TEACHING TRANSLATION AND INTERPRETING 2

BENJAMINS TRANSLATION LIBARY

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 and post-graduate text books and readers in the English language.

ADVISORY B O A R D
Jens Allwood (Linguistics, University of Gothenburg)
Mona Baker (Universy of Birmingham and UMIST)
Morton Benson (Department of Slavic, University of Pennsylvania)
Marilyn Gaddis Rose (CRIT, Binghamton University)
Yves Gambier (Institute of Translation and Interpreting, Turku University)
Daniel Gile (ISIT, Paris)
Ulrich Heid (Computational Linguistics, University of Stuttgart)
W. John Hutchins (Library, University of East Anglia)
Jose 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, UMIST, Manchester)
Maria Julia Sainz (Law School, Universidad de la Republica, Montevideo)
Klaus Schubert (Technical Translation, Fachhochschule Flensburg)
Mary Snell-Hornby (School of Translation & Interpreting, University of Vienna)
Gideon Toury (M. Bernstein Chair of Translation Theory, Tel Aviv University)
Wolfram Wilss (Linguistics, Translation and Interpreting, University of Saarland)
Judith Woodsworth (FIT Committee for the History of Translation,
Concordia University, Montreal)
Sue Ellen Wright (Applied Linguistics, Kent State University)

Volume 5

Cay D o l l e r u p a n d A n n e t t e L i n d e g a a r d

Teaching Translation and Interpreting 2

Insights, A i m s , Visions
TEACHING TRANSLATION
AND INTERPRETING 2
INSIGHTS, AIMS, VISIONS
Papers from the Second Language International Conference
Elsinore, Denmark 4 - 6 June 1993

Edited by

CAY DOLLERUP
ANNETTE LINDEGAARD
University of Copenhagen

JOHN BENJAMINS PUBLISHING COMPANY

AMSTERDAM/PHILADELPHIA
 The paper used in this publication meets the minimum requirements of American
National Standard for Information Sciences — Permanence of Paper for Printed
Library Materials, ANSI Z39.48-1984.

Library of Congress Cataloging-in-Publication Data

Teaching translation and interpreting 2 : insights, aims, visions / [edited by] Cay Dol-
lerup, Annette Lindegaard.
p. cm. -- (Benjamins translation library, ISSN 0929-7316 ; v. 5)
Selection of papers presented at the 2nd Language International Conference which
was held June 1993, Elsinore, Denmark.
Includes bibliographical references and index.
1. Translating and interpreting--Study and teaching-Congresses. I. Dollerup, Cay. II.
Lindegaard, Annette. III. Language International Conference (2nd : 1993 : Helsing0r,
Denmark) IV. Title: Teaching translation and interpreting two. V. Series.
P306.5.T4 1994
418'.02'071--dc20 94-10141
ISBN 90 272 1601 0 (Eur.)/l-55619-682-2 (US) (alk. paper) CIP
® Copyright 1994 - 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 • 821 Bethlehem Pike • Philadelphia, PA 19118 • USA
 TABLE OF CONTENTS

Cay Dollerup and Annette Lindegaard:

Editors' foreword 3
Acknowledgements 8

LANGUAGE AND CULTURE IN COOPERATION

Christian Heynold:
Interpreting at the European Commission 11
Evaristus O. Anyaehie:
Language status and translation studies: a Nigerian perspective 19
Niranjan Mohanty:
Translation: a symbiosis of cultures 25

CULTURAL BARRIERS - TACKLING THE DIFFERENCES

Moses Nunyi Nintai:

Translating African literature from French into English 41
Manouchehr Haghighi:
Supra-lingual aspects of literary translation 47
Meta Grosman:
Cross-cultural awareness: focusing on otherness 51
Christiane Nord:
Translation as a process of linguistic and cultural adaptation 59
Heidrun Witte:
Translation as a means for a better understanding between cultures 69
Gabriele Becker.
Advertisements in translation training 77

TRANSLATION AND CLASS

Andrew Chesterman:
Karl Popper in the translation class 89
VI Teaching Translation and Interpreting 2

Sergio Viaggio:
Theory and professional development:
or admonishing translators to be good 97
Daniel Gile:
The process-oriented approach in translation training 107
Jeanne Dancette:
Comprehension in the translation process:
an analysis of think-aloud protocols 113
Cay Dollerup:
Systematic feedback in teaching translation 121
María Julia Saint.
Student-centred corrections of translations 133
Arnt Lykke Jakobsen:
Starting from the (other) end:
integrating translation and text production 143
Hasnah Ibrahim:
Translation assessment: a case for a spectral model 151
Alexis Nouss:
Translation and the two models of interpretation 157

