Free-form answers about how translator training should be developed
naturally enough repeated many of the points mentioned above. As practical
suggestions many of the translators/interpreters employed by companies or
organisations suggested that teachers of translation should be practitioners, that
more real projects should be used in teaching, and that visits to translation
companies and lectures by visiting professionals, translators and others, visits to
libraries of other university or college departments (information research) should
be encouraged. Better opportunities should be created to study other subjects as
minors. One respondent proposed that training should be provided for (mature)
graduates from other elds (she felt that recent graduates from translations
schools are not suciently mature for the work). Another respondent wanted to
see translator training expanded to give a broad competence in communication
including translating, writing and consultation.
The freelance and self-employed translators suggested many of the same
measures for developing T/I training, but one suggestion was that voluntary
courses in entrepreneurship could be organised for those interested. Exchange of
 POSI-TIVE THINKING ABOUT QUALITY 219

information between the training institutions of dierent countries was also

suggested, as well as developing a tutor (or mentor) system, in which a profes-
sional translator gives a student guidance in practice this would probably
involve some kind of traineeship. The use of real or simulated situations in
translation classes was mentioned again, as well as the need for teachers to be
professional translators.

Professional development

When asked what kind of further training would be most useful to improve their
qualications in order to better meet the requirements of working life, the most
frequent response was deepening of subject knowledge in various elds (the
most commonly mentioned were economics, law and various elds of technolo-
gy). This was followed by improvement of language skills through intensive
courses or language immersion and teaching in the use of modern information
search methods (Internet and data banks). Training in software and the basics of
glossary terminology was surprisingly low in the rating. Probably working
translators/interpreters have acquired these skills on the job.
Other suggestions for further training included courses in administration of
projects, management and leadership, entrepreneurship, marketing skills, found-
ing a company, courses in communication, advertising and journalism. Interpret-
ers, especially in the service of the EU, felt the need for regular courses to keep
up their mother tongue skills. Traditional translation workshops and seminars in
translation were also suggested. Freelancers and self-employed translators wanted
more or less the same further training with the addition of simultaneous interpre-
tation practice.

Students

A preliminary survey carried out among advanced students at training institutions

in Finland aimed to nd out how satised students are that the training they
receive at present will equip them for working life later on. The questions were
based on points that members of the POSI work group considered important for
the practical orientation of training: Translation and interpretation teaching and
tests should be in context (use of real or simulated situations), teamwork,
resource research, self evaluation and revision, giving and receiving of feedback,
should be taught and practised, students should learn about the profession of T/I,
220 ROSEMARY MACKENZIE

there should be possibilities for internships and other links with business and
industry, and students should be taught to use the tools of the trade (information
technology).
The students were asked whether they felt that they had abundant, sucient
(but could be more) or too little opportunity to practice translation/interpretation
in real life or real-life-like situations, to practice working as a member of team,
nding information from various sources, evaluating their own translations and
those of others, and whether translation tests were arranged in such a way as to
give a realistic picture of their skills, whether they received sucient feedback
on their skills, whether they were given a suciently clear picture of the work
of a professional T/I, whether sucient possibilities for internships and other
links with business and industry were available, and nally whether they had
sucient teaching in the use of information technology and its applications to
translation. The last two questions were the only two to which the response was
clearly negative, i.e. the question about traineeships and the teaching of informa-
tion technology. With other aspects of translation teaching varying degrees of
satisfaction were expressed, but no major dissatisfaction, so that no conclusions
could be drawn on a national level.

Implications

I shall not propose a curriculum here, but only make a few proposals for
teaching T/I. It is obvious that all the skills felt to be important by employers of
translators/interpreters and the professionals themselves cannot be tted into a
single rst degree, especially if students are also expected to learn research
methods and write a research thesis. Many of the required skills, e.g. subject
knowledge can only be acquired through years of experience and/or study.
However, considering the importance given to subject knowledge in the survey,
some introduction to special elds should be retained in the curriculum. As T/I
training establishments cannot be expected to provide expert teaching in many
dierent disciplines there is a case for cooperation with other departments, and
students should have easy access to basic courses in subjects taught at other
faculties or institutes. Graduates will have to be prepared to supplement their
knowledge through further education.
IT skills in such areas as information and terminology management are
nowadays essential for all translators entering the profession and need to be part
of the training. Some knowledge of the more advanced computer-assisted
translation tools, such as translation memory tools, is an advantage. Many
</TARGET "mac">

POSI-TIVE THINKING ABOUT QUALITY 221

teachers also need training in these aspects of translation, and this stresses the
need for the development of further training and post-graduate degree program-
mes for translator trainers.
Practical translation classes can be made more motivating by assigning
students simulated and later real translation projects. In this way translators/
interpreters learn the skills of teamwork and cooperation with clients, experts and
colleagues, revision and quality control, i.e. the realities of professional life. The
teachers who manage these projects need to be professional translators with
recent experience in the eld. In this context students can also be given basic
information about such matters as contracts, deadlines, dealing with clients,
pricing etc., but more detailed knowledge, e.g. of bookkeeping, taxation etc. will
have to be acquired through further education. However, a voluntary course in
entrepreneurship for those interested is a feasible suggestion.
The question of internships is considered of the utmost importance by T/I
students, by practising translators and by users. Translation companies have
expressed willingness to take on interns, but there are rather few translation
companies with the resources to do so. The University of Turku Centre for
Translation and Interpreting is setting up a pilot project in 1999, in which the
ground rules will be laid down for translation companies taking interns and for
students and supervisors involved in internships.
<TARGET "ada" DOCINFO

AUTHOR "Beverly Adab"

TITLE "Towards a More Systematic Approach to the Translation of Advertising Texts"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 21

Towards a More Systematic Approach to the

Translation of Advertising Texts

Beverly Adab
Aston University, Birmingham

Introduction

This paper seeks to oer guidelines for the translation of advertising, by looking
at examples of practice and relating these to the communicative aim(s) of
advertising messages. Limits of time and space mean that reported ndings must
of necessity be summative. The purpose of the study, as reported in this paper,
was therefore to compare Source and Target versions of a corpus of advertising
texts, in order to speculate on:
The nature of the dierent factors of text production: these include compa-
ny practice and translation brief, translator prole and place of professional
activity
The nature of the overall approach apparently adopted by individual
translators in the specic circumstances of text production
The potential for communicative success of each target language text,
produced from an original source language text
The potential contribution of the translator in enhancing communicative
eectiveness in the advertising message.
Advertising messages have a similar function across languages and cultures; they
rely on a primarily persuasive form of discourse, based on socio-cultural values
(De Mooij 1998), and use appropriate linguistic and other strategies to achieve
this function. Such texts therefore contain a number of language- and culture-
specic mechanisms, values and associations, which represent potential problems
for translation. Studying how these problems have been dealt with in actual
translation performance could therefore provide insights into the approaches
adopted by individual translators. Hence the use of a comparative analytical
224 BEVERLY ADAB

approach in accordance with the principles of Descriptive Translation Studies

(DTS, Toury 1995; Holmes 1971).
By adopting a socio-contextual perspective within a DTS framework,
analysis can oer insights into the way in which dierent, yet interdependent,
constraints on the translation process can also have an impact on the product and
successful text function. The DTS approach allows the gradual reconstruction of
translation decisions and the constraints under which they were made (Toury
1995: 8688). Toury emphasises the predictive, not prescriptive, value of such
analysis; being able to identify post facto which approaches and/or individual
strategies might be likely to enhance communicative eectiveness and might be
a useful aid to increased translatorial competence.
Advertising texts intended for translation can be produced according to the
following principles: rstly, globalization, where a text adopts the lowest
common denominators (basic values) for product promotion for example, the
Coca Cola approach, one size ts all. Secondly, localization, which is less cost-
eective, since it involves adaptation of the ST based on culture-and language-
specic mechanisms (e.g. computer software). Probably, the optimum compro-
mise is glocalization the production of a globally relevant ST, based on a
message that will have similar impact across dierent cultural contexts. Such
messages will require minimum adaptation, mainly at the level of style and use
of language, to be relevant for any single target community.
One problem in the translation of advertising is the need to realise the
importance of anticipating potential translation problems at the moment of ST
production. This is especially relevant when the translators brief is likely to
involve producing a ST-induced TT which implies the expectation, on the part of
the text commissioner, of a signicant degree of correspondence of content and
form between ST and TT (Nord 1991). The ST author may not have the
necessary translation-related competence (language, culture, transfer competen-
cies) to avoid problems for translation by avoiding, where possible, any reliance
on Source language and culture-specic imagery, references and use of language.
These features may in fact form an integral part of the SL message as dened by
the company, e.g. messages which exploit local or national attractions and
present them according to culture-specic perspectives. The degree of freedom
allowed to the individual translator in the translation process from SLT to TLT
is another important factor of multilingual text production. In accordance with
Holz-Mnttri (1984), the study seeks to encourage an awareness of the role of
the translator and the need for his or her participation at the stage of ST produc-
tion, rather than simply as a human translation tool post facto.
Evaluation of the intended impact of the ST is an integral element of Nords
 THE TRANSLATION OF ADVERTISING TEXTS 225

Translator Oriented ST Analysis (T.O.S.T.A.) (Nord 1991), which encourages the

translator to identify salient features needing to be preserved or recreated in the
TLT, depending on the intended relationship between ST and TT. Analysis of
how this is achieved through text structure, imagery, metaphor and underlying
socio-cultural values is inevitably subjective in part (Chesterman1998). Neverthe-
less, applying Nords T.O.S.T.A. approach to the ST analysis of key intra- and
extra-textual features can help the translator to choose (what s/he considers the
most relevant) corresponding TL units (textual, meaning, language). This form
of initial planning can enhance the potential for successful reception, in the target
culture, of a TT produced on the basis of a given ST, where both have a similar
intended function. TT reception will also depend on conformity to text typolog-
ical conventions within the Target Culture (Trosborg 1997; Chesterman 1998).
Cook comments, in relation to intertextuality on the basis of corresponding text
type features, Each new ad is encountered through knowledge of thousands of
earlier ads (Cook 1992: 4) (cf. Neubert 1985, intertextuality). Evaluation of
potential response, by an experienced SL or TL reader, can be based on aware-
ness, on the part of the evaluator, of prototypical features of message construc-
tion, style and use of language for the text type within each culture. This
evaluation can be a useful preliminary indicator, to be supported by and tested
against post-production consumer surveys (Wilmshurst and Mackay 1999).

The corpus

All the texts in the corpus were designed to promote a non-essential service or
product. Cook (1992: 11) comments that, More expensive items, whose purchase
merits longer consideration, are prone to reason selling and therefore to longer
written copy. The study deliberately focused on messages consisting mainly or
solely of written copy, involving communication of associated values, which are
integral to product promotion, through language signs. Message interpretation for
such texts relies on Grices principle of co-operation between text producer and
text user (Grice 1975) to achieve the persuasive aim. If any images accompanied
the texts, these were simply logos or pictures of the product.
Since the appeal of these messages is based on associated qualities and
lifestyle, analysis could be expected to show evidence of some degree of
adaptation of culture-specic connotative values in order to preserve the intended
text impact. However, it was also anticipated that the ST might be contextualised
within a more international business culture, where fundamental domain-specic
values may not dier signicantly from one target group to another. This is the
226 BEVERLY ADAB

logical outcome of a growing Macworld culture. The McDonaldisation thesis

(Ritzer 1998) asserts inter alia that central planning of policy, as a measure of
cost control, leads to the elimination of intercultural dierences by the imposi-
tion of a dominant set of values from the source culture. There is evidence to
suggest that the world of advertising in France has been inuenced by values
originating in the USA business environment (Mattelart 1989); many of these
values are also present in UK advertising, particularly advertising produced by
multinational companies (De Mooij 1998). In each case, the translator will need
to draw on his/her intercultural competence to determine whether cross-cultural
adaptation is necessary.
The aim, as stated, was to study texts belonging to the extended written text
type, within the discourse of advertising, taking into account dierent constraints
on production. The selected texts promoted a wide range of products, from
holiday destinations (France, Ireland, Australia), to luxury products (watches,
furniture, china, lawn mowers, IT products, whisky), to forms of travel (car hire,
Eurostar) and nancial services; also included was material produced by a
translation agency. All the texts had to establish their USP, dierentiate
themselves from competitors and capture a segment of an active market. For
each ST, there was a corresponding TT, in the language pair French and English.
Both texts were supplied by the product or service provider.

Methodology

Detailed comparative and contrastive analysis was applied to corresponding ST

and TT units of textual meaning. This involved identication of similarities and
dierences of function of comparable text segments (units of meaning) from SL
and TL, in relation to communicative function and (judged) communicative
impact. Experience of (equivalent competence to) an educated native speaker in
this type of text in both Source and Target languages and cultures was essential
to this analysis.
Taking into account the role of norms in determining reader expectations
and ensuring eective text function (Toury 1995; Chesterman 1997, 1998), the
texts were examined for evidence of conformity to or deviation from standards
of text production as described by Neubert (1985). Individual strategies were
dened according to categories proposed by Translation Studies scholars such as
Vinay and Darbelnet (1958), Delisle (1980), Neubert (1985, 1992), Reiss and
Vermeer (1984), Nord (1991, 1997). On the basis of individual translation
strategy types it was possible to speculate on the overall approach adopted by
 THE TRANSLATION OF ADVERTISING TEXTS 227

each translator, as broadly functional or broadly linguistic, representing two quite

dierent frameworks for text production and ST-TT relationship. The stated
translation brief for each text, where known, was also taken into account at this
point. The potential success of each text was then predicted.
In inter-lingual advertising, much is written about the need to construct
individual messages for each target culture in preference to ST induced TT
production. Examples of practice are listed in the next section; these are repre-
sentative of wider feedback reporting a range of practices. It would appear from
the feedback that the default approach for untrained translators, as for text
commissioners having little or no translation awareness, is usually a linguistic
approach (Catford 1965), frequently without reference to socio-cultural context
or underlying values. According to prevailing views in the functionalist approach
to Translation Studies (Vermeer 1996; Nord 1997), a linguistic approach might
be expected to produce a less eective target text for this type, since the focus
on units of language may not also take into account features of context and
culture-specicity. In the functionalist approach, communicative eectiveness is
measured in terms of the intended and potential interaction between text user and
message; this approach also takes into account TL reader prole and needs. De
Mooij (1998) argues that this interaction will be aected in part by the underly-
ing myths and values used to construct the message.

The findings

Detailed analysis conrmed the above. The ndings for each text are evaluated
within the stated parameters dening text production, which include translator
prole as well as individual translation briefs reecting dierent overall
translation policies. A link can be postulated between these factors of text
production, the probable approach adopted (broadly linguistic or broadly
functional) and the potential success of message reception in the target culture.

Dierent translation policies: A summary of types of practice

1. Elida Gibb, producer of pharmaceutical and toiletries products, has a particu-

lar policy of conducting market research into the reception of dierent concepts,
as embodied in product slogans before launching a trial campaign. Due attention
is paid to national regulations governing product information to be included on
or inside packaging and information is given according to need and legal require-
228 BEVERLY ADAB

ments, but this does not require creative or culture-specic concept sensitivity,
simply technical accuracy. Thus the translation aspect of the product image is
included in the marketing process and may be delegated to an agency or to a
marketing expert who is also uent in a target language.
2. Softrans, a company specialising in the translation of computer software,
ensures that translators are trained in the use of the product. In this way transla-
tors are better able to select TL terms and forms of expression within the
constraints of time and space imposed by the task and by the nature of the
software. Translators build up repetition les using MAT to ensure uniform
language use and standardised terminology is agreed on with a representative of
the software producer in the target country, to guarantee appropriacy, acceptabili-
ty and conformity to existing use in that country. The company has an in-house
team of translators, complemented where necessary by translators working free-
lance, who are often former members of the Softrans in-house team. Translators
are recruited according to strict standards and have to submit sample work for
assessment prior to interview. This is a clear example of good practice at the
multinational level combined with the concept of localization.
3. A company which claims to proceed in a similar manner is the 4-T agency,
oering its translation services, which are claimed to be specially tailored to the
needs of the client. However, close examination of the actual style of the TL in
the corpus studied (TL in English) would seem to attest to a lack of linguistic
and socio-cultural competence on the part of the translator and of the agencys
translation reviewer.
4. The Rolex company, manufacturers of quality watches, have adopted a more
global approach, building their advertisements around famous people or types of
people who use the watches (e.g. Kiri Te Kanawa), or around events at which
Rolex is the ocial time-keeper (e.g. Wimbledon). The qualities of the watch
are described both overtly and by implication, through a process of analogy with
the stated qualities of the celebrity user. The same qualities recur in all the
advertisements, giving a recognisable structure to the Rolex campaign, a certain
predictability of style which would, in journalism, be described as a kind of
reader-attraction, creating and then fullling reader expectations. It would appear
that Rolex select their celebrity users on the basis of an international reputation,
thereby avoiding costly reformulation of a particular text and message, relying in
this way on successful translation of a source text into dierent target languages.
Rolex relies therefore on the global market having a system of shared myths and
values, so that target readers will recognise and wish to identify with those
qualities associated with the product in the advertising message. For this to
 THE TRANSLATION OF ADVERTISING TEXTS 229

succeed, the target languages must be suciently in contact for the experience
portrayed to be common to the dierent socio-linguistic cultures, and a means
for expressing the concepts must already exist in these languages.
5. Some companies appear to assume that the target language market will share
the expectations of the SL market, although some socio-cultural knowledge of
the TL community may indicate that this may not be the case, or at least, not to
such an extent as may have been assumed. The Qualcast advertisement would
seem to conrm this observation, as product appeal is based on the expectations
of the British user of the lawn mower, with all the underlying social connota-
tions, which may or may not be of relevance for other target language users.
6. Other companies or organisations, with a low budget, or which give low
priority to translation within their marketing strategy, prefer to farm out
translation work to local university students, who may or may not have any
training in marketing, inter-cultural communication and text type awareness.
Such is the case of the Clermont text.
7. Some companies, such as Eurotunnel, have a team of accredited free-lance
translators who ensure standardised terminology through repetition. Quality
control is undertaken by the public relations oce in the country where the TT
is produced, using native-speakers.
8. Other companies, such as Barclays Bank, may use a specialised agency for
translation work, but rely on company employees with a good level of compe-
tence in the language to proof-read and check translations. However, there is no
pre-specied or required standard of linguistic competence. Some agencies are
able to oer subject knowledge combined with marketing experience, technical
writing and translation, but their services are not always aordable or even
considered necessary.
9. Companies such as Twinings, on the other hand, rely entirely on recruiting
employees for the marketing department who also have near-native or native
competence in a target language, and much of the translation work is done by
these employees. In the case of the Old Charm text, translation into French is
undertaken by one of the Managing Directors, who is a French native speaker,
with quality control being also the responsibility of this person.
10. A further practice is to have a list of translators, who may or may not have
product-specic experience and over whom there is a minimal check of linguistic
competence. Product-specic knowledge (use and terminology) may accrue
through practice. Such is the case of some smaller Staordshire pottery and
ceramics companies (but not Minton, who use an agency).
230 BEVERLY ADAB

11. A new translation services company relies on a bank of tried and tested
professional translators, each a native speaker having expertise in a given eld
(usually information technology in this case). Translators are resident in their
country of origin, translate into their mother tongue, and receive and transmit
their work by fax or by electronic mail. This approach reduces overheads for the
service-provider, whilst ensuring high-quality work from highly motivated sta
with the relevant expertise in the subject area. Rewards are high in return for
eort and motivation; professional status is taken as the norm. No texts were
provided by this company.

The variables of the translation situation

Translator Prole
These details relate to the level, or lack of, professional translation training, also
to the role of the translator in text production and text evaluation.
A. Agency:
1. Undertakes translation and quality control.
2. Undertakes translation, company has input for marketing and informativity.
B. In-house translator:
1. Undertakes translation with some quality control and input for marketing
and informativity from the company, often from an agent of the company
in the target culture.
2. Undertakes translation with some quality control from other translators
within the company.
3. Undertakes translation and quality control.
C. In-house translator, non-linguist, usually marketing specialist:
1. Undertakes translation, with some quality control from the company, usually
from an agent of the company in the target culture.
2. Takes full responsibility for translation and quality control.
D. Freelance translator:
1. Undertakes translation, with some quality control from the company, usually
from an agent of the company in the target culture.
2. Takes full responsibility for translation and quality control.
In the cases studied, there was a range of translator proles, but little indication
of any systematic quality control of the TT.
 THE TRANSLATION OF ADVERTISING TEXTS 231

Translation brief:
According to information obtained from dierent text providers, the following
range of translation briefs could be said to represent the types of instructions
given to the translator, by either the text producer or the translation commission-
er, who may or may not be the same.
1. Total delity to ST form and content
2. Fidelity to ST formats and content but some exibility for form, to ensure
TL acceptability
3. Fidelity to ST function within norms of TL acceptability
4. Adaptation of ST for TL function in terms of text type norms, linguistic and
stylistic conventions
5. Working from a global mission, information and concepts but no xed ST.
The majority of texts studied were produced on the basis of the rst and second
types of instruction. This had the predictable limiting eect on potential choices
of TL forms to express SL units of meaning. This also meant that there was little
scope for adaptation of ST units according to Target Addressee prole, needs or
expectations.

Summary of findings

From a total of twenty-ve texts studied, the following points were noted:
Native Speaker of TL: 17
Internal or External quality control: all underwent some quality control, but to
diering degrees. There was no systematic criterion-referenced evaluation.
Situation of text production:
Agency: 19; In-house Translator (trained linguist): 2;
In-house Translator (non-linguist): 1; Freelance translator: 3.
Brief (see list above):
1st Brief: 0; 2nd Brief: 17; 3rd Brief: 4; 4th Brief: 2; 5th Brief: 0.
Overall approach (apparent):
Functional: 5; Linguistic: 15; Combination of both: 3.
As might be expected, translator prole (professional translation training, or lack
of this) was a factor in determining successful target text production. However,
in cases where experienced translators were faced with the obligation to create
a closely similar relationship of form and structure between Source and Target text,
this was observed to act as a constraint on freedom of choice of transfer strategies.
232 BEVERLY ADAB

Overall approach adopted and estimated degree of potential success/impact

This table seeks to link the overall approach, whether linguistic or functional, to
the potential success of text reception. Taken in conjunction with the other
information reported above, conclusions can be drawn about aspects of good
practice.

Text Overall Summary of success of approach

approach
01 F lack of appropriacy and acceptability in places
02 L closely follows SL but generally acceptable
03 L some interference from SL, generally acceptable
04 L generally successful but some less appropriate choices
05 L/F generally successful but some less appropriate choices
06 L generally successful but some less appropriate choices
07 L lack of acceptability of TL but works overall
08 L generally successful despite some culture-specic problems for TL reader
09 L lack of appropriacy and acceptability
10 L generally successful but some less appropriate choices
11 L generally successful but some less appropriate choices
12 L some lack of acceptability for TL
13 F/L generally successful but some less appropriate choices
14 F/L generally successful but some less appropriate choices
15 F generally successful, some loss of imagery and symbolic sign function
16 L not completely acceptable for TL but works overall
17 F generally successful but some less appropriate choices
18 L some lack of acceptability but adequate for purpose
19 L some lack of acceptability but adequate for TL function
20 L TL acceptability but not completely adequate for TL function
21 F generally successful and adequate for function
22 F generally successful and adequate for function
23 L generally successful and adequate for function but some lack of acceptability
24 L lack of acceptability, not completely adequate for function
25 L lack of acceptability, not completely adequate for function
F = Functional; L = Linguistic

It can be seen that adopting a linguistic approach as a result of a specic

translation brief from the text giver or due to lack of training in translation
competence tends to lead to the production of a functionally less than adequate
TT. However, in the texts studied, adoption of a linguistic approach did not lead
to inadequate TT production, simply to a less acceptable TT than might have
been produced. In our opinion, each TT might have been more eective, had
 THE TRANSLATION OF ADVERTISING TEXTS 233

there been less reliance on the creation of a clear relationship of similarity of

form between Source and Target texts and greater attention to the creation of
similarity of message function and intended impact. The factors reported as
constraints on text production also indicate variations in practice which could be
more, or less, conducive to eective text production.
This seems to conrm the views of functionalist translation scholars (Nord
1997; Holz Mnttri 1984), about the importance of situational factors in the
production of an acceptable and functionally successful. To ensure production of
an eective advertising message, translators need to have appropriate training,
be aware of comparative intercultural and inter-linguistic problems in translation
and be allowed to exercise their expertise and judgement.

Recommendations

It is hoped that the ndings of this study, however limited, will encourage
further similar analysis, so that it will eventually be possible to formulate a set
of guidelines for the industry, to serve as a framework for the translation of
advertising texts. The recommendations of this paper can be summarised as a set
of DOs and DONTs, based on the ndings of the study.

DO
* Use suitably trained and qualied translators.
* Include translators early on in the process of producing the SL advertising
message.
* Where possible, develop a formula for SL advertising texts which can be
adopted for TT production with minimal adaptation to TL reader needs.
* Keep records of previously translated texts.
* Devise and promote a framework of criteria for evaluation of the potential
success of a TT.
* Use suitably qualied editors to proof-read and evaluate all translations
within these criteria.
* Give clear information to the ST producer and to the translator about:
the nature of the product
the qualities to be associated with the product for the target culture (SL
or TL)
(it is assumed that marketing research will have identied the prole of the
target reader, whatever the language community).
* Allow and trust translators to exercise their professional expertise.
</TARGET "ada">

234 BEVERLY ADAB

DO NOT
* Underestimate the TTs potential impact (positive or negative) on both the
product to be promoted and the company itself within the target culture.
* Constrain the translator and prejudice the TTs chances of success by
insisting on a particular methodology or approach to the task of translation.
* Fail to appreciate the need for proper translator expertise, to be acquired
through appropriate training programmes which include not only practical
development of competence but also an understanding of how the theoretical
concepts of Translation Studies can contribute to improved performance.
<TARGET "fue" DOCINFO

AUTHOR "Adrin Fuentes Luque"

TITLE "The Translator as Mediator in Advertising Spanish Products in English-Speaking Markets"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 22

The Translator as Mediator in Advertising

Spanish Products in English-Speaking Markets

Adrin Fuentes Luque and Dorothy Kelly

Universidad de Granada

Introduction

There has recently been growing interest in the translation of advertising in

Translation Studies, and it is indeed a eld which highlights the futility of some
of the traditional debates in our discipline. Here is a eld in which there is no
doubt that the translation must full a function in the target culture, must
inuence the target reader in a specic fashion. As Resch (1998) puts it: Instead
of obsessively contemplating the source text target text relationship and
endlessly discussing delity or equivalence, translation studies should advocate
an understanding of translation as expert intercultural text production within the
theoretical framework of intertextuality. In this same vein, Sguinot (1994) has
placed emphasis on the intercultural competence (legislation, marketing, cultural
conventions in advertising, etc.) of the translator as an essential gure in the
commercialisation of imported and exported products. In this paper special
attention will be paid to one particular aspect of this process, that is the exploita-
tion of national or cultural identity in the advertising of exports. It is generally
accepted in studies on advertising that the identication of a product with a
specic national or cultural origin may inuence the success of marketing
campaigns (see for example Cannon 1998). In order to underline the importance
of the role of the translator as a mediator in these cases, we analyse the results
of a small-scale case study of advertisements for Spanish products in English-
speaking markets, linking our ndings to those of studies on the national image
of Spain. In so doing, we adopt a denition of the role of the translator which
clearly transcends the traditionally-accepted role of bilingual scribe. Despite the
(not so recent) advances in Translation Studies and within the translation
profession (see for example the BD Memorandum 1986), it is still very much
236 ADRIN FUENTES LUQUE AND DOROTHY KELLY

the case that many national markets, including Spains, often limit the translator
to purely linguistic matters in the translation of advertising, or even directly
farm entire responsibility for their marketing campaigns out to agencies in the
target countries (Beltrn 1999).

The translator as mediator in international advertising

Increasing globalization of the world economy, including the consolidation of the

European single market through the eective implementation of the single
currency in the European Union will undoubtedly imply greater demand for the
translation of advertising texts. In this paper, we fully subscribe Sguinots
recommendations (1994), paying particular attention to one intercultural aspect
which she does not mention explicitly, that of mutual perceptions. It is our belief
that, if the translator is to full her role eciently, it is not sucient to be
aware of cultural dierences, but also of how the source and target cultures view
each other. Clearly, this is a complex issue which intervenes in decision-making
in many dierent translation situations. In our case, a translator with sound
awareness of the mutual perceptions of source and target cultures may exploit
them in dierent ways in order to make her target text more eective, that is in
order to persuade the consumer to buy, avoiding blunders of the kind which
abound in advertising manuals around the world.

Mutual perceptions and stereotypes of national identity

The human brains need to use shortcuts in the process of apprehending the
many, varied and complex messages which it receives every day leads to its
developing a series of stereotypes within which it is able to interpret messages.
These stereotypes may be positive, negative or neutral, although there is a strong
tendency for them to be negative when they are perceptions of the other, as
opposed to self-perceptions. Sangrador comments de hecho el extranjero es
caricaturizado casi siempre, y muy pocos son los que perciben favorablemente a
los miembros del pas vecino: es algo, al parecer, connatural a las relaciones
humanas1 (1981: 99100). This is probably due to the need to generate a
positive self-perception, which is easier to develop by distancing oneself from all
that is undesirable in the other. As several authors have pointed out, these
stereotypes are tremendously powerful: In extreme cases a single word can
suce to trigger not only an entire stereotype but also (at least elements of) the
 THE TRANSLATOR AS MEDIATOR 237

macrodiscourse in the readers mind (ODonnell 1994: 359). They are also
extremely longlasting: Distintos investigadores han comprobado cmo las
imgenes de los distintos pueblos persisten durante aos, e incluso muestran una
gran similaridad de unos lugares a otros2 (Sangrador 1981: 27).
The use of this power, deliberate or otherwise, by the mass media has been
studied on numerous occasions. In this case, we are interested in its use in the
world of international marketing, where it has become a well-established tool in
the hands of advertisers. Let us now turn specically to the image of Spain in
English-speaking countries, particularly the United Kingdom, in order to see how
this image aects the commercialisation of Spanish products on these markets.

The image of Spain

A study, commissioned in 1996 by the then Spanish Ministry for Trade and
Tourism to establish how Spain was seen abroad as regards trade, tourism and
investment, oers the following conclusion on the image of Spain in its four
largest European neighbours:
una imagen poltica muy buena para un pas latino y mejorando, con niveles
crecientes y altos de conanza hacia el pas y hacia las personas. Una muy
buena imagen cultural y artstica. Un fuerte estereotipo, simplicado y rodeado
de un espacio de gran ignorancia hacia la realidad del pas. Perjuicios arrai-
gados de tendencia incierta. Una imagen econmica buena pero no excelente,
en la cola del pelotn de cabeza y en cierto modo inferior a la capacidad
econmica real del pas. Un made in regularmente valorado a consecuencia de
un muy escaso conocimiento de productos y marcas espaolas y de cierta
imagen de rusticidad. Y nalmente una excelente imagen como destino
turstico slidamente asentada.3
(Lamo de Espinosa 1996: 212)

Despite the apparent optimism with which these ndings are presented, the
authors of this study clearly recognise that the image of Spain abroad is not
really the one the country deserves, and in fact the details of their study
contain data which are much more worrying. Specically, the British perception
of Spain is more negative than the French, the Italian or even the German
(although in this case there are more similarities): in these perceptions, Spain is,
for example, a lazy, inecient, and corrupt country (the famous Spanish
practices), which is technologically underdeveloped. These ndings coincide
with those of Kelly (1997) in a large corpus study of the discursive construction
of Spain in the British press. In addition to this perception and construction of
238 ADRIN FUENTES LUQUE AND DOROTHY KELLY

Spain, there is an almost complete lack of knowledge among European consum-

ers regarding Spanish brands: SEAT is the only one which is recognised by more
than 20% of the interviewees in the Ministry of Trade study (Lamo de Espinosa
1996: 17). We refer here to perceived rather than actual nationality, given the
inter-, multi- and transnational nature of many of todays companies, as per-
ceived nationality has considerable inuence on sales (Cannon 1998: 122). Other
studies (Peris et al. 1993) indicate that the British consumer tends to associate
Spain with a very limited range of products and services, most of them agricul-
tural or related to tourism. In actual fact, in 1997, the sectors which exported
most from Spain were those of semimanufactured goods, capital goods, and
automobiles, not the food sector (Anuario El Pas 1998: 346).
Quite the opposite occurs in the case of products of other countries, which
have shown skill in using stereotypes to establish a positive national identity. To
name but a few, German products are associated with eciency and state-of-the-
art technological development, allowing the countrys emblematic engineering
sector to use these images to great eect in the marketing of cars, electrical
appliances, or more specialised capital goods for smaller audiences. French
products are associated with good taste, romance and sex. Italy makes excellent
use of its image of elegance in its agship design sector.
Cannon (1998) suggests several possible reasons for Spains non-use of a
positive national identity to sell its products, amongst them the interventionism
and protectionism of the Franco regime, the considerable weight of the national-
ised industries, the lack of experience of private enterprise, the traditional
tendency of Spanish companies only to export when they run out of national
market, the absence of specically international departments in companies, this
last point linked to the preponderance of small and medium-sized companies in
Spain. She suggests, nally, While one cannot ignore the publics propensity to
generalise on the basis of perceptions, much of the problem lies with the Spanish
business community itself and a decient internal policy which has failed to
promote a coordinated national image (Cannon 1998: 122).
It is clear from our case study (see below) that there is eectively a lack of
national image in the marketing campaigns of Spanish companies from the
private sector, although we stress as a factor precisely the publics propensity
to generalise, which we believe is inuenced strongly by the medias persis-
tence in constructing a negative image of Spain in the UK. It is interesting to
note that there is awareness of this issue amongst those currently responsible for
the promotion of Spain and Spanish products abroad, especially the Instituto de
Comercio Exterior or ICEX, and advertising specialists in general within Spain
(see for example, Valdelomar 1998).
 THE TRANSLATOR AS MEDIATOR 239

Case study of advertisements for Spanish products in English-speaking

markets

Corpus

Our corpus consists basically of two main groups of advertising texts: advertise-
ments used, rstly, by major private Spanish companies (some of which have a
degree of state participation), and secondly, by dierent Spanish state and
regional institutions, including the ocial tourist promotion campaigns. It is
taken from English-language publications from the UK (The Times, Economist,
Financial Times, Daily Express), Canada (Macleans), the USA (Time, News-
week), and from Iberias bilingual in-ight magazine, Ronda Iberia, over the
period 1989 to 1997. Our intention was to detect presence or absence of explicit
or implicit reference to Spain in the texts. Space limits mean that our ndings
can be described only very briey here.

The advertising of the private sector

In the case of the private sector, the corpus includes, amongst others, texts
produced to advertise some of the major Spanish companies: Telefnica, Iberia,
Banco Central Hispano, Banco Bilbao Vizcaya, El Corte Ingls, SEAT, etc. As
Cannon suggests, we also nd that the general tendency here is for the country
of origin, Spain, to be absent, or only very marginally present in the texts. As
examples, we will mention the banking and wine sectors. In the former, there is
absolutely no reference (verbal or iconic) to Spain in any of the texts in the
corpus. Similarly, for instance, in the case of Freixenet cavas, we nd absolutely
no mention of Spain, to the point that the company is identied by its British
address at Freixenet House, with the exception of one advertisement, which
was published in the Spanish edition of the Daily Express, for British tourists in
Spain, for whom the origin of the drink was perhaps considered of some
relevance. In the same vein, several UK supermarket catalogues consulted
contained references to the nationality of practically all the wines they sold, with
the exception of those from Spain (cavas included). This is possibly linked to a
tendency detected amongst Spanish wine producers to entrust their entire export
marketing campaigns to agencies in the target countries (Beltrn 1999).

Institutional advertising

When it comes to institutional advertising, however, we nd quite the opposite:

the deliberate use of some of the more positive of the stereotypes regarding
240 ADRIN FUENTES LUQUE AND DOROTHY KELLY

Spain. The most outstanding (and numerous) of the texts in this group of our
corpus belong to the dierent campaigns used by Turespaa to promote Spain
as a tourist destination. These are very professional texts, which make excellent
use of photographs and other images to attract the potential consumers attention.
During the period of the corpus, there are two separate sets of texts, the rst
centred on the slogan Passion for Life, and the second, Spain by. The rst
is particularly interesting in its skillful exploitation of single key words associat-
ed with Spain (soul, ol) which appear in the background of each of the ads.
The second invokes the stereotyped image, to a lesser degree, through the use of
photography (monuments, countryside, women in traditional attire). In both
cases, the campaigns, far from shying away from the impact of stereotypes, make
use of them (and the fact that they are so deep-rooted) to the benet of the
promotion of Spanish tourism. Since our initial study, Turespaa has begun a
new campaign under the slogan Bravo, which plays on a very successful
combination of traditional and modern images.
An interesting parallel may be drawn here with the institutional advertising
of the Basque Country and Catalonia, destined mainly to consolidate the separate
identity of these areas as sites for good investment opportunities. The texts use
the concepts of Europe and the Mediterranean (for the latter) as strong selling
points. Indirect, or even on occasion direct, reference to the negative stereotypes
of Spain is deliberately used here also, but in this case to dissociate these two
regions from the image of the rest of the State.

Conclusions: what the mediator can do

We would like to end by oering a simple example of dierent ways in which

the translators mediation might work. The Spanish airline Iberia uses an ad
which is based on the companys punctuality: a very attractive photograph in
pleasing pastel tones shows from above an Iberia aeroplane ying over clouds
tainted by sunlight, with the slogan in small letters alongside: Como un reloj.
There is little need to tell readers that lack of punctuality is one of the most
well-established clichs of Spanish-ness, as headlines of the kind Expo 92 will
be ready maana (Sunday Telegraph, 12th April 1992) insist constantly on
reminding UK readers. This stereotype is closely related to others such as
ineciency, which are considered to be particularly negative in English- or
German-speaking cultures.
Our example poses few problems to the translator from a linguistic point of
view: the simile is relatively easy to render in English, like clockwork being
 THE TRANSLATOR AS MEDIATOR 241

only one of several idiomatic expressions which could be used. But how would
the target readers react? Would this text really be able to persuade them that
Iberia is one of the most punctual commercial airlines, when the inability to
arrive on time is an essential part of the macrodiscourse of Spanish-ness in their
culture? As we suggested above, the rst role of the translator is to identify this
diculty as an intercultural issue and to advise the commissioner on the subject.
There will be a variety of possible solutions, ranging from engaging the stereo-
type in a fairly outright fashion So who said Spaniards are never on time?, or
the less aggressive When maana is a thing of the past (both of which use the
stereotype as a marketing tool, as Spain institutional advertising seems to do), to
a much more neutral The quickest way to Spain, As the crow ies (which
avoid the clich, as Spains private companies seem to do).
These are not the only solutions open to the translator, however. As Resch
(1998) points out, the advertising text must be taken as a whole in which the
verbal elements are but a small part. As verbal and visual elements form an
inseparable whole, the translators intervention could, or indeed should, also
aect the visuals. In this case, the translator may decide that there is not an
English-language slogan to complete the text suitably, and that the most appro-
priate action is to discard this particular idea for the English-speaking market.
Unfortunately, there are often immediate nancial considerations which lead the
commissioner to reject such solutions, despite the fact that they may make for
more ecient marketing and thus be more protable in the long run.
To conclude, then, in general the role of the translator in international
advertising is a complex one, which can in no way be limited to purely
linguistic issues. Increasing recognition of translators as intercultural experts,
and awareness within the profession and on training courses of this aspect of our
professional activity, will ensure better results, and doubtless avoid intercultural
blunders of the kind which abound as anecdotes in the pages of advertising
manuals. More specically, translators working for Spanish companies launching
export campaigns abroad might do well to consider the current limitations of the
advertising of Spanish products and take the bull by the horns, so to speak, in
the thorny issue of negative national stereotypes.