INTERPRETING AND CLASS

Margareta Bowen:
Interpreting studies and the history of the profession 167
David Bowen:
Teaching and learning styles 175
Robin Setton:
Experiments in the application of discourse studies
to interpreter training 183
Bistra Alexieva:
On teaching note-taking in consecutive interpreting . 199
Viera Makarová:
Whose line is it anyway? or teaching improvisation in interpreting 207
Nancy Schweda Nicholson:
Training for refugee mental health interpreters 211
Leonor Zimman:
Intervention as a pedagogical problem in community interpreting 217
Sylvia Kalina:
Analyzing interpreters' performance: methods and problems 225
 Table of Contents vii

Franz Pöchhacker:

Quality assurance in simultaneous interpreting 233

SCREEN TRANSLATION

Irena Kovacic:
Relevance as a factor in subtitling reductions 245
Ian Roffe & David Thome:
Transcultural language transfer: subtitling from a minority language 253
Henrik Gottlieb:
Subtitling: people translating people 261
Yves Gambier:
Audio-visual communication: typological detour 275
TOOLS

M.K.C. Uwajeh:
Teaching linguists translation 287
Peter Baumgartner:
Technical translation: putting the right terms in the right context 295
Robert Clark:
Computer-assisted translation: the state of the art 301
Dieter Waltermann:
Machine translation systems in a translation curriculum 309

WORKS CITED

Works Cited 321

INDEX

Index 343
EDITORS' FOREWORD
Jagtpavillonen
 EDITORS' FOREWORD

The present volume comprises a selection of papers presented at the Second

Language International Conference: 'Teaching Translation and Interpreting:
Insights, Aims, Visions', which took place at Elsinore, Denmark, in June 1993.
The volume is dedicated to the teaching of translation and interpreting as an
activity which is discussed internationally. In editing the volume, we have chosen
to start out from insights into multi- and plurilingual settings, then proceed to
discussions of aims for practical work with students, and to end with visions of
future developments within translation for the mass media and the impact of
machine translation.
Multilingual settings, and the central role of translation, are discussed in the
three first articles.
The first of these presents the interpreting services and the training programme
at the European Commission. The interpreting services were established with the
European Community - now better known as the European Union - and they have
grown with it. The services at the Commission are now by far the largest of their
kind in the world. Looking back at the development of the interpreting services,
Christian Heynold deals with the Commission's experience and concludes with
a discussion of problems in managing interpreting with a multitude of languages.
The next article takes us to another continent where several languages exist
within one nation, but the perspective is different, for in multilingual Nigeria
described by Evaristus O. Anyaehie, it is a foreign language, English, that serves
as the official language throughout the nation. Against this background, Anyaehie
outlines a translation programme preparing future professionals for dealing with
many working languages among their users. In the last article in this section
Niranjan Mohanty from yet another country with numerous languages, India,
stresses that translation is in itself a symbiosis of cultures and argues that for a
fusion to take place, translators must be cognizant of numerous societal para­
meters in both source and target languages if they are to bridge the cultural gaps.
The next contributions focus on linguistic barriers and how they may be over­
come. Moses Nunyi Nintai from Cameroon deals with problems in translating
Francophone African literature which mirroring African realities, and proposes
a number of approaches for teaching students consciousness of the complexities
of their task. Manouchehr Haghighi (Iran) argues that the Renaissance constitutes
the major cultural divide between the West and other parts of the world and that
4 Teaching Translation and Interpreting 2

an awareness of the impact of the Renaissance on European thinking is a

prerequisite for successfully teaching translation from European languages. Meta
Grosman (Slovenia) suggests that an understanding of the otherness of cultures
is required by would-be translators if their perceptions of texts for translation are
to be appropriate. Christiane Nord (Germany and Austria) discusses adaptation
and translation from a functional point of view and argues that an all-out
functional approach should be used for translation teaching. The point of
departure for Heidrun Witte (Spain) is also functional, but her line of enquiry is
the actual translational commission. She posits that translators must be taught to
be above naive intercultural communication and that they must sometimes take
upon them the job of enlightening clients. And finally, Gabrielle Becher (Spain)
discusses the use of advertisements for introducing students to other cultures as
well as teaching them factors to be taken into account in culture-oriented
translation,
The subsequent articles deal with principles and specifics of translation teaching
- and the quality of it. Andrew Chesterman (Finland) sets the scene by discussing
translation processes - and translation learning - in terms of the philosopher Karl
Popper's schema for the scientific method with its never-ending spiralling
towards unattainable perfection. Chesterman is followed up by Sergio Viaggio
(UN), who has translation tasks and approaches reflect on advances in translation
theory. Daniel Gile (France) deals with the usefulness of the process-oriented
approach in training programmes and points out that, among other things, it leads
to a considerable reduction in student errors due to faulty analysis of texts.
Jeanne Dancette (Canada) presents readers with preliminary findings from think-
aloud protocols of student translations of a text concerned with economics, and
concludes that although successful translation may well take place without a
major conceptual framework, it does presuppose that the correct textual and
extratextual linkages and inferences are made. Cay Dollerup (Denmark) discusses
some procedures for systematic feedback to translation students which combine
individual and collective assessments. Marfa Julia Sainz (Uruguay) elaborates on
this point from other angles, including references to student rights and explicit
information to students about the underlying 'ideologies' and taxonomies. In his
contribution, Arnt Lykke Jakobsen (Denmark) suggests that it may be more
beneficial for translation teaching to focus on the end product, notably in terms
of tasks which are, in his terminology, 'warm' (authentic), rather than on
translation of texts outside a situational context. The section is closed by Alexis
Nouss (Canada) who discusses two dominant theories of interpretation of texts,
and concludes that the validity of an interpretation is ultimately dependent on the
community's acceptance of it.
 Cay Dollerup and Annette Lindegaard, Denmark 5