Notes

1. In fact, the foreigner is almost always caricatured, and only very rarely are those who come
from a neighbouring country perceived in a favourable light: this would seem to be an inherent
part of human relations. [Our translation]
</TARGET "fue">

242 ADRIN FUENTES LUQUE AND DOROTHY KELLY

2. Several researchers have found that the images of dierent peoples persist for years, and are
even very similar from one place to another. [Our translation]
3. a very good political image for a Latin country, with a tendency to improve, high levels of
condence in the country and its people, with a tendency to improve. A very good cultural and
artistic image. A strong, simplied stereotype, together with considerable ignorance of what the
country is really like. Deep-rooted prejudice with no clearly dened tendency for the future. A
good, but not excellent image of the economy at the back of the leading group, in some way
underestimating the real economic potential of the country. Low value given to made in Spain
due to very little knowledge of Spanish products and brands, together with a certain rustic
image. And nally, an excellent image as a well-established tourist destination. [Our translation]
<TARGET "par" DOCINFO

AUTHOR "Joan Parra"

TITLE "Translation as a Component of Software Localization Projects"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 23

Translation as a Component
of Software Localization Projects

Joan Parra
Universitat Autnoma de Barcelona

Introduction

Today, software localization is one of the biggest and fastest growing sectors
within the language industries, and in all probability it is also the fastest chang-
ing. Localization has played a seminal and central role in all the changes that
have occurred in recent years in professional translation practice. This entails
implications both for translation theory in its broadest sense and the teaching of
translation.
The term localization, which has now entered other languages as a loan
from English, refers to the process of producing specic versions of software
programs1 for use in markets other than those for which the original was
developed. In other words, localization is the process by which the original
characteristics of a program designed for one market are modied so that it can
be used or used more successfully in a dierent market. Among the many
denitions existing, of special interest for their conciseness are those of Seoane
(1997): to localize a product is to make it usable in a local environment and
Esselink (1998): Localization is the process of adapting and translating a
software application into another language in order to make it linguistically and
culturally appropriate for a particular local market.

Localization and Translation

In most cases, modifying a program for its sale in another market will mean
translating it: Purchasers perceive [the] product as having been developed and
produced in their own country not as something that has been developed
244 JOAN PARRA

abroad, converted and then imported (Way 1996).

Of all the operations which concur in the localization process, what we
could refer to as translation in the strict sense is the most strategically important
for the following reasons:
a. The translation phase is the one which involves the greatest volume of
human resources and co-ordination.
b. It represents the highest cost in economic terms.
c. It is one of the main causes of errors in localized product functioning.
d. It is one of the main causes of delays in nal product delivery date.
e. Its results are the most visible to the nal user.
If we understand translation as the operation undertaken by translators within
a production process, then we soon see that in software localization translation
tends to include a wider range of tasks than in, say, translation in the publishing
sector. In localization, the technological and professional background plus the
nature of the production process itself have a much more direct eect on the
actual translation work than in, say, literary or legal translation, which, to
mention just one aspect, require mastery of a more limited number of tools and
technologies.
When we speak of localization then, we mean a complex process in which
translation in the strict sense interacts with other operations. In this light,
localization exhibits a number of characteristics which mark it out as dierent
from other forms of translation. These features I have classied into three
groups: the nature of what is translated, the media and format in which it is dealt
with, and the peculiarities of the localization industry and the software market.

What makes translation in software localization dierent from other forms of

translation?

Analysis of a software program for the purpose of localization requires a series

of preliminary distinctions. The rst and most important dividing line is the one
which separates the program components in terms of their immediate visibility.
Moreno-Torres (1998) denes these two categories as:
the set of algorithms which make up the basis for the program, and
the interface which acts as intermediary between the user and the system.
Translation in the strict sense is concerned with the second of these categories.
Among the typical components of a software localization project are the
 TRANSLATION IN SOFTWARE LOCALIZATION PROJECTS 245

user interface, the on-screen help program, the various manuals, the software
license contract, the promotional materials and the packaging.
Among other items the user interface may include menus, dialog boxes,
screen messages and other textual elements, plus graphic and multimedia
components. It requires specic treatment because of its technical diculty, its
high visibility and a possible lack of context.
Localizing the on-screen help entails translating codied text and modifying
graphics, all of this as part of a process which can include preliminary break-up
and subsequent re-assembly of the materials.
Many software packages include documents intended to be used as exam-
ples or as templates for the creation of documents by the nal user. The
translation of these documents often entails great complexity, because it may
require cultural and functional adaptation.
Localization of manuals tends to be a complex task by virtue of the sheer
volume involved, the fact that text must be dealt with in desktop publishing
formats, and the fact that graphics must be handled (Uren et al. 1993: 8083). A
distinction must be drawn between manuals aimed at system administrators and
those designed for individual users.
Translation of the software license contract calls for legal translation skills.
Promotional materials come in all shapes and sizes and may include paper
documents and webpages. In addition to handling codied text and graphics, the
translator must have some degree of expertise or air for advertising; indeed,
along with the sample documents, this is the area of the localization process
which calls for most creativity on the part of the translator.
Software may include images, sounds, video sequences and other elements
often included under the general heading of multimedia. Each type of material
calls for specic treatment from the technical and translational point of view.
Software is by nature dynamic. Programs carry out tasks, unlike other
communicational products which are static in nature (Uren et al. 1993: 1.). The
functional complexity of some programs means that the nature of the task is
often unforeseeable since each of the translatable elements may operate in a
multiplicity of situations. Thus, the same text segment may appear in a large
number of dierent micro-contexts.
All the elements of visual interaction between the software and the user
(and that includes most of the components to be translated) must t on the
computer screen. In some cases the text expansion which arises from translation
(for example, from English into French) leads to space problems which must be
resolved by means of changes in the programming or the use of abbreviations
and other resources.
246 JOAN PARRA

Functional peculiarities include certain sensitive parts of the user interface

text. Hotkeys are among these, but there are others. The names of the menu
options usually contain a bold or highlighted letter by means of which the
command can be activated without using the mouse. This letter cannot be
repeated in the same micro-context, since it would hinder the softwares perfor-
mance. It is of course obvious that the highlighted letters often change from one
language to another.
The proportion of text which is repeated, that is, the number of re-occur-
rences of a single expression, tends to be high and reaches especially high
proportions in the on-screen help modules and the manuals. Therefore, translation
costs can be cut and turn-around time can be shortened through the use of
assisted translation technology especially translation memories.
A signicant number of the texts that constitute or accompany the original
software products suer from defects, such as being badly written, incomplete,
and/or terminologically and stylistically inconsistent, among others. These aws
can usually be attributed to the lack of eective technical writing practice, to
rectications and expansions arising during the development phase, decient
project organization or other factors. The problem of defective originals is also
common to other types of translation; however, in localization these problems are
more marked, due to the dispersion of the materials, the short economic life-
cycle of the products and the fact that the originals are in most cases written by
more than one person.
Every localization project must include a testing phase (see Kaner et al.
1993: 169202). In this, localization is no dierent from any other type of
translation. Technical checks and testing must be carried out to determine
whether the localization process has in any way aected the programs function-
ing. Furthermore, the interaction between user interface, on-screen help and other
elements, demands terminological consistency and cross-checks, such as for
example, between the user interface and the manual, the user interface and on-
screen help, and so on.
The material to be translated is often presented to translators in a codied
format. The natural language text, that is, the part to be translated in the strict
sense, is combined with elements which are concerned exclusively with the
programs functioning, denition of text format and links to other objects in the
text.
To translate software, it is necessary to take it apart or to reduce it to
elements which can be dealt with by the translator. In the majority of cases a
very high number of les of dierent types are involved. The sequence in which
they are to be translated must be established and so too must the tools that will
 TRANSLATION IN SOFTWARE LOCALIZATION PROJECTS 247

be necessary. Similarly, terminology management must be provided for and

consistency problems anticipated.
These are some of the tasks that must be performed by the project manager,
who is the specialist increasingly employed by localization providers and who in
some cases may have a translation background. The individual translator only
sees a relatively small part of the whole, while the project manager must have an
overview of the entire process.
The fact that the product must be disassembled prior to translation means
that very often the translator is forced to work without sucient context in the
broadest sense of the word (Madell et al. 1994: 6.). In certain cases the
translator may be unaware of the operative function of certain expressions, the
elements, textual and otherwise, which will accompany the text when the
localized software is actually in use, and the sequence of operations that will
precede the appearance of an expression when the program is being run.
In short, the translator is obliged to work on the basis of disparate segments
which are not ordered in accordance with criteria of textual coherence but rather of
production eciency.
In software localization, the author of the original product, the initiator of the
translation process and the publisher of the translation generally tend to be one and
the same entity. Amongst other things, this means that the client normally has a
clearly focussed idea of what he or she wants. This is to the benet of the initiator,
and sometimes of the localization provider, the translator and the nal user of the
localized product. Project management, whether whole or partial, tends to be the
responsibility of the localization provider.
The software and localization industries have developed a considerable number
of specic tools to help deal with the complexity of the task in hand (Kano 1995:
121123). In addition to the already mentioned assisted translation tools and machine
translation technologies, there are other tools designed specically for use in the
localization industry (see Rinsche 1997; Esselink 1998).
A number of localization providers have developed their own localization
tools and in some cases they have put them on the market. The same is true of
the localization initiators: a number of them have set up localization departments,
which also develop assisted localization technology, which can too end up being
marketed commercially. It frequently happens that initiators expect localization
providers to use certain tools, for technical, production or commercial reasons.
The provider then often nds himself faced with a client who is not only
interested in the result of the process but who also wishes to control a number of
its variables.
The internationalization techniques applied in the development phase of the
248 JOAN PARRA

original product aim to facilitate translation (Uren et al., 1993: 1191). The
original product is seen as a translatable object from the earliest planning stages.
As a result, the translatable elements of the software are isolated and made
separately available to the translator. The purpose of this is to cut costs, speed up
turn-around and marketing time and improve the quality of the nal result.
It is very common, indeed, inevitable, that the original software will
undergo further changes once the localization process is underway, for in the
world of software there is no such thing as a nal version and programs are
always liable to further development and modication. Hence, the user interface
may change half-way through the translation process and this will entail further
changes in the on-screen help and user manuals, etc. For this reason, translation
work begins on software, on-screen help programs and documentation well
before they are nished. Indeed, there is enormous pressure to have simultaneous
launches of original and localized software programs.
The dynamics of the software market tend to press for as early a product
launch as possible, and as a result most localization projects, many of which
involve enormous volumes of words, must be completed in a relatively short
space of time. This demands in-depth planning, organization of group work, the
use of assisted translation technologies and other time-saving strategies.
Software manufacturers, especially those with high outputs, have made and
continue to make eorts to standardise terminology, for a range of dierent
reasons. Therefore localization is glossary-based and this involves a number of
advantages, such as less time spent on terminological research work. However,
it also entails disadvantages (see Gardner 1992).
In software localization, the source language is in most cases American
English, and this is increasingly the case, at least in global terms. On the other
hand however, the number of target languages involved never stops growing and
internationalization technologies such as double byte character sets greatly
facilitate localization into Far Eastern, Arab and South Asian languages, etc.
(Kano 1995: 57113). In this context however, there are few grounds for opti-
mism as regards the situation of minoritised languages in the eld of software
localization, despite the existence of some promotional initiatives on a local
scale. Some characteristic features of localization into minoritised languages have
been described (see Griths 1995).
</TARGET "par">

TRANSLATION IN SOFTWARE LOCALIZATION PROJECTS 249

Conclusion

From the above, it can be seen that at very least, the specic conditions which
accompany translation in software localization and the situations and settings in
which it takes place set it apart from other types of translation. The consequences
of this for translation theory are not at all easy to predict, but there is one certain
conclusion: translation teaching must pay attention to the phenomenon, and
translator training institutions must either develop or acquire the resources that
are necessary to teach the specic skills required. Both localizable software and
localization technologies and tools are evolving at great speed. The changes
taking place are so diverse and complex, and their consequences so far-reaching,
that it will be impossible for the academic world to broach the area alone. For
this reason, any approach that is to achieve valid results will need to obtain
funding and support from the professional sector.

Translated by Carl Mac Gabhann

Note

1. For the purposes of this paper, software is taken to be the standard and widely-used computer
programs run on PC/Windows platforms.
<TARGET "gar" DOCINFO

AUTHOR "Carlos Garrido"

TITLE "Traduccin de los nombres vernculos ingleses de animales en los textos de divulgacin cientifca"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 24

Traduccin de los nombres vernculos ingleses de

animales en los textos de divulgacin cientfica

Carlos Garrido
Universidade de Vigo

Abstract

Problems and strategies in the translating English common names for animals in
scientic texts for non-specialists. Common, non-scientic, names for animals
make up an important part of the terminology used in scientic texts for non-
specialists in English. This is due to both the genres thematic orientation
toward nature as well as to the wide variety of common names available to the
English language which is, in turn, partly a consequence of the long-standing
English tradition of amateur and specialist study of fauna which has led to the
coining of a great number of articial, non-scientic animal names. After
briey explaining the basics of taxonomic naming and vernacular zoonymy,
this paper goes on to discuss the problems which arise when translating such
animal names into Spanish. Bearing in mind a series of translation errors
detected in the Spanish version of several articles taken from newspapers,
Scientic American and National Geographic Magazine, a number of transla-
tion strategies are proposed in order to overcome three main following
problems: (1) the fact that bilingual English-Spanish dictionaries fail to include
many common names for animals; (2) the frequent absence of scientic names
to accompany their popular equivalents in English texts for non-specialists; and
(3) the frequent lack of a popular Spanish equivalent for the corresponding
common name in English.

Introduccin

Las designaciones vernculas, no cientcas, de especies y de grupos de animales

son un elemento terminolgico muy frecuente de la divulgacin cientca en
252 CARLOS GARRIDO

lengua inglesa, tanto por la inherente conexin temtica del gnero con la
naturaleza, como por la circunstancia de que, segn ms adelante se expondr,
la rica tradicin anglosajona de observacin de la fauna, por parte de acionados
y de especialistas, ha determinado una masiva y sistemtica instauracin de
nombres vernculos de organismos, incluso para especies y grupos exticos,
inconspicuos o desconocidos para el gran pblico.
El hecho de que la mayora de los nombres vernculos ingleses de animales
no sean catalogados en los diccionarios bilinges ingls > espaol ni, a veces, en
los monolinges, que sus equivalentes vernculos no existan en espaol, y que
frecuentemente no aparezcan acompaados en el texto original de sus correspon-
dientes designaciones cientcas o semicientcas (vid. infra), los torna en un
verdadero escollo en el proceso de traduccin de la divulgacin cientca.1 A
modo de botn de muestra de este problema, que puede comprometer seriamente
la validez de una traduccin, nos remitimos (cf. Maillot 1997: 195) al siguiente
juicio de Fernando Parra contenido en su recensin de El ltimo panda, versin
espaola de una monografa faunstica del conocido zologo George B. Schaller:
Un s[o]lo reproche a la edicin espaola. Correctamente traducido[, el texto]
debera, sin embargo, haber sido revisado por un bilogo; los nombres de las
especies son a menudo incorrectos en castellano y las notas a pie de pgina del
traductor a menudo suman incorrecciones a lo que pretendan aclarar. (Babelia,
El Pas, 8 de noviembre de 1997: 16)

Con el n de llamar la atencin del traductor espaol no especializado sobre las

dicultades planteadas por los nombres vernculos ingleses de animales, el
presente trabajo, tras exponer sucintamente los fundamentos de la designacin
taxonmica y las divergencias de la zoonimia, analiza los problemas que suscita
su traduccin y propone estrategias para solucionarlos. En cada caso, los
problemas se ilustran mediante casos conictivos y errores de traduccin
detectados en varios artculos de divulgacin cientca aparecidos en las revistas
Investigacin y Ciencia y National Geographic Magazine Espaa (versiones en
castellano de, respectivamente, Scientic American y National Geographic
Magazine),2 as como en las pginas de ciencia de El Pas, diario de informacin
general.

Fundamentos de la designacin taxonmica

Los seres vivos pueden clasicarse siguiendo criterios diversos, lo que constituye
el objeto de estudio de la disciplina biolgica de la sistemtica o taxonoma.
sta, de acuerdo con un criterio y una nalidad a la vez tericos (la clasicacin
 TRADUCCIN DE LOS NOMBRES VERNCULOS 253

debe reejar la lognesis o historia evolutiva de los seres vivos, es decir, el

parentesco) y prcticos (la clasicacin facilita la comprensin de la biodiver-
sidad), delimita grupos de organismos o txones, a los que tradicionalmente se
asigna una categora (de especie a lo). Las designaciones de taxon pueden
adoptar las siguientes formas y niveles de uso (cf. Tabla 1; cf. Garrido 1998):
1. Forma cientca e internacional (trminos en latn, griego clsico trans-
literado al alfabeto latino, o de morfologa adaptada a esos dos idiomas). En
general, esta forma es, para cada unidad taxonmica, nica (a pesar de la
existencia ocasional de sinnimos taxonmicos). Ej.: Plathelminthes.
2. Forma semicientca, cuasi internacional (semejante a la cientca,
aunque algo adaptada a las lenguas vernculas. Esta adaptacin atae sobre
todo al empleo de sujos patrimoniales, acentos y signos diacrticos y
transcripciones de grupos de letras. Cf. Alvarado 1983). A cada unidad
taxonmica, por lo general, corresponde una nica designacin semicien-
tca. Ej.: ing. Platyhelminths, esp. Platelmintos.
3. Forma verncula, muy variable segn las diversas lenguas. Es frecuente la
pluralidad de denominaciones para cada unidad. Ej.: ing. atworms, esp.
gusanos planos.

Tabla 1: Nombres semicientcos y vernculos en varias lenguas correspondientes a la

clase Scyphozoa (nombre cientco)
Nombre semicientco Nombre(s) vernculo(s)
Alemn Skyphozoen Quallen, Schirmquallen
Espaol Escifozoos medusas, aguamalas
Ingls Scyphozoans jellysh
Portugus Cifozorios alforrecas

Riqueza zoonmica y traduccin

Existen en la actualidad alrededor de un milln de especies de animales conocidas,

a las que los zologos han atribuido nombres cientcos latinos o latinizados que
siguen las prescripciones registradas en el Cdigo Internacional de Nomenclatura
Zoolgica. De este ingente nmero de especies, que, gracias a la labor de los
taxnomos, crece anualmente a un ritmo considerable, tan slo una reducidsima
fraccin es conocida por la generalidad de los componentes de cualquier comunidad
lingstica y, en consecuencia, ha merecido en la correspondiente lengua comn
designaciones que puedan calicarse de nombres populares o vulgares.
254 CARLOS GARRIDO

Si bien una notable inferioridad numrica caracteriza sin excepcin en todas

las lenguas naturales el caudal de las designaciones vernculas de animales en
relacin con el de nombres cientcos zoolgicos (o de especies conocidas por
la ciencia), por otro lado resulta patente que entre las diversas culturas y lenguas
se registran diferencias en la riqueza zoonmica. Esta disparidad entre las lenguas
puede deberse, en un primer momento, a una serie de causas naturales , como
son la disimilitud de los entornos naturales ( ecologa ) en que se insieren las
diferentes culturas, las diferentes intensidades de exploracin cultural de dicho
entorno natural, la diversa amplitud del dominio geogrco de las lenguas, o los
distintos grados de receptividad de prstamos lxicos procedentes de otros
idiomas que ellas muestran. En un segundo momento, las diferencias en la
riqueza zoonmica pueden deberse a que en algunas lenguas propias de comuni-
dades culturales de cierto desarrollo cientco se ha procedido por parte de los
especialistas a una masiva y sistemtica instauracin de nombres vernculos para
especies animales exticas, prehistricas, raras, inconspicuas o de descubrimiento
reciente, de manera que la mayor parte del lxico vernculo de especies en esas
lenguas es, paradjicamente, de exclusivo uso cientco.
El ingls es una de las lenguas que ms abunda en este tipo de nombres
vernculos de animales de instauracin articial , debido sin duda a la rica
tradicin anglosajona de observacin de la naturaleza, tanto por parte de aciona-
dos como de especialistas, y a la habitual renuencia de sus hablantes a utilizar
los nombres cientcos, que se sienten en general como excesivamente tcnicos,
largos y, por tanto, impronunciables. Adems, esta tendencida se ve agudizada en
los textos de divulgacin cientca, que estn destinados a un pblico lego en
zoologa, y en ellos es frecuente que el nombre vernculo de la especie no vaya
acompaado de la correspondiente designacin cientca.
Debido a que esta extrema riqueza zoonmica del ingls no tiene parangn en
castellano,3 el traductor espaol de textos ingleses de divulgacin cientca o de
temtica naturalista se enfrenta a menudo al siguiente inconveniente: los diccionarios
bilinges no contienen el nombre vernculo ingls cuyo equivalente espaol se desea
determinar. En estas circunstancias se impone una estrategia de traduccin basada en
el conocimiento del correspondiente nombre cientco (o semicientco), de uso
internacional, cuya casustica y desarrollo exponemos en el siguiente apartado.

Tratamiento translativo de los nombres vernculos ingleses de animales

En la traduccin al espaol de los nombres vernculos ingleses de animales

podemos considerar dos etapas: primera, la propia determinacin del equivalente,
 TRADUCCIN DE LOS NOMBRES VERNCULOS 255

si existe, en la lengua de llegada y, segunda, la formulacin de la traduccin en

el texto nal de la unidad de procesamiento en cuestin. En las lneas que
siguen presentamos un anlisis detallado de la estrategia de traduccin que
juzgamos ms apropiada para tratar los nombres vernculos ingleses que
designan especies o grupos de animales, ilustrando con ejemplos de traducciones
publicadas algunos puntos de inters.

Determinacin del equivalente

Para determinar el equivalente espaol de un nombre vernculo ingls que resulta

desconocido o dudoso al traductor, ste deber primero consultar un buen
diccionario bilinge.4 Si esta consulta resulta fructfera (caso, por desgracia,
bastante infrecuente), el traductor har bien en cerciorarse de la bondad de la
equivalencia hallada consultando un diccionario o enciclopedia monolinges en
cada uno de los idiomas de la traduccin para contrastar las respectivas deni-
ciones. En esta confrontacin de deniciones es importante reparar en la coin-
cidencia de nombres cientcos, ilustraciones y mbitos ecolgicos y geogrcos
atribuidos a la especie o grupo de animales en cuestin, para garantizar la
exactitud y propiedad de la traduccin.
As, si en el original se habla de unos pjaros denominados warblers, el
traductor tendr que ser consciente de que la equivalencia espaola puede ser ora
currucas/mosquiteros (familia Slvidos), ora reinitas (familia Parlidos), depen-
diendo del mbito geogrco (Europa o Amrica) en que se siten estas aves,
como puede observarse consultando la citada voz en un diccionario de lengua
comn inglesa:
warbler. [] 2. Any of several small, chiey Old World songbirds of the
family Sylviidae. Cf. blackcap (def. 1), reed warbler. 3. Also called wood
warbler, any of numerous small, insectivorous, New World birds of the family
Parulidae, many species of which are brightly colored. Cf. yellow warbler
(Dendroica petechia). (Websters Encyclopedic Unabridged Dictionary of the
English Language, 1994)

Esta circunstancia no ha sido tenida en cuenta por el traductor del artculo en que
se incluye el siguiente trecho, pues, si bien ha traducido correctamente warblers
por reinitas, dado que el contexto geogrco es norteamericano, en su ampli-
cacin explicativa del nombre vernculo (vid. infra) habla de la familia Sylviidae
(la de las currucas europeas) en relacin con las reinitas, en vez de la familia
Parulidae (parlidos) que correspondera:
256 CARLOS GARRIDO

(1) NGA, 192 (4), p. 83: The brown-headed cowbird is an egg dumper, deposit-
ing its eggs in the nests of warblers [] > NGE, 1 (1), p. 83: El boyero de
cabeza parda es un pjaro parsito, que deposita sus huevos en los nidos de
las reinitas (slvidos parecidos a las currucas), []
El siguiente ejemplo (2), extrado del mismo artculo, ilustra la necesidad de que
el traductor, mediante la documentacin, profundice en la adscripcin taxon-
mica, en la identidad, del animal citado en el texto original y en el nal, de
manera que se pueda asegurar de la exactitud de la equivalencia propuesta:
(2) NGA, 192 (4), p. 90: To complete their life cycle, trematode larvae lodged
in a mud snail must nd their way into a beach ea and, from there, into
the digestive system of a bird. [] When a shorebird eats the ea [] >
NGE, 1 (1), p. 90: Para completar su ciclo biolgico, las larvas de trema-
todo alojadas en un caracol del fango tienen que labrarse camino en una
nigua y, desde all, hacia el aparato digestivo de un ave. [] Cuando un
ave limcola traga la nigua []
Aqu el traductor se ha dejado engaar por el apelativo ea (>pulga) que gura
en el texto original, pues ha convertido beach ea (>pulga de mar), un crustceo
anfpodo no parasitario que habita el litoral de las regiones templadas-fras (por
ejemplo, la especie Talitrus saltator), en un insecto sifonptero tropical y
parasitario, la nigua (Tunga penetrans). La confusin tal vez se haya debido a
que sand ea, nombre prximo al original beach ea, designa en ingls, adems
de la pulga de mar, la nigua (tambin denominada chigoe, chigger o jigger ea).
En caso de que el nombre vernculo ingls no sea registrado por los
diccionarios bilinges, habr de determinarse, para proseguir en la investigacin
del equivalente, el correspondiente nombre cientco (o semicientco). ste
puede hallarse a veces acompaando al nombre vernculo en el texto original, o
en un apartado previo del mismo, pero, si no es as, podr encontrarse en una
enciclopedia o en un buen diccionario monolinge de ingls en la denicin
encabezada por el pertinente nombre vernculo.5
Una vez determinada la designacin cientca (o semicientca), de mbito
internacional, que corresponde al nombre vernculo original, el problema de
traduccin puede resolverse total o parcialmente recurriendo a bibliografa
adecuada compuesta en castellano: si existe un equivalente vernculo espaol,
ste puede encontrarse consultando el nombre cientco pertinente (o el vern-
culo del grupo que lo incluye) en una enciclopedia o en el ndice alfabtico
remisivo de un manual universitario de zoologa, de una gua de la naturaleza o
de una monografa/enciclopedia faunstica (ej.: Rodrguez de la Fuente 1970;
 TRADUCCIN DE LOS NOMBRES VERNCULOS 257

Barnes 1984). As, por ejemplo, para determinar el equivalente vernculo espaol
de hartebeest puede consultarse un diccionario o enciclopedia redactados en
ingls, donde se descubrir que se trata de un antlope africano perteneciente a
la especie Alcelaphus buselaphus; en el artculo encabezado por la voz antlope
de una enciclopedia espaola podr verse a continuacin que una de las especies
de antlope es el alcelafo o bubal, de nombre cientco Alcelaphus buselaphus.
Si no existe equivalente vernculo espaol para el nombre ingls original y,
por tanto, la estrategia que se acaba de describir resulta infructuosa, deber
utilizarse en la traduccin el correspondiente nombre cientco (o semicientco).
As, para verter al espaol en un texto sobre pesca la denominacin inglesa
roughy, correspondiente a un pez tropical, deber recurrirse a su nombre
cientco (Trachichthys australis), pues, como atestigua el diccionario multi-
linge de especies marinas de Vera Kirchner (1992), para esta especie no existen
denominaciones vernculas en espaol (ni en francs, ni en alemn).
Esta sustitucin de nombre vernculo por nombre cientco debe realizarse
con cuidado porque, incluso en el caso de que la correspondencia sea correcta y
no conlleve un cambio de la designacin original, en ciertas ocasiones puede
alterar innecesariamente el registro del texto y causar extraeza. Considrese, por
ejemplo, la siguiente traduccin:
(3) SciAm, noviembre 1996, p. 110: The aardvark > IyC, enero 1997, p. 80: El
mamfero nocturno de la especie Orycteropus afer.
Aqu lo apropiado hubiese sido, para traducir aardvark, utilizar las equivalencias
verncula o semicientca cerdo hormiguero u oricteropo, con lo que se hubiese
conservado el efecto pragmtico del original. Tngase en cuenta, por otra
parte, que la elevacin de registro respecto al original que acarrea la sustitu-
cin de un nombre vernculo por otro (semi)cientco resulta en muchos casos
inevitable y no hace sino conrmar que la divulgacin cientca en espaol, por
el carcter exclusivamente grecolatino de gran parte de su vocabulario especial-
izado, se halla ms alejada de la lengua comn que la que se compone en ingls
(y, sobre todo, en alemn).6
Lo que nunca se debe hacer en caso de inexistencia de equivalente vern-
culo espaol para el nombre original es acuar un falso nombre espaol mediante
la traduccin palabra por palabra del nombre ingls, pues el uso del nombre
vernculo inventado por el traductor, aunque en algunos casos pueda resultar
ms descriptivo que el nombre (semi)cientco, en general obscurece el texto
traducido al impedir tanto la identicacin inmediata del animal por parte del
especialista como una consulta provechosa de bibliografa adicional por parte del
lego interesado. En el ejemplo siguiente se muestra este tipo de instauracin
258 CARLOS GARRIDO

espuria de falsos nombres vernculos por parte del traductor:

(4) NGA, 192 (4), p. 90: the thorny-headed worm robs the roach of its normal
escape response [] > NGE, 1 (1), p. 90: El gusano de cabeza espinosa le
hurta a la cucaracha su respuesta de huida []
En el ejemplo (4), el sintagma vernculo gusano de cabeza espinosa, aunque
descriptivo, no reviste carcter terminolgico, y es que para traducir thorny-
headed worm (=spiny-headed worm) hay que recurrir al nombre semicientco
espaol (de etimologa griega) acantocfalo (cf., p. ej., Barnes 1984: 327; Brusca
y Brusca 1990: 366).
La tentacin de traducir palabra por palabra las designaciones vernculas
inglesas puede resultar tan perniciosa que, por desidia del traductor, se llegue
incluso a sustituir verdaderos nombres vernculos espaoles por otros falsos y
carentes de signicado para el lector del texto nal, como ilustra el ejemplo (5):
(5) El Pas, 14 de diciembre de 1995, p. 34 (ttulo del artculo traducido del
ingls a partir del original proporcionado por Nature News Service): Un
parsito de complicada vida sexual inaugura una nueva categora de seres
vivos. La criatura ahora descubierta habita en la boca de la langosta nrdica
[<Norway lobster]
El ejemplo (5) se corresponde con un artculo aparecido hace algunos aos en las
pginas de sociedad de un diario de informacin general y representa la tradu-
ccin de un original periodstico ingls en que se comunica el descubrimiento de
un nuevo grupo de organismos. Este hallazgo haba sido inicialmente anunciado
por sus autores en un artculo de investigacin publicado en la revista inglesa
Nature (Funch y Kristensen 1995), donde se puede leer que el animal descrito
habita en the mouthparts of the Norway lobster Nephrops (p. 711). Ahora
bien, a pesar de la correspondencia literal de la denominacin, langosta nrdica
no puede darse como equivalente espaol de Norway lobster, pues esta especie
de crustceo (Nephrops norvegicus) es conocida entre nosotros como cigala (cf.
Campbell 1983: 214).

Formulacin de la traduccin

Una vez determinada la equivalencia de la unidad de procesamiento consti-

tuida por el nombre vernculo de animal en ingls, interesa ahora comentar
algunos aspectos relativos a la correspondiente formulacin de la traduccin en
el texto nal. Si en el original el nombre vernculo aparece acompaado de su
nombre cientco (en general, colocado entre parntesis), en la traduccin, si
 TRADUCCIN DE LOS NOMBRES VERNCULOS 259

existe en espaol el equivalente vernculo, tambin deben constar (en la misma

disposicin) los dos. Si, por el contrario, no existe equivalente vernculo en
espaol y se ha de utilizar en la traduccin un equivalente (semi)cientco, en
muchos casos es conveniente acompaar ste de una amplicacin explicativa,
que puede consistir en el nombre vernculo del grupo que incluye al animal
citado (hipernimo) o en los caracteres descriptivos que estn presentes en
muchos nombres vernculos ingleses (ej.: sea-potato > el erizo de mar Echino-
cardium cordatum, que tiene forma de patata, []). As:
(6) SciAm, noviembre de 1996, p. 63: One of us (Beck) has found molecules
resembling IL-1 and IL-6 in the tobacco hornworm. > IyC, enero de 1997,
p. 8: Uno de los autores (Beck) ha encontrado molculas parecidas a la IL-1
y a la IL-6 en el lepidptero Manduca sexta.
En el ejemplo (6) el traductor ha procedido correctamente sustituyendo en el
texto nal el nombre vernculo ingls (tobacco hornworm), sin equivalencia
verncula en espaol, por su nombre cientco (Manduca sexta), acompaado, a
modo de amplicacin explicativa, del pertinente grupo zoolgico en que se
incluye (lepidptero).

Conclusiones

Como estrategia general para traducir al espaol los nombres vernculos ingleses
de animales que con gran frecuencia guran en los textos de divulgacin
cientca, el presente trabajo propone: (1) no conarse exclusivamente a los
diccionarios bilinges y, para cerciorarse de la exactitud de las equivalencias,
realizar una confrontacin de deniciones en los dos idiomas (diccionarios y
enciclopedias monolinges) que incluya nombres cientcos, ilustraciones y
mbitos ecolgicos y geogrcos de los animales en cuestin; (2) en caso de que
el nombre vernculo ingls no sea registrado en los diccionarios bilinges,
recurrir al respectivo nombre cientco o semicientco, o al vernculo del grupo
supraordinado, para acceder, mediante la consulta de enciclopedias, manuales o
monografas, al equivalente vernculo espaol; (3) ante la frecuente inexistencia
en espaol de nombres vernculos para traducir los nombres vernculos ingleses,
utilizar en el texto nal los correspondientes nombres cientcos o semicien-
tcos, teniendo cuidado de no alterar innecesariamente el registro y de no
acuar falsos nombres vernculos espaoles por traduccin literal de los ingleses;
(4) al formular la traduccin de un nombre vernculo ingls que carece de
</TARGET "gar">

260 CARLOS GARRIDO

equivalente vernculo en espaol, conviene hacer acompaar el respectivo

nombre (semi)cientco de una amplicacin de carcter descriptivo.

Notas

1. Entendemos por divulgacin cientca aquella transmisin de informacin relativa a las ciencias
naturales y a la tcnica que tiene por emisor un especialista o mediador especializado (perio-
dista cientco) y por receptor un lego interesado. Debido a la creciente curiosidad del hombre
por el mundo que le rodea y a la transcendencia que la ciencia y la tecnologa han adquirido en
nuestras sociedades, hoy en da los conocimientos cientco-tcnicos forman parte importante
del bagaje cultural del ciudadano medio, de manera que nunca como ahora la divulgacin
cientca haba conocido una poca de tanto cultivo (artculos de prensa, artculos de revistas,
libros, documentales televisivos, vdeos, etc.).
2. Ttulos aqu abreviados como, respectivamente, IyC, NGE, SciAm y NGA.
3. No es tanto que el castellano carezca de nombres vernculos para animales exticos (la
magnitud de su actual extensin geogrca, prxima a la del ingls, lo ha puesto en contacto
con muy diversas ecologas), sino que el dcit frente al ingls (compartido por la mayora
de las otras lenguas) se registra ms bien en los mbitos de la instauracin erudita de, por una
parte, denominaciones vernculas para grupos de animales raros o poco llamativos (aunque no
necesariamente exticos) y, por otra, de nombres de categora especca en el seno de grupos
ms o menos conocidos por el gran pblico.
4. Los diccionarios bilinges de la lengua general no suelen contener sino los nombres de los
animales ms comunes y conocidos. Existen tambin algunos diccionarios multilinges, que
incluyen el ingls y el espaol, de nombres de animales (p. ej., para los peces: Vera Kirchner
1992; AA.VV. 1995), pero su nmero y su caudal lxico, por diversas causas, resulta escaso.
5. Los diccionarios ingleses de lengua comn suelen ofrecer, en contraste con, por ejemplo, la
mayora de sus homlogos espaoles (DRAE), portugueses (Porto Editora) y alemanes (Duden,
Wahrig), las designaciones cientcas de las especies y grupos de especies que corresponden a
los nombres vernculos lematizados. Otra fuente muy til de nombres cientcos son las guas
faunsticas especializadas en diversos grupos de animales (p. ej.: Campbell 1983; Peterson,
Mountfort y Hollom 1989).
6. Consltese, por ejemplo, la Nota a la edicin espaola que J. Ros i Aragons antepuso (p.
3) a su traduccin de una gua de la naturaleza inglesa (Campbell 1983), o, para el caso del
alemn, Rlker (1972: 52, 53).
<TARGET "ter" DOCINFO

AUTHOR "Maribel Tercedor-Snchez"

TITLE "A Pragmatic Approach to the Description of Phraseology in Biomedical Texts"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

C 25

A Pragmatic Approach to the Description

of Phraseology in Biomedical Texts*

Maribel Tercedor-Snchez
Universidad de Granada

Introduction

Specialized information in context is now the focus of an increasing number of

terminological studies and more attention is being paid to the environment of
terms in the creation of lexicographic and terminographic tools. In the encoding
and decoding processes, phraseological information is of paramount importance
for the translator who is not an expert in the specialized domain.
Scientic terminology has traditionally focussed on the study of information
for specialist-to-specialist communication. The vital need to access specialized
knowledge as well as the necessity that such knowledge be in a form that can be
easily understood, demand a switch in the work of the terminologist towards a
more pragmatic approach in order to give evidence of terminological and
phraseological variation. In a parallel way, it also demands a multidimensional
and interdisciplinary approach aimed at the creation of tools that allow access to
the relations between concepts. Here we are going to focus primarily on the
pragmatic aspect of terminology.
This paper underlines the usefulness of phraseological information for
conceptual and discourse analysis. One of the things that should be taken into
account in the creation of terminographic tools is the existence of context-
dependent phraseological units. The automatic processing of contextual informa-
tion in texts of specialized content requires a descriptive study of phraseological

* We are grateful to Dr. Abada Molina of the University of Granada and Dr. Cabrera Ibolen of
the Palliative Care Unit (Hospital Clnico Universitario, Granada) for their kind help in choosing
sources and supplying expert knowledge.
262 MARIBEL TERCEDOR-SNCHEZ

units in dierent contexts. This article gives an example of such a phraseological

analysis as applied to a corpus of English and Spanish biomedical texts in the
domain of Oncology.

Description and objectives

In the domain of Oncology the specialist often tends to use deliberately abstruse
language; this means that the patient is faced with many obstacles as a result of
the specialists concern to inform to the minimum in order to avoid iatrogenic
illnesses (Grinev et al. 1991: 183).
In contrast to the often secretive and esoteric nature of biomedical language,
the scientic writer and the translator have to adapt terminology to dierent
receivers in order to facilitate understanding.
The goal of terminology insofar as its function in society goes is the
ecient transmission of specialized information. In this regard, it is concerned
with the combinations of terms and words from the general language that vary
according to dierent parameters such as type of text, communicative aim of the
discourse, text sender and receiver, register and situation.
In the domain of Oncology there is important terminological and phraseo-
logical variation as a result of the sheer quantity of information available at all
levels in response to the desire of people to access information about cancer.
This situation demands a greater attention to the social aspect of terminological
work. In order to assist the translator, terminological description should be done
within the most frequent types of contexts in which the translator carries out
his/her work. For the accomplishment of this task, the description of phraseology
is essential given the possibilities it oers in translation-oriented discourse
analysis. Among the pragmatic components, special attention should be paid to
the notion of register as a parameter that allows the text to be coherent with the
situation (Halliday and Hasan 1976: 23) and which thus is essential for the
ecient transfer of information to specic users. A descriptive study of
phraseological units following such pragmatic approach is a necessary step prior
to the elaboration of tools for assisting the translator in decoding source texts and
helping the translator and technical writer to encode texts with specic pragmatic
features.
 PHRASEOLOGY IN BIOMEDICAL TEXTS 263

Corpus selection

Translation oriented terminological work can benet greatly from corpus studies.
We believe that in order for the corpus to be representative it must contain as
many text types and genres as are relevant within the domain. For our study we
are in the process of elaborating a corpus of comparable (original) and parallel
(translated) biomedical texts in English and Spanish on dierent aspects within
the domain of Oncology. The texts are checked for adequacy from a terminologi-
cal point of view; in this regard experts in biology and oncology have validated
the selection of specialized sources. Since the corpus must also be adequate from
a translational perspective, it is indicative of communicative situations that are
typical of professional contexts in translation. The text types included are:
1. Specialized publications in Oncology
2. Specialized publications in the area of General Medicine
3. Texts on cancer aimed at the general public
4. University textbooks
5. Information leaets

Methodology

Terms and phraseological units have been assigned to the particular concepts
they are related to and located within the dierent dimensions in the domain of
Oncology following an onomasiological approach. With the aid of a KWIC (Key
Word in Context) tool, the phraseological information present in the corpus has
been extracted within the dierent types of texts classied according to the
degree of specialization. The data have rst been extracted and described in each
language and then contrasted interlinguistically.
The concept apoptosis used to illustrate this analysis presents the following
conceptual features:
1. According to the literature, it is closely linked to cancer and more speci-
cally to its causes and treatments; apoptosis or programmed cell death plays
an important part in several processes to prevent illnesses and maintain
homeostasis, therefore, defects in the mechanisms of programmed cell death
can lead to dierent alterations.
2. It is a multidisciplinary concept since it is part of scientic domains such as
Cell Biology, Genetics, Biochemistry, Developmental Biology and medical
domains such as Oncology, Infectious Diseases, Immunology, Neurology,
264 MARIBEL TERCEDOR-SNCHEZ

etc. This characteristic appears to be common to many concepts in the

domain of Oncology, a typically multidisciplinary domain.
3. It has quite recently begun to be commonplace in scientic literature. It is
signicant to underline that apoptosis was the most cited subject in 1997
(three of the 10 most cited articles in this year dealt with apoptosis, accord-
ing to the Institute of Scientic Information of Philadelphia).
The following is an extract of concordances of the term apoptosis:

1 ma, conlleva a un aumento de la apoptosis con dosis subptimas de isp

2 n la regulacin del fenmeno de
3 e mama se asocia ala prdida de
4 -2 (Bcl-2, Bax, Bax, Bcl-x), la
5 cin de p53 como la muerte por
6 resin de Bcl-2 y la prdida de
7 e Bcl-2 junto con la prdida de
8 enciada por la regulacin de la
9 resin de Bcl-2 y la prdida de
10 pliquen su capacidad de inducir
11 p53, que regulan el proceso de
12 -Brain1 mientras que aumenta la
13 por lo que la regulacin de la
14 en procesos en los cuales la
15 ad que se traduce en prdida de
16 oncogenes rho demuestran que la
17 la importancia de la prdida de
apoptosis. Cuando se cultivan en edio
apoptosis de las clulas, hemos anali
apoptosis de las clulas tumorales, a
apoptosis de las clulas BAF3 ometida
apoptosis del tumor se asocia a un um
apoptosis del tumor aumenta el riesgo
apoptosis del tumor. Entre las protei
apoptosis del tumor se asocia a un um
apoptosis. Dentro de este contexto, e
apoptosis. El tratamiento de las lul
apoptosis en las clulas que etastati
apoptosis en subpoblaciones tumorales
apoptosis supone un acontecimiento fu
apoptosis en las clulas 435-Brain1 i
apoptosis en fibroblastos requiere ac
apoptosis en la progresin metasttic

1.leukemic cells to corticosteroids. Apoptosis after glucocorticoid the

2.human colon cancer cells underwent apoptosis after hyperthermia and v
3.c SHE cells. Chrysotile induced apoptosis after a considerably lon
4.ervation in order to demonstrate apoptosis after denervation. METHO
5.h cellular differentiation but not apoptosis after RA treatment in
6.indle, causes mitotic arrest and apoptosis after a prolonged incuba
7.ies increase in cells undergoing apoptosis after exposure to 4HPR.
8.novel function of IGFBP-3 as an apoptosis-inducing agent and show
9.eatment with TGF-beta1, a known apoptosis-inducing agent, resulted
10.chemistry and molecular biology of apoptosis already published on th
11.index of 50%) in adenocarcinoma. Apoptosis also increased in hyperp
12.rs of programmed cell death and apoptosis. Alterations in the expr
13. blocks programmed cell death (apoptosis). Although bcl-2 has bee
14. arrest or programmed cell death (apoptosis). Although evidence exis
15.rating cell cycle and induction of apoptosis. Although there are two

Figure 1: Concordances of the term apoptosis from texts in Spanish and English, marked
as formal and addressed to the researcher.
 PHRASEOLOGY IN BIOMEDICAL TEXTS 265

Units for description

The description of phraseology in biomedical texts should take into account all
units that transfer specialized information. In this sense, as Rey (1995: 131)
points out, terminology deals with units (terms, nouns) of sociolinguistically
variable nature, ranging from terms only known to a few users of a restricted
eld of knowledge or a professional group to terms known by the general
public. The author reveals the importance of focussing on the specialized
content and not on the specialized form of the language that might well be that
of a general word. Furthermore, it is impossible to separate words from terms,
[] there are a large number of lexical units which are both words and terms
(De Bess 1995: 3). The analysis of phraseology can reveal the interaction
between general and specialized language for dierent communicative situations
and is crucial in the sociolinguistic aspect of terminology. It is crucial to consider
all types of texts in terminological description for the translator and the technical
writer who work in dierent contexts. The analysis of concordances reveals the
relevant collocators for a particular unit. These must be inserted into the database
or information system for the access of the translator. Some of the most promi-
nent collocational data for the central term of this analysis are:
(1) V + N (apoptosis): induce, increase, inhibit, trigger (Eng); inducir, provocar,
aumentar, inhibir (Spa).
(2) Nphrase + N (apoptosis): increase of, degradation of, inhibition of (Eng);
aumento de la, prdida de, fenmeno de, control de (Spa).
(3) N(apoptosis) + N phrase: in #cancer cells, control (Eng); en la progresin
metastsica, en clulas hematopoyticas, en clulas 435 brain (Spa).
A close study of concordances can reveal the activation and interrelation of
dimensions within the domain. In this sense, the term in the example is related
to dierent dimensions in the domain. These conceptual dimensions or subelds
can be observed through the collocators that co-occur with the term. For
example, the numerous occurrences of the collocator #cell reveal that the term
apoptosis, within the domain of Oncology, is relevant in the subeld of Cell
Processes.
An analysis of the main implications of such phraseological analysis for
terminology and translation is set out below.
266 MARIBEL TERCEDOR-SNCHEZ

Phraseology and concept representation

The text sender, in order to explain concepts, resorts to forms that compare the
concept described with others in more familiar domains. It is common practice
to use a simple denition before or after introducing a specialized term, as
happens with cell suicide or apoptosis or anti-sickness drugs or antiemetics.
These paraphrases or denitions are phraseological units that replace the term
and carry out the same functions in the texts addressed to the non-specialist. The
units with a high metaphoric component are often those that transfer the notion
in a clearer way and oer valid information about the conceptual structure of the
domain.
In contrast with the data oered by other authors, in our corpus we have
found that the language of Oncology is full of phraseology made up of words
from the general language. More specically there are also many metaphors and
similes with a denitional function that bring specialized information closer to
the lay person. In this sense, in the concept APOPTOSIS, the metaphor cell
suicide claries its meaning if we consider that the cell, once it has commited
suicide, cannot reverse the process. Surrounding this concept, there are metaphors
that are informative about ways of representing knowledge in the domain; for
example, the p53 murder mystery. Complex processes are frequently explained
with similes: A picture is emerging of a complicated protein, which in some
ways resembles p53 in that it has multiple independent functions, rather like a
Swiss army knife (Reed 1997: 773).
The analysis of these units helps to locate the concepts within the domain
and also marks their relations to other domains. Moreover, it can highlight the
conceptual schema for the lexical units of the domain. In this regard, termino-
graphy as well as translation can benet from context-oriented terminological
studies that reveal the conceptual structure of the domain, with certain limitations
as to the routes for term formation: Any attempt to discover regularities in term
formation must be fully aware of the limited usefulness of this enterprise [] the
linguistic sign for a concept can be quite arbitrarily chosen and often is (Sager
1990: 62).

Phraseology and concept identification

One of the greatest obstacles in identifying concepts is the existence of concepts

represented by a word of the general language. As will be seen, the phrase cell
suicide appears to be closely linked to the causes of cancer, and in this context
 PHRASEOLOGY IN BIOMEDICAL TEXTS 267

the collocational patterns reveal the specialized content of the concept.

Furthermore, this problem also occurs in the case of complex terminological
units. When there is a component of the general language in a compound term,
the diculties of tagging it as specialized are considerable, as would be the case
with cell suicide program. Although the studies of frequency can be of some
help to see when a concept is specialized, these are not so reliable as they are in
lexicography (Meyer 1996: 8).
It is nevertheless important to distinguish between the common name of a
term (programmed cell death for apoptosis) and the name it is given as a
creative resource in order to give a particular audience access to the text
(cellular suicide for apoptosis). In the rst case, programmed cell death acts
as a full synonym, whereas in the second case, cellular suicide oers register
information, acting as a variant for the term apoptosis only suitable in certain
contexts.

Verbal Phrases in Terminological analysis

Most specialized vocabularies designed for the translator still contain names and
nominal phrases and rarely verbs and verb phrases (Pavel 1993: 29). Verb
analysis in specialized texts is crucial since there are terminology proper verbs
and verbs from the general language which acquire terminological meaning when
combined with certain collocators:
VPhr- V+ N:
recibir alquitrn: el pulmn recibe el alquitrn equivalente a
to receive tar: the patient receives an amount of tar comparable to
entrar a quirfano:no siempre el enfermo est en condiciones de entrar a
quirfano
to enter the operating room: not always is the patient ready to enter the
operating room
cometer suicidio: la apoptosis es una respuesta genticamente controlada para que
la clula cometa suicidio.
to commit suicide: apoptosis is a genetically controlled response for cells to
commit suicide.
268 MARIBEL TERCEDOR-SNCHEZ

Results

The need for transferring specialized information not only among specialists but
at all levels implies adapting the terminology to the particular audience. This
occurs for example through the use of lexical units from the general language
and the use of metaphors that acquire terminological functions.
The concordance analysis of the concept under study has identied the
following morpho-syntactic features:
The concept APOPTOSIS is introduced by the term apoptosis alone or together
with several phraseological units through the following patterns:
[X llamado/called Y] [X o/or Y]
[Y o/or X] [X (Y)]
[Y (X)], [X ,Y,]
[X] [Y]
[X (X1)]
In the above example, X represents the term apoptosis and Y is the phraseo-
logical unit that can take the following relevant forms in Spanish and English
from a more formal or technical register to an informal register:
mecanismo programado de muerte programmed form of cell death
muerte programada de la clula programmed cell death
muerte celular programada
suicidio celular cell suicide
cellular suicide
programa de suicidio celular cellular suicide programme
muerte siolgica o programada
Frequently, compound terms, representative of a concept, are not considered
under the scope of phraseology in terminological studies. However, there are
many simple terms that can be replaced by a type of unit that is representative
of one concept. Such a unit is not a typical phraseological unit, which is usually
the combination of two or more concepts. This is the case of the concept
APOPTOSIS, which is represented by the term apoptosis and the various phraseo-
logical units shown. Furthermore, these units, far from being xed, which is a
typical characteristic of compound terms, manifest a high degree of variation.
Since they encode a specialized concept and transfer specialized knowledge to
dierent audiences, their terminological treatment is desirable from the interest
of skopos-dependent translation.
For the creation of a terminographical tool, information on this type of units
 PHRASEOLOGY IN BIOMEDICAL TEXTS 269

can be encoded in a hierarchical manner following a monolingual approach.

Results in the languages under analysis can be contrasted. This approach avoids
oering the user false interlingual equivalents and reveals the existence of
terminological gaps:

Procesos Celulares Cell Processes

muerte celular: proceso de destruccin cell death: process of cell destruction.
celular
apoptosis o apoptosis or
muerte celular programada: programmed cell death:

muerte celular controlada cell death genetically controlled

genticamente e inducida por and induced by the cell.
la propia clula.
mecanismo programado de muerte celular programmed form of cell death
muerte celular programada programmed cell death
muerte programada de la clula
muerte siolgica o programada cellular suicide programme
+ proceso de suicidio celular cellular suicide
suicidio celular cell suicide
Figure 2: Contrastive onomasiological representation of apoptosis. (For a vision of the
onomasiological structuring of the oncological event, see Faber, Forthcoming).

This way of representing information allows access to denitional information

surrounding concepts in every dimension (cell processes in our example). The
rapid access to information is guaranteed thanks to the economy of the deni-
tions and the hierarchical structure of the concepts classied with onomasio-
logical criteria. However, given the interdisciplinary nature of the concepts in
many biomedical areas, the registers must allow access to multidimensionality
(this concept is considered part of the causes and treatments of cancer) and
interdisciplinary features (the concept belongs to many dierent areas, having
distinguishing functions in each of them). This latter task is beyond the limits of
this paper.
This type of units vary according to the pragmatic features of the texts
where they appear and therefore it is important to include all relevant varieties
and mark them with a pragmatic tag in the database, so as to give evidence of
the spectrum of registers included in the corpus. The information on every unit
can be held in a phraseological data model which should include (1) information
on register, tone, intention, connotation as well as the audience the text is aimed
at, and (2) as many contextual examples as well as validation information on
</TARGET "ter">

270 MARIBEL TERCEDOR-SNCHEZ

bibliographical sources. Attention should be paid to the exchange ability of

variants in dierent contexts.

Conclusions

The language of biomedicine is full of phraseological units that vary according

to the pragmatic features of the texts where they appear and are an important
marker of the type of discourse. Before compiling phraseological data, it is
crucial to carry out a descriptive study of phraseology in the domain under
analysis.
This ongoing research has shown that terms and units from the general
language represent specialized information within the domain of Oncology.
Phraseological information reveals the combination of both as well as the
preferable use of one or another in dierent contexts. Similar studies could be
carried out to extrapolate these conclusions to other specialized domains and
terminology in general.
Translation-focussed terminological work should reect the pragmatic and
conceptual idiosyncrasies of the domain under study. In the pragmatic aspect of
terminology, more attention should be paid to the representation of usage
information in term banks and other information systems. The translator and the
technical writer can benet from obtaining access to phraseological variation
through phraseological banks at their disposal.
With regard to the conceptual aspect of the discipline, monolingual ap-
proaches should be harnessed in order to compare domains among languages and
study conceptual and terminological gaps. Finally, the terminologist should
search for ways of representing concept multidimensionality and interdisciplinar-
ity and their role in the dierent ways of naming concepts in discourse. This
analysis tries to underline the importance of a symbiotic work between the
translator, the terminographer and the terminologist in all areas of specialization.
<TARGET "ref" DOCINFO

AUTHOR ""

TITLE "References"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

References

Abella, R. 1996. La vida cotidiana bajo el rgimen de Franco. Madrid: Temas de Hoy.
Agar, M. 1991. The Biculture in Bilingual. Language in Society 20: 167181.
Agar, M. 1992. The Intercultural Frame. Manuscript.
Agost, R. 1995. The Colloquial Register and Dubbing. In P. Jansen (ed.), Translation
and the Manipulation of Discourse, Selected Papers of the CERA Research Seminars
in Translation Studies 19921993. Leuven: Leuven Research Center for Translation,
Communication and Cultures.
Aguilar, C. 1991. Gua del vdeo-cine. Madrid: Ctedra.
Ahrens, B. 1994. Belastung beim Konsekutivdolmetschen: Strategisches Vorgehen und
nonverbales Verhalten. Diploma thesis. Ruprecht-Karls-Universitt Heidelberg.
Alln, S. 1997. The Re-establishment of Good Taste. On the Role of Academies in the
21st century. Parlaments pronunciats en lacte acadmic del dia 18 de juny de 1997.
Barcelona: Institut dEstudis Catalans.
Alvarado, R. 1983. Los nombres de los taxones y su espaolizacin: estudio del
problema sobre un caso prctico. Boletn de la Real Academia Espaola 63:
227239.
Ammann, M. 1990. Grundlagen der modernen Translationstheorie. Ein Leitfaden fr
Studierende. 2nd ed. Heidelberg: Institut fr bersetzen und Dolmetschen.
Anderson, J. R. 1983. The Architecture of Cognition. Cambridge: Harvard University
Press.
Anuario El Pas. 1998. Madrid: PRISA.
Anzalda, G. 1987. Borderlands/La Frontera. San Francisco: aunt lute books.
Aparicio, F. 1991. Versiones, interpretaciones y creaciones. Gaiterburg: Ediciones
Hispamrica.
Arnau i Gras, J. 1995. Diseos experimentales en esquemas. Barcelona: Publicacions de
la Universitat de Barcelona.
Aschenbach, G. 1992. Forschungsmethoden. In R. Asanger and G. Wenninger (eds),
Wrterbuch der Psychologie. Weinheim: Psychologie Verlags Union, 184194.
Atari, O. F. 1994. The place of communicative strategies in translation. Babel 40(2):
6576.
vila, A. 1997. La censura del doblaje cinematogrco en Espaa. Barcelona: CIMS.
Bacard, M., Fontcuberta, J. and Parcerisas, F. (eds). 1998. Cent anys de traducci al
catal (18911990). Barcelona: Eumo.
272 REFERENCES

Baddeley, A. and Hitch, G. 1974. Working Memory. In G. H. Bower (ed.), The

psychology of learning and motivation. Vol. 8. London: Academic Press.
Baker, M. 1992. In Other Words. A Coursebook on Translation. London, New York:
Routledge.
Ballester, A. 1995. La poltica del doblaje en Espaa (Eutopas 2 poca, Documentos de
trabajo, Vol. 94). Valencia: Episteme.
Barnes, R. D. 1984. Zoologa de los Invertebrados. Trans. R. Elizondo Mata. Mxico:
Interamericana.
Barthes, R. 1973. O Prazer do Texto. Lisboa: Edies 70.
Bassnett, S. and Lefevere, A. (eds). 1990. Translation, History and Culture. London:
Pinter.
Bassols, M. (ed.). 1995a. Adapting Translation for Dubbing to The Customers Require-
ments. In Y. Gambier (ed.), Communication audiovisuelle et transferts linguistiques.
Actes du colloque de Strasbourg. Translatio, December 1995: 410423.
Bassols, M. 1995b. Le modle de langue des chanes de tlvision catalanes (TV3, Canal
33). In Y. Gambier (ed.), Communication audiovisuelle et transferts linguistiques.
Actes du colloque de Strasbourg. Translatio, December 1995: 478492.
Baumann, K.-D. 1992. Die Fachlichkeit von Texten als eine komplexe Vergleichsgre.
In K.-D. Baumann and H. Kalverkmper (eds), Kontrastive Fachsprachenforschung.
Tbingen: Narr, 2948.
BD. 1986. Mmorandum. Enseignement et ralit professionnelle. Unpublished
document.
Beattie, J. 1997. Sutil, falso y traidor Tres versiones de Richard III de Shakespeare.
In F. Lafarga (ed.), Traducci i literatura. Homenatge a ngel Crespo. Barcelona:
Eumo, 163172.
Beaugrande, R. de. 1978. Factors in a theory of poetic translating. Assen: Van Gorcum.
Beaugrande, R. de. 1980. Text, Discourse and Process. London: Longman.
Beaugrande, R. de and Dressler, W. 1981. Introduction to Text Linguistics. London:
Longman.
Beeby, A. 1996. Teaching Translation from Spanish to English. Ottawa: Ottawa University
Press.
Begoa, F. M. 1956. El Sacerdote y el Cine. Revista Internacional del Cine 26: 1319.
Blanger, G. 1992. tude des relations cohsives grammaticales: perspectives traducto-
logiqe et typologique. Doctoral thesis. University of Sherbrooke.
Bell, R. 1991. Translation and Translating. London: Longman
Bell, R. 1997. Psycholinguistic/Cognitive Approaches to Translation. In M. Baker (ed.),
Encyclopaedia of Translation Studies. London, New York: Routledge.
Beltrn, Roco. 1999. La comercializacin de los vinos espaoles en el mercado britnico.
Final year dissertation. Universidad de Granada.
Benjamin, W. 1969. The Work of Art in the Age of Mechanical Reproduction. In H.
Arendt (ed.), Illuminations. Trans. Harry Zohn. New York: Scocken, 217252.
Bennett, T., Martin, G. and Mercer, C. (eds). 1981. Culture, Ideology and Social Process.
London: Batsford.
 REFERENCES 273

Berman, A. 1985. Lauberge du lointain. In A. Berman (ed.). Sur les tours de Babel.
Mauvezin: Trans-Europ.
Black, G. D. 1994. Hollywood Censored: morality codes, Catholics and the movies.
Cambridge: Cambridge University Press.
Bonet, A. 1957. Texto completo de la Encclica Miranda Prorsus . Revista Inter-
nacional del Cine 30: 2032.
Bourdieu, P. 1979. Distinction: Critique Sociale du Jugement. Paris: Minuit.
Bourdieu, P. 1986. Distinction. Trans. Richard Nice. London: Routledge.
Broch, A. 1991. Literatura catalana dels anys vuitanta. Barcelona: Edicions 62.
Brooks, F. 1992. Can we talk?. Foreign Language Annals 25: 5971
Brusca, R. C. and Brusca, G. J. 1990. Invertebrates. Sunderland: Sinauer Associates.
Bhler, H. 1984. Textlinguistische Aspekte der bersetzungsdidaktik. In W. Wilss and
G. Thome (eds), Die Theorie des bersetzens und ihr Aufschluwert fr die ber-
setzungs- und Dolmetschdidaktik. Tbingen: Narr, 250259.
Burke, J. A. 1955. El magisterio de la Iglesia y la clasicacin moral de las pelculas.
Revista Internacional del Cine 1314: 8790.
Bush, P. 1997. The Translator as Reader and Writer. Donaire 8: 1318.
Butler, J. 1990. Gender trouble: feminism and the subversion of identity. New York:
Routledge.
Cabrera Infante, G. 1998. LEER. El magazine literario 2: 4851.
Campbell, A. C. 1983. Gua de campo de la ora y fauna de las costas de Espaa y de
Europa. Trans., adapt. J. Ros i Aragons. Barcelona: Ediciones Omega.
Canale, M. 1983. From communicative competence to communicative language
pedagogy. In J. C. Richards and R. W. Schmidt (eds), Language and Communica-
tion. London: Longman.
Cannon, J. 1998. Spain as a marketing tool: an examination of perceptions of Spanish
products. International Journal of Iberian Studies 11(2): 112122.
Catford, J. C. 1965. A Linguistic Theory of Translation. London: Oxford University Press.
Chamberlain, L. 1988. Gender and the metaphorics of translation. Signs 13(3): 454
472.
Chen, S. Q. 1990. A study of communication strategies in interlanguage production in
Chinese EFL learners. Language Learning 40: 155187.
Chesterman, A. 1997. Memes of Translation. Amsterdam, Philadelphia: John Benjamins.
Chesterman, A. 1998. Contrastive Functional Analysis. Amsterdam, Philadelphia: John
Benjamins.
Chuquet, H. and Paillard, M. 1989. Approche linguistique des problemes de traduction.
Paris: Ophrys.
Cixous, H. 1986. Sorties. In The Newly Born Woman. Trans. B. Wing. Minnesota,
Minn.: The University of Minnesota Press, 63132.
Coelho, N. N. 1981. Literatura infantil; histria, teoria, anlise: das origens orientais ao
Brasil de hoje. So Paulo: Quron.
274 REFERENCES

Collados Ais, A. 1996. La entonacin montona como parmetro de calidad en la

interpretacin simultnea: la evaluacin de los receptores. Doctoral thesis. Universi-
dad de Granada
Comissi de Normalitzaci Lingstica 1993. Orientacions lingstiques per al doblatge.
Barcelona: Televisi de Catalunya.
Cook, G. 1992. The Discourse of Advertising. London: Routledge.
Cooper, J. C. 1986. Cuentos de hadas; alegoras de los mundos internos. Trad. X. Huasi.
Mlaga: Sirio.
Corbin, R. M. 1980. Decisions that might not get made. In T. E. Wallsten (ed.), Cognitive
processes in choice and decision behaviour. Hillsdale NJ: Erlbaum, 4767.
Damasio, A. R. and Damasio, H. 1992. Brain and Language. Scientic American
September: 8995.
Danaher, P. 1992. Deaf and blind but not dumb: translation theorists. Language
International 4(2): 1516.
Dancette, J. 1995. Parcours de Traduction: tude exprimentale du processus de compr-
hension. Lille: Presses Universitaires de Lille.
Danks, J. 1991. The psycholinguistics of reading and translation. Fundamental Ques-
tions in Translation Theory. Universitt Leipzig. Manuscript.
De Bess, B. 1995. A paradoxical terminologist. In J. C. Sager (ed.), Essays on Termi-
nology. Amsterdam, Philadelphia: John Benjamins.
De Mooij, M. 1998. Global Marketing and Advertising: Understanding Cultural Paradox-
es. London: Sage.
Delisle, J. 1980. Analyse du Discours comme Mthode de Traduction. Ottawa: Ottawa
University Press.
Delisle, J. 1988. Translation: an interpretive approach. Trans. P. Logan and M. Creery.
Ottawa: Ottawa University Press.
Delisle, J. and Woodsworth, J. (eds). 1995. Translators through History. Amsterdam,
Philadelphia: John Benjamins.
Demers, G. 1992 Les statistiques au service de la pdagogie de la traduction. TTR 5(1):
101-111.
Dillinger, M. L. 1989. Component Processes of Simultaneous Interpreting. Doctoral thesis.
Mc Gill University, Montreal.
Drnyei, Z. 1995. On the Teachability of Communication Strategies. TESOL Quarterly
29(1): 5584.
Dressler, W. U. 1975. Textgrammatische Invarianz in bersetzungen? In E. Glich and
W. Raible (eds), Textsorten. Wiesbaden: Athenaion, 98106.
Dreyfus, H. L. and Dreyfus, S. E. 1986. Mind over machine. Oxford: Blackwell.
Dubuc, R. and Lauriston, A. 1997. Terms and contexts. In S. E.Wright and G. Budin
(eds), Handbook of Terminology Management. Vol. 1: Basic Aspects of Terminology
Management. Amsterdam, Philadelphia: John Benjamins.
Eco, H. 1993. Apocalpticos e Integrados. So Paulo: Perspectiva.
Eggins, S. 1994. An Introduction to Systemic Functional Linguistics. London: Pinter.
 REFERENCES 275

Elena Garca, P. 1990. Aspectos tericos y prcticos de la traduccin. Salamanca:

Publicaciones de la Universidad.
Ensinger, D. 1997. Vermittlung und Verarbeitung von landeskundlichem Wissen am
Computer. In E. Fleischmann, W. Kutz and P. A. Schmitt (eds), Translations-
didaktik. Grundfragen der bersetzungswissenschaft. Tbingen: Narr, 368377.
Esselink, Bert. 1998. A practical Guide to Software Localization. Amsterdam, Philadel-
phia: John Benjamins.
Even-Zohar, I. 1978. The Position of Translated Literature in the Literary Polysystem.
In J. S. Holmes, J. Lambert and R. van den Broeck (eds), Literature and Translation:
New Perspectives in Literary Studies with a Basic Bibliography of Books on Transla-
tion Studies. Leuven: Acco, 117127 .
Fabbro, F. and Gran, L. 1997. Neurolinguistic Aspects of Simultaneous Interpretation.
In I. Gambier, D. Gile and C. Taylor (eds), Conference Interpreting: Current Trends
in Research. Amsterdam, Philadelphia: John Benjamins, 927.
Faber, P. Forthcoming. Concept specication and analysis in the conguration of
specialized domains. In P. Fernndez Nistal (ed.), A Spectrum of Translation
Studies. Valladolid: Universidad de Valladolid.
Faerch, C. and Kasper, G. 1983. Plans and strategies in foreign language communica-
tion. In C. Faerch and G. Kasper. Strategies in Interlanguage Communication.
London: Harlow.
Ferreira, J. P. 1993. O Livro de So Cipriano: uma Legenda de Massas. So Paulo:
Perspectiva.
Fitch, B. T. 1988. Beckett and Babel. Toronto, Bualo, London: University of Toronto
Press.
Frankfort-Nachmias, C. and Nachmias, D. 1992. Research Methods in the Social Sciences.
London, Melbourne, Auckland: Edward Arnold.
Fraser, J. 1994. Translating Practice into Theory: A Practical Study of Quality in
Translator Training. In C. Picken (ed.), ITI Conference 7 Proceedings, London:
Institute of Translation and Interpreting.
Fraser, J. 1996. The Translator Investigated. Learning from Translation Process Analy-
sis. The Translator 2(1): 6579.
Funch, P. and Kristensen, R. M. 1995. Cycliophora is a new phylum with anities to
Entoprocta and Ectoprocta. Nature 378: 711714.
Fusco, C. 1995. English is broken here. New York: The New Press.
Fuss, D. 1989. Essentially speaking. London: Routledge.
Fuss, D. 1995. Identication papers. London: Routledge.
Garca Lpez, R. 1996. Lo implcito y lo explcito y su repercusin en la traduccin.
Doctoral thesis. Universidad de Las Palmas de Gran Canaria.
Gardner, R. 1992. Terminology Management at WordPerfect Corporation. In E.-F. Losa
(ed.), Frontiers: Proceedings of the 33rd Annual Conference of the American
Translators Association. Medford: Learned Information, vii, 311316.
Garrido, C. 1998. La traduccin al espaol de las designaciones alemanas de grupos
taxonmicos. In L. Flix Fernndez and E. Ortega Arjonilla (coord.), II Estudios
276 REFERENCES

sobre traduccin e interpretacin, vol. III. Mlaga: Universidad de Mlaga, 1057

1063.
Gile, D. 1987. Les exercices dinterprtation et la dgradation du franais: une tude de
cas. Meta 32(4): 420428.
Gile, D. 1995a. Regards sur la recherche en interprtation de confrence. Lille: Presses
Universitaires de Lille.
Gile, D. 1995b. Basic Concepts and Models for Interpreter and Translator Training.
Amsterdam, Philadelphia: John Benjamins.
Gile, D. 1995c. Fidelity Assessment in Consecutive Interpretation: An Experiment.
Target 7(1): 151164.
Gile, D. 1995d. La lecture critique en traductologie. Meta 40(1): 514.
Gile, D. 1999. Variability in the perception of delity in simultaneous interpretation.
Hermes 22: 51-79.
Gpferich, S. 1995. Textsorten in Naturwissenschaft und Technik. Pragmatische Typologie
Kontrastierung Translation. Tbingen: Narr.
Goris, O. 1993. The Question of French Dubbing: Towards a Frame for Systematic
Investigation. Target 5 (2): 162190.
Gouanvic, J.-M. 1997. Translation and the Shape of Things to Come: The Emergence of
American Science Fiction in Post-War France. The Translator 3(2): 125152.
Gran, L. and Taylor, C. (eds). 1990. Aspects of Applied and Experimental Research on
Conference Interpretation. Udine: Campanotto Editore.
Gran, L. and Viezzi, M. 1995. Development of Research Work at SSLM, Trieste.
Target 7(1): 107118.
Grice, H. 1975. Logic and Conversation. In P. Cole and J. Morgan (eds), Syntax and
Semantics 3: Speech Acts. New York: Academic Press, 4158
Griths, D., Heppell, S., Millwood, R. and Mladenova, G. 1995. Translating software:
what it means and what it costs for small cultures and large cultures. The LISA
Forum Newsletter IV(3): 411.
Grinev, S. V., Jermakow, J. S., Morozow, B. V. and Grinev, A. S. 1991. The problems of
medical terminology. In Proceedings of the Third INFOTERM symposium: Termi-
nology Work in Subject Fields. Viena: Termnet.
Gubern, R. 1981. La censura. Funcin poltica y ordenamiento jurdico bajo el franquismo
(19361975). Barcelona: Pennsula.
Gumpert, C. 1995. Conversaciones con Antonio Tabucchi. Barcelona: Anagrama.
Haarstrup, K. and Phillipson, R. 1983. Strategies in Interlanguage Communication.
London: Longman.
Halliday, M. A. K. 1985. An Introduction to Functional Grammar. London: Edward
Arnold.
Halliday, M. A. K. and Hassan, R. 1976. Cohesion in English. London: Longman.
Hansen, G. et al. 1998. The translation process: from source text to target text. In:
Hansen, G. (ed.), Copenhagen Working Papers in LSP. Copenhagen: University of
Copenhagen, 5971.
Hansen, K. (ed.). 1983. Studien zur Sprachkonfrontation. Berlin: Humboldt-Universitt.
 REFERENCES 277

Hardin, G. and Picot, C. 1990. Translate. Initiation la pratique de la traduction. Paris:

Dunod.
Hartmann, R. R. K. 1980. Contrastive Textology: Comparative Discourse Analysis in
Applied Linguistics. Heidelberg: Groos.
Hatim, B. and Mason, I. 1990. Discourse and the Translator. London: Longman.
Hellwig, P. 1984. Titulus oder ber den Zusammenhang von Titeln und Texten. Zeit-
schrift fr germanistische Linguistik 12(1): 120.
Holmes, J. S. 1988 The Name and Nature of Translation Studies. In Translated! Papers
on Literary Translation and Translation Studies. Amsterdam: Rodopi, 6680.
Holz-Mnttri, J. 1984. Translatorisches Handeln. Theorie und Methode. Helsinki:
Suomalainen Tiedeakatemia.
Hnig, H. G. 1988. Wissen bersetzer eigentlich, was sie tun? Lebende Sprachen 33(1):
1014.
Hnig, H. G. 1991. Holmes mapping theory and the landscape of mental translation
processes. In K. Leuven-Zwart, and T. Naaijkens (eds), Translation Studies: The
State of the Art. Amsterdam: Rodopi, 7790.
Hnig, H. G. and Kussmaul, P. 1982. Strategie der bersetzung: Ein Lehr- und Arbeits-
buch. Tbingen: Narr.
Hurtado Albir, A. 1996a. La cuestin del mtodo traductor. Mtodo, estrategia y tcnica
de traduccin. Sendebar 7: 3957.
Hurtado Albir, A. 1996b. Modalidades y tipos de traduccin. Vasos comunicantes 4: 1927.
Jskelinen, R. 1993. Investigating Translation Strategies. In S. Tirkkonen-Condit and
J. Laing (eds), Recent Trends in Empirical Translation Research. Joensuu:
University of Joensuu.
Jskelinen, R. 1997. Think-aloud protocols. In M. Baker (ed.), Enclyclopedia of
Translation Studies. London: Routledge.
Jakobson, R. 1960. Linguistics and Poetics. In T. A. Sebeok (ed.), Style in Language.
Cambridge Mass.: MIT Press, 350377.
Jones, R. 1998. Conference Interpreting Explained. Manchester: St. Jerome.
Jovanovic, M. 1990. On translating titles. Babel 36(4): 213222.
Kalina, S. 1996. Strategische Prozesse beim Dolmetschen. Theoretische Grundlagen,
empirische Fallstudien, didaktische Konsequenzen. Doctoral thesis. Ruprecht-Karls-
Universitt Heidelberg.
Kaminsky, A. K. 1993. Feminist Criticism and Latin American Women Writers. Minnesota,
Minn.: University of Minnesota Press.
Kaner, C., Falk, J. and Nguyen, H. Q. 1993. Testing Computer Software. New York: Van
Nostrand Reinhold.
Kano, N. 1995. Developing International Software for Windows 95 and Windows NT.
Redmond: Microsoft Press.
Kelly, D. 1997. Prensa e identidad nacional. La imagen de Espaa en la prensa britnica.
Doctoral thesis. Universidad de Granada.
Kiraly, D. C. 1990. Toward a systematic approach to translation skills instruction. Doctoral
thesis. Urbana IL.
278 REFERENCES

Kiraly, D.C. 1995. Pathways to Translation. Pedagogy and Process. Kent, London: The
Kent State University Press.
Kiraly, D. C. 1997. Think-aloud protocols and the construction of a professional self-
concept. In J. H. Danks et al. (eds), Cognitive processes in translation and interpret-
ing. Thousand Oaks CA: Sage.
Kohn, K. and Kalina, S. 1996. The strategic dimension of interpreting. Meta XLI(1):
11837.
Koller, W. 1992. Einfhrung in die bersetzungswissenschaft. 4th ed. Heidelberg: Quelle
und Meyer.
Koller, W. 1995. The Concept of Equivalence and the Object of Translation Studies.
Target 7(2): 4964.
Komissarov, V. 1985. The practical value of translation theory. Babel 31(4): 208212.
Kopczynski, A. 1980. Conference Interpreting. Some Linguistic and Communicative
Problems. Nauk, Wydawn, Poznan: UAM.
Kourgano, V. 1958. La recherche scientique. Paris: PUF.
Krawutsche, P. (ed.). 1989. Translator and Interpreter Training and Foreign Language
Pedagogy. Binghamton: The State University of New York at Binghamton.
Kreutzer, M. and Neunzig, W. 1998. En torno a la investigacin emprica en el campo
de la traductologa. Quaderns. Revista de Traducci 1: 127134.
Krings, H. P. 1986. Was in den Kpfen von bersetzern vorgeht. Eine empirische Unter-
suchung der Struktur des bersetzungsprozesses an fortgeschrittenen Franzsisch-
lernern. Tbingen: Narr.
Krings, H. P. 1988. Blick in die Black Box Eine Fallstudie zum bersetzungs-
prozess bei Berufsbersetzern. In R. Arntz (ed.), Textlinguistik und Fachsprache:
Akten des Internationalen bersetzungswissenschaftlichen AILA-Symposiums.
Hildesheim: Olms, 393412.
Kuhn, T. S. 1970. The Structure of Scientic Revolutions. Chicago: University of Chicago
Press.
Kumarajiva. 1984. Wei Sengrui lun xifang ci ti (A talk to Sengrui about language style
of the West). In Luo Xinzhang (ed.), 32.
Kundera, M. 1992. LImmortalit. Paris: Gallimard.
Kurz, I. 1983. Berufszufriedenheit und soziales Prestige. FIT Newsletter 2(4): 130135.
Kurz, I. 1996. Simultandolmetschen als Gegenstand der interdisziplinren Forschung.
Wien: WUV Universittsverlag.
Kussmaul, P. 1986. bersetzen als Entscheidungsprozess. Die Rolle der Fehleranalyse
in der bersetzungsdidaktik. In M. Snell-Hornby (ed.), bersetzungswissenschaft.
Eine Neuorientierung. Tbingen: Francke.
Kussmaul, P. 1991. Creativity in the Translation Process: Empirical Approaches. In
K. M. van Leuven and T. Naaijkens (eds), Translation Studies: The State of the Art.
Amsterdam: Rodopi.
Lamberger-Felber, H. 1998. Der Einuss kontextueller Faktoren auf das Simultandol-
metschen. Eine Fallstudie am Beispiel gelesener Reden. Doctoral thesis. Karl-
Franzens-Universitt Graz.
 REFERENCES 279

Lambert, J. 1994. The Cultural Component Reconsidered. In M. Snell-Hornby (ed.),

Translation Studies An Integrated Approach. Amsterdam, Philadelphia: John
Benjamins.
Lambert, J. 1995. A Traduo. In M. Angenot, J. Bessire, D. Fokkema and E. Kushner
(eds), Teoria Literria. Problemas e Perspectivas. Lisboa: Publicaes Dom Quixote.
Lambert, S. and Moser-Mercer, B. (eds). 1994. Bridging the Gap: Empirical research in
simultaneous interpretation. Amsterdam, Philadelphia: John Benjamins.
Lamo de Espinosa, Emilio (dir). 1996. Conclusiones. La imagen exterior de Espaa.
Aspectos comerciales, tursticos e inversores. Manuscript.
Lauer, I.-A. 1986. Fachtextlinguistische Untersuchungen zum Kommunikationsbereich der
Pdagogischen Psychologie dargestellt an ausgewhlten Fachtextsorten im
Englischen. Doctoral thesis. Universitt Leipzig. Revised edition: Busch-Lauer, I.-A.
1991. Englische Fachtexte in der Pdagogischen Psychologie. Eine linguistische
Analyse. Frankfurt/M. etc.: Peter Lang.
Leech, G. and Short, M. H. 1981. Style in Fiction. A Linguistic Introduction to English
Fictional Prose. London: Longman.
Lefevere, A. 1992. Translation, Rewriting, and the Manipulation of Literary Fame.
London: Routledge.
Lev, J. 1967. Translation as a Decision Process. In To Honor Roman Jakobson: Essays
on the Ocasion of his Seventieth Birthday. Vol. 2. The Hague: Mounton, 11711182.
Localisation Resources Centre Yearbook 1997. Dublin.
Lpez Guix, J. G. and Minett Wilkinson, J. 1997. Manual de traduccin ingls/castellano.
Teora y prctica. Barcelona: Gedisa.
Lrscher, W. 1991. Translation Performance, Translation Process, and Translation
Strategies. Tbingen: Narr.
Lrscher, W. 1992a. Investigating the Translation Process. Meta XXXVII(3): 426439.
Lrscher, W. 1992b. Process-oriented research into translation and its implications for
translation teaching. Interface: Journal of Applied Linguistics 6(2): 105117.
Lrscher, W. 1993. Translation Process Analysis. In Y. Gambier and J. Tommola (eds),
Translation and Knowledge. Turku: University of Turku.
Lrscher, W. 1996. A psycholinguistic analysis of translation processes. Meta XLI(1):
2632.
Lu, Cheng. 1979. Zhongguo Fo xue liuyuan lue jiang (Lectures on the origine and spread
of Buddhism in China). Beijing: Zhonghua shuju.
Lugones, M. 1994. Purity, impurity, and separation. Signs 19(2): 458479.
Luo, Xinzhang. (ed.). 1984. Fanyi lunji (Collected essays on translation). Beijing:
Shangwu yinshuguan.
Lvvskaya, Z. 1992. Sistema de ejercicios en la enseanza de la traduccin. Guini-
guada 3, 309328.
Lvvskaya, Z. 1997. Problemas actuales de la traduccin. Granada: Lingstica y Mtodo
Ediciones.
Macdonald, D. 1983. Against the American Grain. New York: Da Capo.
280 REFERENCES

Madell, T., Parsons C. and Abbeg, J. 1994. Developing and Localizing International
Software. Englewood Clis: Prentice Hall.
Maier, C. 1994. Women in translation: current intersections, theory, and practice. Delos
16 (2): 2939.
Maier, C. 1996. On translation: the translators. In R. Christ (ed.), On translation: The
Translators. Atlanta: Atlanta College of Art Gallery, 5767.
Maillot, J. 1997. La traduccin cientca y tcnica. Trans. J. Sevilla Muoz. Madrid:
Gredos.
Mallafr, J. 1991. Llengua de tribu i llengua de polis. Barcelona: Quaderns Crema.
Malmkjaer, K. 1997. Linguistics in Functionland and through the Front Door. A
contribution to a Symposium on Translation Quality Assessment held at Aston
University, Birmingham. Manuscript.
Malone, J. L. 1988. The science of linguistics in the art of translation. Albany, NY: SUNY
Press.
Manchn R. 1994. Las estrategias del aprendiz de una L2: el estado de la cuestin. In
Serie sobre estrategias de aprendizaje y uso del lenguaje. Sevilla: Universidad de
Sevilla, 717
Mar, A. 1997b. Lexperincia de lautotraducci. In Quaderns Divulgatius, 8. V
Seminari sobre la Traducci a Catalunya. Barcelona: Associaci dEscriptors en
Llengua Catalana, 5363.
Mason, I. 1994. Techniques of Translation Revisited: a text-linguistic review of
borrowing and modulation. In A. Hurtado (ed.), Estudis sobre la traducci.
Castell: Universitat Jaume I, 6172.
Mason, I. 1995. Translation <<Techniques>>: A Procedural Approach. In I. Mason and
C. Pagnoulle (eds), Cross Words. Issues and Debates in Literary and Non-literary
Translation. Lige: Universit de Lige, 5968.
Mayoral Asensio, R. and Muoz Martn, R. 1997. Estrategias comunicativas en la
traduccin intercultural. In P. Fernndez Nistal and J. M. Bravo Gozalo (eds),
Aproximaciones a los Estudios de Traduccin. Valladolid: SAE.
McDonald, C. 1980. Myths, Evidence and Evaluation. Admap 16(11): 546555.
McDonald, C. 1993. How Advertising Works: A Review of Current Thinking. Henley upon
Thames: The Advertising Association, in association with NTC Publications Ltd.
Medawar, P. 1981. Advice to a Young Scientist. London, Sydney: Pan Books.
Meyer, I. and Mackintosh, K. 1996. Rening the terminographers concept-analysis
methods: how can phraseology help?. Terminology 3(1): 126.
Miller, G. A., Galanter, E. and Pribram, K. H. 1960. Plans and the Structure of Behavior.
New York: Holt, Rinehart and Winston.
Milton, J. 1995. A Traduo de Romances Clssicos do Ingls para o Portugus no
Brasil. In Trabalhos em Lingstica Aplicada 24. Instituto de Estudos da Lingua-
gem, UNICAMP, Campinas, Brazil.
Milton, J. 1996. The Translations of O Clube de Livro. In TradTerm 3: 4765.
Mondahl, M. and Jensen, K. A. 1992 Information Processing in a Translation Task.
Multilingua 11(2).
 REFERENCES 281

Mondahl, M. and Jensen, K. A. 1996. Lexical search strategies in translation. Meta. XLI
(1): 97113.
Moreno-Torres, I. 1996. Textos dentro del ordenador. Trans 1: 97109.
Multilingual Dictionary of Fish and Fish Products. 1995. Cambridge: Organisation for
Economic Cooperation and Development-Fishing News Books.
Muoz Martn, R. 1994. La lingstica cognitiva, la teora de la traduccin y la navaja
de Occam. Actas del I Taller Internacional sobre la Traduccin. La Habana:
Universidad de La Habana.
Muoz Martn, R. 1996. Lingstica para traducir. Barcelona: Teide.
Nedergaard-Larsen, B. 1993. Culture-bound problems in subtitling. Perspectives. Studies
in Translatology 2: 207241.
Neubert, A. 1968. Pragmatische Aspekte der bersetzung. In A. Neubert (ed.),
Grundfragen der bersetzungswissenschaft. Fremdsprachen, Beiheft II. Leipzig:
Enzyklopdie, 2133.
Neubert, A. 1973. Theorie und Praxis fr die bersetzungswissenschaft. In Proceedings
of the Third International Congress of Applied Linguistics. Vol. 3. Heidelberg:
Winter.
Neubert, A. 1985. Text and Translation. bersetzungswissenschaftliche Beitrge 8.
Leipzig: Enzyklopdie.
Neubert, A. 1996. Textlinguistics of Translation: The Textual Approach to Translation.
In Marilyn Gaddis Rose (ed.), Translation Horizons: Beyond the Boundaries of
Translation Spectrum. Binghamton: State University of New York at Binghamton,
87106.
Neubert, A. 1997. Postulates for a Theory of Translation. In J. H. Danks, G. M. Shreve,
S. P. Fountain and M. K. Mcbeath (eds), Cognitive Processes in Translation and
Interpreting. Thousand Oaks, London, New Delhi: Sage, 124.
Neubert, A. and Shreve, G. 1992. Translation as Text. Kent: The Kent State University
Press.
Neunzig, W. 1997a. Der Computer als Hilfsmittel beim Erwerben kognitiver ber-
setzungsstrategien. In E. Fleischmann, W. Kutz and P. A. Schmitt (eds), Trans-
lationsdidaktik. Grundfragen der bersetzungswissenschaft. Tbingen: Narr,
377384.
Neunzig, W. 1997b. Die Ezienz computergesttzter bungsformen: eine Untersuchung
im Rahmen des bersetzungsunterrichts. In J. Kohn, B. Rscho and D. Wol
(eds), New Horizons in Call. Szombathely: Eurocall, 303312.
Neunzig, W. 1998. El ordenador como profesor virtual en la formacin de traductores.
Simulacin de una clase de traduccin. In P. Orero (ed.), III Congrs Internacional
sobre Traducci. Bellaterra: Universitat Autnoma de Barcelona, 565579.
Newmark, P. 1988. A Textbook of Translation. New York, London: Prentice Hall.
Nida, E. A. 1964. Towards a Science of Translating. Leiden: E. J. Brill.
Nord, C. 1991. Text Analysis in Translation. Theory, Methodology and Didactic Applica-
tion of a Model for Translation-Oriented Text-Analysis. Amsterdam, Atlanta: Rodopi.
282 REFERENCES

Nord, C. 1993. Einfhrung in das funktionale bersetzen. Am Beispiel von Titeln und
berschriften. Tbingen: Francke.
Nord, C. 1997. Translating as a Purposeful Activity. Functionalist Approaches Explained.
Manchester: St. Jerome.
ODonnell, H. 1994. Mapping the mythical: a geopolitics of national sporting stereo-
types. Discourse and Society 5: 34580.
Ortega y Gasset, J. 1947. Miseria y esplendor de la traduccin. In Obras Completas de
Jos Ortega y Gasset. Vol. V. Madrid: Revista de Occidente, 433452.
Ortega y Gasset, J. 1983. La deshumanizacin del arte y otros ensayos de esttica. Madrid:
Revista de Occidente, Alianza.
Ortega y Gasset, J. 1994. La rebelin de las masas. Madrid: Revista de Occidente,
Alianza.
Padilla, P. 1995. Procesos de memoria y atencin en la interpretacin de lenguas. Doctoral
thesis. Universidad de Granada.
Paizis, G. 1998. Category Romances: Translation, Realism and Myth. The Translator
4(1): 124.
Parcerisas, F. 1997. Traducci, edici, ideologia. Aspectes sociolgics de les traduccions
de la Bblia i de lOdissea al catal. Doctoral thesis. Universitat Autnoma de
Barcelona.
Paribakht, T. 1986. On the pedagogical relevance of strategic competence. TESL,
Canada Journal 3: 5366.
Pascua Febles, I. 1994. Estudio sobre la traduccin de los ttulos de pelculas. In M.
Raders and R. Martn Gaitero (eds), IV Encuentros Complutenses en torno a la
traduccin. Madrid: Editorial Complutense, 349354.
Pascua Febles, I. 1996. La adaptacin en la traduccin de la literatura infantil. Doctoral
thesis. Universidad de Las Palmas de Gran Canaria.
Pascua Febles, I. and Peate Soares, A. L. 1991. Introduccin a los Estudios de
Traduccin. Las Palmas de Gran Canaria: Corona.
Pavel, S. 1993. Neology and Phraseology as Terminology-in-the-making. In H. B.
Sonneveld and K. L. Loening (eds), Terminology: Applications in Interdisciplinary
Communication. Amsterdam, Philadelphia: John Benjamins.
Pericay, X. and Toutain, F. 1996. El malents del noucentisme. Barcelona: Proa.
Peterson, R., Mountfort, G. and Hollom, P. A. D. 1989. Gua de campo de las aves de
Espaa y de Europa. Trans. adapt. M. Gonzlez Dez. Barcelona: Ediciones Omega.
Pchhacker, F. 1995. Those Who Do: A Prole of Research(ers) in Interpreting.
Target 7(1): 4764.
Pontiero, G. 1997. The Task of the Literary Translator. In P. Orero and J. Sager (eds),
The Translators Dialogue Giovanni Pontiero. Amsterdam, Philadelphia: John
Benjamins, 5566.
Popovic, A. 1973. The Contemporary State of the Theory of Literary Translation. Babel
24: 111113.
Presas, M. 1996. Problemes de traducci i competncia traductora. Bases per a una
pedagogia de la traducci. Doctoral thesis. Universitat Autnoma de Barcelona.
 REFERENCES 283

Pym. A. 1997. Pour une tique du traducteur. Artois: Presses de lUniversit dArtois.
Rabadn, R. 1991. Equivalencia y traduccin. Problemtica de la equivalencia translmica
ingls-espaol. Len: Universidad de Len.
Rabadn, R. 1994. Traduccin, intertextualidad, manipulacin. In A. Hurtado Albir
(ed.), Estudis sobre la traducci. Castell: Publicacions de la Universitat Jaume I,
129139.
Rabadn, R. and Chamosa, J. L. 1997. Traduccin y construccin cultural. In Trasvases
culturales: literatura, cine, traduccin 2. Vitoria: Universidad del Pas Vasco.
Rajaud, V. and Brunetti, M. 1992. Traducir. Initiation la practique de la traduction.
Paris: Dunod.
Rastall, P. 1994. Communication strategies and translation. The example of the geni-
tive in Russian. Babel 40(1): 3848.
Reed, J. C. 1997. Double identity for proteins of the Bcl-2 family. Nature 387:
773776.
Reiss, K. 1976. Texttyp und bersetzungsmethode. Der operative Text. Kronberg: Scriptor.
Resch, R. 1998. Translating persuasive discourse. Intertextuality in multi-semiotic texts.
Paper presented at the EST Congress, Granada.
Rey, A. 1995. Essays on Terminology. Ed. by J. C. Sager. Amsterdam, Philadelphia: John
Benjamins.
Riba, C. 1954. Prefaci. In Diccionari general de la Llengua Catalana de Pompeu Fabra.
Barcelona: Lpez Llauss.
Richards, I. A. 1972. Functions of and factors in language. Journal of Literary Semantics
1: 2540.
Rinsche, A. 1997. Translation Technology: The State of the Art. Localisation Resources
Centre Yearbook,132140.
Ritzer, G. 1998. The McDonaldisation Thesis. London: Sage.
Rivera Garretas, M. 1994. Nombrar el mundo en femenino: pensamiento de las mujeres y
teora feminista. Barcelona: Icaria.
Robinson, D. 1997. Becoming a translator. London, New York: Routledge.
Robyns, C. 1990. The Normative Models of Twentieth Century Belles Indles:
Detective Novels in French Translation. Target 2(1): 2342.
Rodrguez De La Fuente, F. 1970. Enciclopedia Salvat de la fauna. Pamplona: Salvat.
Rojas, J. 1987. A survey on job satisfaction of free-lance (and permanent) conference
interpreters domiciled in the Geneva area. Term-paper, Integrated Studies in Human
Resources, Webster University.
Rost, M. 1994. Communication strategies: are they teachable?. 28th Annual TESOL
Convention. Baltimore, Maryland.
Rost, M. and Ross, S. 1991. Learner use of strategies in interaction: typology and
teachability. Language Learning 41: 235273.
Rlker, K. 1972. Einige Probleme der bersetzung naturwissenschaftlich-technischer
Literatur unter besonderer Bercksichtigung des pragmatischen Aspekts. In H.
Spitzbardt (ed.), Spezialprobleme der wissenschaftlichen und technischen ber-
setzung. Halle (Saale): Niemeyer, 4458.
284 REFERENCES

Sager, J. C. 1990. A Practical Course in Terminology Processing. Amsterdam, Philadel-

phia: John Benjamins.
Sangrador Garca, J. L. 1981. Estereotipos de las nacionalidades y regiones de Espaa.
Madrid: CIS.
Santoyo, J. C. 1996. El delito de traducir. Len: Secretariado de Publicaciones de la
Universidad de Len.
Saramago, J. 1997. To Write is to Translate. In P. Orero and J. Sager (eds), The
Translators Dialogue Giovanni Pontiero. Amsterdam, Philadelphia: John
Benjamins, 8586.
Sarraz, V. 1992. Das Experiment. In R. Asanger and G. Wenninger (eds), Wrterbuch
der Psychologie. Weinheim: Psychologie Verlags Union, 167174.
Schner, C. 1993. Meaning and Knowledge in Translation. In Y. Gambier and J.
Tommola (eds), Translation and Knowledge. Turku: University of Turku.
Schmid, A. 1994. Gruppenprotokolle ein Einblick in die black box des bersetzers?
TexTconText 9(2): 121146.
Schmitt, P. A. 1990. Was bersetzen bersetzer? Eine Umfrage. Lebende Sprachen 35:
97106.
Schnpug, U. 1994. Sprache und Kultur. In K. Wessel and F. Naumann (eds),
Kommunikation und Humanogenese. Bielefeld: Kleine, 370383.
Scott, J. W. 1992. Experience. In J. Butler, and J. W. Scott (eds), Feminists theorize the
political. London: Routledge.
Scott, J. W. 1996. Feminism and History. Oxford: Oxford University Press.
Sguinot C. 1991. A study of student translation strategies. In S. Tirkonnen-Condit
(ed.), Empirical research in translation and intercultural studies. Tbingen: Narr,
7988.
Sguinot, C. 1994. Translation and Advertising: Going Global. Cultural Functions of
Translation. Monographic of Current Issues in Language and Society 1(3): 24966.
Sguinot, C. 1996. Some thoughts about think-aloud protocols. Target 8(1): 7595.
Seleskovitch, D. 1978. Interpretation, a Psychological Approach to Translating.
Translation, Application and Research. New York: R. W. Brislin, 92116.
Seleskovitch, D. 1985. Interprter pour traduire. Paris: Didier.
Seoane, J. 1997. Internacionalizacin y localizacin. In: http://selva.dit.upm.es/~joaquin
/internacionalizacion/inter.html
Setton, R. 1997. A Pragmatic Theory of Simultaneous Interpretation. Doctoral thesis. The
Chinese University of Hong Kong.
Shi, Daoxuan. 1891. Gao seng zhuan er ji (Biographies of eminent monks: collection II).
Jingbei kejing chu.
Shi, Huiguan. 1984. Fahua zong yao xu (Preface to introduction to the Lotus sutra). In
Luo Xinzhang (ed.), 3233.
Shi, Huijiao. 1992. Gao seng zhuan (Biographies of eminent monks). Beijing: Zhonghhua
shuju.
Shi, Sengrui. 1984a. Xiaopin jing xu. In Luo Xinzhang (ed.), 34.
Shi, Sengrui. 1984b. Si yi jing xu. In Luo Xinzhang (ed.), 37.
 REFERENCES 285

Shi, Sengyou. 1995. Chu san zang ji ji (Collection of records concerning the tripitaka).
Beijing: Zhonghua shuju.
Shi, Sengzhao. 1984. Bailun xu (Preface to Satasatra). In Luo Xinzhang (ed.), 39.
Shi, Zanning. 1987. Song gao seng zhuan (Song Biographies of Eminent Monks). Beijing:
Zhonghua shuju.
Shipman, M. 1988. The Limitations of Social Research. 3rd ed. London, New York:
Longman.
Shlesinger, M. 1989. Simultaneous Interpretation as a factor in eecting shifts in the
position of texts on the oral-literate continuum. M. A. thesis. Tel-Aviv University.
Shlesinger, M. 1995. Shifts in Cohesion in Simultaneous Interpreting. The Translator
1(2): 193214.
Shreve, M. G. 1997. Cognition and the Evolution of Translation Competence. In J. H.
Danks et al. (eds), Cognitive Processes in Translation and Interpreting. Thousand
Oaks: Sage.
Snell-Hornby, M., Pchhacker, F. and Kaindl, K. (eds). 1994. Translation Studies: An
Interdiscipline. Amsterdam, Philadelphia: John Benjamins.
Sohr, A. 1996. Traduoes Fictcias de Ficao Cientca na Hungria. Boletim do
Citrat 4(1): 25.
Sohr, A. 1997. The Cultural Importation Process of Popular Genres: The Case of SF and
Fantasy in Hungary (19891995) from the point of view of Translation Studies.
Doctoral thesis. Katholicke Universiteit Leuven.
Soriano, M. 1977. Les Contes de Perrault; culture savante et traditions populaires. Paris:
Gallimard.
Stenzl, C. 1983. Groundwork towards a Comprehensive Model of Simultaneous Interpreta-
tion. M. A. thesis. University of London.
Sternberg, R. J. 1996. Cognitive Psychology. Fort Worth, etc.: Harcourt Brace.
Suleri, S. 1992. Woman skin deep: feminism and the postcolonial condition. Critical
Inquiry 18: 75859.
Tannen, D. 1980. Implications of the oral/literate continuum for cross-cultural communi-
cation. In J. Alatis (ed.), Current issues in bilingual education. GURT: Washington
D. C. Georgetown U. P., 32447.
Tannen, D. 1982. Oral and literate strategies in imaginative ction. In D. Tannen (ed.),
Spoken and written language. Ablex.
Tarone, E. 1984. Teaching strategic competence in the foreign language classroom. In
A. Wenden. 1991. Topic 8: An Action Plan for Strategies. Learner Strategies for
Learner Autonomy. New York: Prentice Hall International.
Techtmeier, B. 1984. Das Gesprch. Funktionen, Normen und Strukturen. Berlin: Aka-
demie-Verlag.
Televisi de Catalunya. 1997. Criteris lingstics sobre traducci i doblatge. Barcelona:
Edicions 62.
The IR(TI)N Bulletin, issues 1 to 16, 19891998.
The Random House Dictionary of the English Language, 2nd ed. 1987.
286 REFERENCES

Thelen, M. 1990. From interpretation to re-expression of meaning in the translation

process. In M. Thelen and B. Lewandowska-Tomascyck (eds), Translation and
Meaning, Part 1. Maastricht: Euroterm Maastricht.
Thomas, P. 1993. Choosing Headwords from LSP Collocations. In H. B. Sonneveld and
K. L. Loening (eds), Terminology: Applications in Interdisciplinary Communication.
Amsterdam, Philadelphia: John Benjamins.
Tirkkonen-Condit, S. 1989. Profesional Versus Non-professional Translation: A Think-
aloud Protocol Study. In C. Sguinot (ed.), The Translation Process. Toronto: H. G.
Publications.
Tirkkonen-Condit, S. 1992. The Interaction of World Knowledge and Linguistic
Knowledge in the Processes of Translation. A Think-Aloud Protocol Study. In B.
Lewandowska-Tomaszczyk (ed.), Translation and Meaning. Maastricht: Faculty of
Translation and Interpreting.
Tommola, J. (ed.). 1995. Topics in Interpreting Research. Turku: University of Turku.
Tommola, J. and Lakso, T. 1997. Source Text Segmentation, Speech Rate and Language
Direction: Eects on Trainee Simultaneous Interpreting. In K. Klaudy and J. Kohn
(eds), Transferre Necesse est. Proceedings of the 2nd International Conference on
Current Trends in Studies of Translation and Interpreting. Budapest: Scholastica.
186191.
Toury, G. 1980. In Search of a Theory of Translation. Tel Aviv: Porter Institute for Poetics
and Semiotics.
Toury, G. 1991. Experimentation in Translation Studies: Achievements, prospects and
some pitfalls. In S. Tirkkonen-Condit (ed.), Empirical research in Translation and
Intercultural Studies. Tbingen: Narr.
Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam, Philadelphia:
John Benjamins.
Trosborg, A. 1997. Text Typology and Translation. Amsterdam, Philadelphia: John
Benjamins.
Tytler, A. 1978. An Essay on the Principles of Translation. Amsterdam: John Benjamins.
Uren, E., Howard, R. and Perinotti, T. 1993. Software Internationalization and Localiza-
tion: An Introduction. New York: Van Nostrand Reinhold.
Valdelomar, R. 1998. Los publicistas consideran que la imagen de Espaa en el
extranjero debe basarse en los tpicos. ABC 3099.
Vallverd, F. 1987. Els problemes de la traducci. In Una approximaci a la literatura
catalana i universal. Barcelona: Fundaci Caixa de Pensions, 95107.
Van Dijk, T. A. (ed.). 1985. Handbook of Discourse Analysis. 4 vols. London, San Diego,
New York: Academic Press.
Van Dijk, T. A. 1980. Macrostructures: An Interdisciplinary Study of Global Structures in
Discourse, Interaction and Cognition. Hillsdale NJ: Lawrence Erlbaum.
Vzquez Ayora, G. 1977. Introduccin a la Traductologa. Georgetown DC: U. P.
Venuti, L. 1991. Genealogies of Translation Theory: Schleiermacher. TTR IV(2):
125150.
 REFERENCES 287

Venuti, L. 1995. The Translators Invisibility. A History of Translation. London, New

York: Routledge.
Vera Kirchner, J. 1992. Diccionario multilinge de especies marinas para el mundo.
Madrid: Ministerio de Agricultura, Pesca y Alimentacin.
Vermeer, H. J. 1983. Translation theory and linguistics. In P. Roinila, R. Orfanos and S.
Tirkkonen-Condit (eds), Nkkohtia knmisen tutkimuksesta. Joensuu: University
of Joensuu, 110.
Vermeer, H. 1996. A Skopos Theory of Translation: Some Arguments for and against.
Heidelberg: TextConText Verlag.
Vinay, J.-P. and Darbelnet, J. 1958. Stilistique compare du franais et de langlais. Paris:
Didier.
Walsh, F. 1996. Sin and Censorship. The Catholic Church and the Motion Picture Industry.
New Haven: Yale University Press.
Way, A. 1996 Localisation. In http://www.compapp.dcu.ie/~away/talk.html.
Willems, G. 1987. Communication strategies and their signicance in foreign language
teaching. System 15: 351364.
Wilmshurst, J. and Mackay, A. 1999. The Fundamentals of Advertising. Oxford: Butter-
worth Heinemann.
Wilss, W. 1982. The Science of Translation. Tbingen: Narr
Wilss, W. 1983. Translation strategy, translation method, and translation technique:
Towards a clarication of three translational concepts. Revue de Phontique
Aplique 66/68: 14352.
Wilss, W. 1988. Kognition und bersetzen: Zu Theorie und Praxis der menschlichen und
maschinellen bersetzung. Tbingen: Niemeyer.
Wilss. W. 1992. bersetzungsfertigkeit. Annherungen an einen komplexen bersetzungs-
praktischen Begri. Tbingen: Narr.
Wilss, W. 1996. Knowledge and Skills in Translators Behaviour. Amsterdam: John
Benjamins.
Wotjak, G. 1981. Tcnicas de traslacin. In M. Medina, L. Caballero and F. Martnez,
comps. Aspectos fundamentales de la teora de la traduccin. Trans. M. Carreo. La
Habana: Ed. Pueblo y Educacin, 197230.
Zabalbeascoa, P. 1993. Finding the proper place for descriptiveness and prescriptiveness
in translation. Sintagma 5: 4958.
Zhi, Qian 1984. Fa ju jing xu (The Preface to Fa ju jing (Dhamapada)). In Luo
Xinzhang (ed.), 22.
'@>R"D,>8@, E. K. (1988): E<4:4F<4R,F846 ">":42 4Fn">F8@?@ F<4N@<&@D>@?@
<,8F<". ;@F8&": %ZFT"b T8@:"
3.&,>8@, %. %. (1992): ?F>@&Z 8@>R,nR44 *,<,D<4>"R44 & n,D,&@*, (>" <"<,D4":,
4Fn">F8@-DJFF84N 4 DJFF8@-4Fn">F84N n,D,&@*@&). !&<@D,L. *@8<. *4FF.:
;@F8&".
9\&@&F8"b, ]. ). (1985): G,@D,<4R,F84, nD@$:,<Z n,D,&@*". ;@F8&": %ZFT"b
T8@:".
Q,D>&@, '. %. (1987): ?F>@&Z F4>ND@>>@?@ n,D,&@*". ;@F8&": %ZFT"" T8@:".
</TARGET "ref">

288 REFERENCES
<TARGET "ni" DOCINFO

AUTHOR ""

TITLE "Name Index"

SUBJECT "Benjamins Translation Library, Volume 32"

KEYWORDS ""

SIZE HEIGHT "220"

WIDTH "150"

VOFFSET "4">

Name Index

A H
Agar 204 Halliday 66, 262
Anzalda 37ff Hansen 17
Hartmann 17
B Hasan 262
Bassnett 168 Holz-Mnttri 224
Baumann 206 Hnig 129, 134
Beaugrande 18, 134ff Hurtado 130
Bourdieu 172
J
C Jensen 133
Cannon 238
Catford 14, 120 K
Chamberlain 39 Kaindl 22
Chesterman 225ff Kaminsky 39
Chomsky 18 Kiraly 120, 134
Cook 225 Krings 24, 132ff
Kuhn 17
D Kussmaul 129, 132ff
Danks 108
Darbelnet 117ff, 131 L
Delisle 131ff Lambert 175
De Mooij 223ff Lauer 206
Leech 67
E Lefevere 118, 168
Elena 132 Lrscher 108, 134
Lugones 38
F
Fusco 38 M
MacDonald 173
G Maier 41
Gpferich 206 Malmkjaer 17
Grice 225 Malone 131
</TARGET "ni">

290 NAME INDEX

Marcuse 173 S
Mason 121, 124, 132 Sager 266
Mondahl 133 Sguinot 132ff, 235
Short 67
N Shreve 23
Neubert 17, 23, 225ff Snell-Hornby 22
Newmark 119, 130ff Sohr 177
Nida 15
Nord 133ff, 224ff T
Techtmeier 206
O Toury 118, 226
Ortega y Gasset 172 Trosborg 225

P V
Pchhacker 22 Van Dijk 18
Vzquez Ayora 131
R Venuti 171
Rabadn 168 Vermeer 204
Rastall 134ff Vinay 117ff, 131
Reiss 130
Resch 235 W
Ritzer 226 Wilss 129ff, 133
Wotjak 131
<TARGET "si" DOCINFO
AUTHOR ""
TITLE "Subject Index"
SUBJECT "Benjamins Translation Library, Volume 32"
KEYWORDS ""
SIZE HEIGHT "220"
WIDTH "150"
VOFFSET "4">
Subject Index
A communicative competence 101
acquiring translation competence 103;
see teaching and training
action theory 29; see speech acts
adaptations 183
cultural adaptation 153; see also
domesticating
linguistic adaptation 155
adapted translations 174; see also
domesticating
addressees (of texts) 22, 196, 203
adequacy 129
adequate translation 196, 232
applied linguistics 16
aristocratic translation 171; see also
highbrow translation
autocensura 163
B diaculture 204
bitext 126
bitextual pairs 121
bitextual unit 122
bottom-up analysis 118; see also top-
down analysis
broad context 118ff; see also narrow
context
C deverbalization 123
censorship 174
censura 162; see also autocensura
competence 35, 100, 113, 224; see
also skills and translator
(knowledge) prole
extralinguistic competence 101
instrumental professional
competence 101
intercultural competence 226, 235
literary translator competence 56
psychophysiological competence
102
strategic competence 102
systemic competence 16
transfer competence 102
contrastive linguistics 16, 132
corpus of parallel texts 20, 203
corpus of texts 20
corpus studies 263
cultural identity 235ff
culture 204
cultureme 204
idioculture 204
paraculture 204
D
declarative knowledge 22, 102; see
also procedural knowledge
descriptive translation studies 224
designacin taxonmica 252ff
directionality 99
domesticating 130; see also
foreignizing
292 SUBJECT INDEX

E interdisciplinary paradigm 22
empirical experimental research 105 internationalization 248; see also
environmental validity 91 globalization, glocalization,
equivalence 4, 5, 123ff, 129 localization
communicative equivalence 31 interpreting research 77
functional equivalence 35, 122 intertextuality 201ff, 225, 235
lexical equivalence 122 intracultural communication 201
linguistic equivalence 29 intuition 133

F K
factory translation 171; see also kitsch 173
aristocratic translation
feminine subject translator 39 L
eld experiment 92 laboratory experiment 92
foreignizing 130, 159, 172; see also language pair 15
domesticating language studies 14
freedom 59 lexical density 66
function 129 linguistic approach 227, 232; see also
functionalist approach 195, 227, functionalist approach
232; see also linguistic literary translation 146
approach literary canons 142
literary translator 57
G localization 224, 243; see also
gender/text 41 globalization, glocalization,
general strategy 69; see also internationalization
translation method localization tools 247
genres 17, 200ff, 204
globalization 224, 236; see also M
glocalization, internationalization, matching the meaning 50; see also
localization, sophisticated translation
glocalization 224; see also McDonaldisation thesis 226; see also
globalization, internationalization, globalization
localization, measuring instrument 92
metacommunication 206ff
H method
highbrow translation 173; see translation method 117ff
aristocratic translation translation research method,
methodology 84, 88, 91
I interpreting research tools and
imitation 118 methods 80,
initiator 118 midcult 173; see also kitsch
intercultural communication 30, 201
 SUBJECT INDEX 293

N S
narrow context 118ff; see also broad shifts 117, 121; see also operations,
context procedures, replacements,
national identity 235ff techniques
norms 156, 226 simple translation 44; see also
initial norms 118 sophisticated translation
linguistic norms 142 skills 214ff; see also competence and
style norms 200 translator (knowledge) prole
textual norms 17 skopos 129
translation norms 33, 156 skopos-dependent translation 268
TT-norms 142 solution 68, 108, 121ff
solution-type 122ff
O sophisticated translation 44; see also
operations 117; see also procedures, simple translation
replacements, shifts, techniques speech acts 33
oral discourse 66 standardization 175
strategies 99, 108, 117, 120ff, 129ff
P
interpreting strategies 81
parallel texts 19
stylebook 157
patronage 118
phatic function 206
T
phraseological analysis 265
taxonomic naming 251ff; see also
phraseological units 261ff
designacin taxonmica
pragmatics 17
teaching 35, 134, 249
problem 68, 108, 121, 251ff
computer-assisted teaching
problem solving 132
programmes 94; see also training
procedural knowledge 22, 102; see
techniques 117, 121ff; see also
also declarative knowledge
operations, procedures,
procedures 117, 123; see also
replacements, shifts
operations, replacements, shifts,
addition 122, 125
techniques
compensation 125
process 99, 108, 118ff, 132
deletion 122
pseudotranslation 178
displacement 125
psycholinguistics 132
explicitation 125
Q omission 125
quality of translation 195 rearrangement 126
substitution 125
R transposition 121
receivers (of texts) 196 terminology 251ff, 261ff
reescritura 168 text production 197
replacement 117; see also operations, text segment 130ff; see also translation
procedures, shifts, techniques unit
replication 87 text types 17, 200ff
</TARGET "si">

294 SUBJECT INDEX

textual macrostructure 199 translation policies 227

textual paradigm 17 translation unit 121, 130
textuality 17 translator prole 230; see also
thematic progression 199 competence, skills and translator
theory 3, 25, 84ff knowledge prole
action theory 29 translator knowledge prole 20f; see
communicative theory 29 also competence, skills and
general theory of translation 33 translator prole
linguistic theory 29
think aloud protocols, TAPs 120, 133 V
ttulos 163ff variation
top-down analysis 119; see bottom-up terminological and phraseological
analysis variation 262
traduction intralinguale 181
training W
translator training 16, 109, 122, written discourse 66
215, 230; see also teaching written protocols 113
research training 86
In the BENJAMINS TRANSLATION LIBRARY the following titles have been published
thus far or are scheduled for publication:
1. SAGER, Juan C: Language Engineering and Translation: Consequences of automation,
1994.
2. SNELL-HORNBY, Mary, Franz POCHHACKER and Klaus KAINDL (eds): Transla-
tion Studies: An interdiscipline. Selected papers from the Translation Congress, Vienna,
912 September 1992. 1994.
3. LAMBERT, Sylvie and Barbara MOSER-MERCER (eds): Bridging the Gap: Empiri-
cal research on simultaneous interpretation. 1994.
4. TOURY, Gideon: Descriptive Translation Studies and beyond. 1995.
5. DOLLERUP, Cay and Annette LINDEGAARD (eds): Teaching Translation and Inter-
preting 2: Insights, aims, visions. Selected papers from the Second Language Interna-
tional Conference, Elsinore, 46 June 1993. 1994.
6. EDWARDS, Alicia Betsy: The Practice of Court Interpreting. 1995.
7. BEAUGRANDE, Robert de, Abdulla SHUNNAQ and Mohamed Helmy HELIEL
(eds): Language Discourse and Translation in the West and Middle East. Selected and
revised papers from the conference on Language and Translation, Irbid, Jordan 1992.
1994.
8. GILE, Daniel: Basic Concepts and Models for Interpreter and Translator Training.
1995.
9. REY, Alain: Essays on Terminology. 1995.
10. KUSSMAUL, Paul: Training the Translator. 1995.
11. VINAY, Jean Paul and Jean DARBELNET: Comparative Stylistics of French and
English: A methodology for Translation. 1995.
12. BERGENHOLTZ, Henning and Sven TARP: Manual of Specialised Lexicography: The
preparation of specialised dictionaries. 1995.
13. DELISLE, Jean and Judith WOODSWORTH (eds): Translators through History. 1995.
14. MELBY, Alan with Terry WARNER: The Possibility of Language. A discussion of the
nature of language, with implications for human and machine translation. 1995.
15. WILSS, Wolfram: Knowledge and Skills in Translator Behavior. 1996.
16. DOLLERUP, Cay and Vibeke APPEL: Teaching Translation and Interpreting 3. New
Horizons. Papers from the Third Language International Conference, Elsinore, Den-
mark 911 June 1995. 1996.
17. POYATOS, Fernando (ed.): Nonverbal Communication and Translation. New perspec-
tives and challenges in literature, interpretation and the media. 1997.
18. SOMERS, Harold (ed.): Terminology, LSP and Translation. Studies in language engi-
neering in honour of Juan C. Sager. 1996.
19. CARR, Silvana E., Roda P. ROBERTS, Aideen DUFOUR and Dini STEYN (eds): The
Critical Link: Interpreters in the Community. Papers from the 1st international confer-
ence on interpreting in legal, health and social service settings, Geneva Park, Canada,
14 June 1995. 1997.
20. SNELL-HORNBY, Mary, Zuzana JETTMAROV and Klaus KAINDL (eds): Trans-
lation as Intercultural Communication. Selected papers from the EST Congress
Prague 1995. 1997.
21. BUSH, Peter and Kirsten MALMKJR (eds): Rimbauds Rainbow. Literary transla-
tion in higher education. 1998.
22. CHESTERMAN, Andrew: Memes of Translation. The spread of ideas in translation
theory. 1997.
23. GAMBIER, Yves, Daniel GILE and Christopher TAYLOR (eds): Conference Inter-
preting: Current Trends in Research. Proceedings of the International Conference on
Interpreting: What do we know and how? 1997.
24. ORERO, Pilar and Juan C SAGER (eds): Translators on Translation. Giovanni Pontiero.
1997.
25. POLLARD, David E. (ed.): Translation and Creation. Readings of Western Literature
in Early modern China, 18401918. 1998.
26. TROSBORG, Anna (ed.): Text Typology and Translation. 1997.
27. BEYLARD-OZEROFF, Ann, Jana KRLOV and Barbara MOSER-MERCER (eds):
Translator Strategies and Creativity. Selected Papers from the 9th International Confer-
ence on Translation and Interpreting, Prague, September 1995. In honor of Jir Levi and
Anton Popovic. 1998.
28. SETTON, Robin: Simultaneous Interpretation. A cognitive-pragmatic analysis. 1999.
29. WILSS, Wolfram: Translation and Interpreting in the 20th Century. Focus on German.
1999.
30. DOLLERUP, Cay: Tales and Translation. The Grimm Tales from Pan-Germanic
narratives to shared international fairytales. 1999.
31. ROBERTS, Roda P., Silvana E. CARR, Diana ABRAHAM and Aideen DUFOUR
(eds.): The Critical Link 2: Interpreters in the Community. Papers from the Second
International Conference on Interpreting in legal, health and social service settings,
Vancouver, BC, Canada, 1923 May 1998. 2000.
32. BEEBY, Allison, Doris ENSINGER and Marisa PRESAS (eds.): Investigating Transla-
tion. Selected papers from the 4th International Congress on Translation, Barcelona,
1998. 2000.
33. In preparation.
34. In preparation.
35. In preparation.
36. SCHMID, Monika S.: Translating the Elusive. Marked word order and subjectivity in
English-German translation. 1999.
37. TIRKKONEN-CONDIT, Sonja and Riitta JSKELINEN (eds.): Tapping and Map-
ping the Processes of Translation and Interpreting. Outlooks on empirical research.
2000.
38. SCHFFNER, Christina and Beverly ADAB (eds.): Developing Translation Compe-
tence. 2000.
39. CHESTERMAN, Andrew, Natividad GALLARDO SAN SALVADOR and Yves
GAMBIER (eds.): Translation in Context. Selected papers from the EST Congress,
Granada 1998. 2000.
40. ENGLUND DIMITROVA, Birgitta and Kenneth HYLTENSTAM (eds.): Language
Processing and Simultaneous Interpreting. Interdisciplinary perspectives. 2000.