The volume then concentrates on the teaching of interpreting. Holding the view
that a historical perspective will enhance the understanding of the profession both
among trainees and full-fledged practictioners, Margareta Bowen (USA) deals
with the history of interpreting. David Bowen (USA) discusses the merits and
disadvantages of various trends of interpreting teaching, also touching on students
and student backgrounds. Robin Setton (Republic of China) presents pedagogics
from a non-Indoeuropean language perspective and deals with considerations
underlying the progression in the interpreter programme in Taiwan. Referring to
results and observations from her own classes, Bistra Alexieva (Bulgaria) dis­
cusses the pros and cons of teaching note-taking techniques in interpreting and
concludes that this is useful when combined with analytic training. Using per­
sonal experience from professional work, Viera Makarová (Slovakia and United
Kingdom) details how she teaches students to manage under poor working con­
ditions. Nancy Schweda Nicholson (USA) describes the teaching of interpreting
to trainees whose mother tongues are unknown to the teacher. And Leonor
Zimman (United Kingdom) discusses a number of methods for making would-be
community interpreters face the dilemma of intervention vs non-intervention in
community interpreting situations. Sylvia Kalina (Germany) discusses the prob­
lems involved in tackling authentic interpreting material for the purpose of re­
search, and hence for the improvement of teaching. And Franz Pochhacker (Aus­
tria) attempts to establish criteria for assessing quality in simultaneous interpret­
ing, touching upon a large number of parameters whose pertinence is evaluated.
Then follows a number of articles dealing with screen translation, a fairly new
field in translatology. Irena Kovačič (Slovenia) argues that relevance theory
provides an explanatory framework for reductions in subtitling, but stresses that
mere familiarity with the theory itself will not make for perfect subtitling. Ian
Roffe and David Thorne (United Kingdom) describe the societal and linguistic
background for the subtitling programme at the University of Wales and present
a series of transfer problems used for teaching purposes. In the next contribution
Henrik Gottlieb (Denmark) discusses courses for subtitlers as well as the para­
meters which must be taken into account for an assessment of quality from a pro­
fessional point of view. This section is rounded off by Yves Gambier (Finland)
who deals with the larger context of audiovisual translation and suggests the
establishment of an international course in language transfer in the media.
The last contributions focus on translation and tools. In the first article, M. K.
C. Uwajeh (Nigeria) emphasises the usefulness of translational activity as a tool
for linguistic insight into the workings of the source languages, and the potentials
for teaching translation at the same time. Peter Baumgartner (Germany) describes
the methods used in translation training at the Flensburg Polytechnic for ensuring
6 Teaching Translation and Interpreting 2

the proper technical terminology, including, for instance, parallel texts and break­
down into textual components.
The last two articles concentrate on machine translation. Robert Clark (United
Kingdom) suggests that most prejudices against machine translation do not hold
good. He stresses the importance of new technologies and discusses a number of
available translation programmes in terms of their usefulness for the professional,
arguing that today it is necessary for trainee translators to be familiarised with
machine translation. The final word is left to Dieter Waltermann (USA), who
describes the machine translation training programme at Carnegie Mellon
University. He presents the course and its components and provides a detailed
discussion of the integration of the Translator's Workstation in the course.
All told, then, the volume reflects an ongoing debate about the role of teaching
translation and interpreting at the end of the twentieth century.
This discussion is international. At first glance, it may therefore seem as if we
are Eurocentric in choosing the European Union (EU) as the point of departure.
But this is a deliberate choice, indeed a statement: There are good reasons for
focusing on the EU in a book which reaches out to language professionals all
over the globe:
We suggest that the formation of the EU has been the strongest single factor
behind the present professionalisation and internationalisation of translation work,
giving translation studies new vitality and creating the keen interest in teaching
translation and interpreting, and - possibly as a ripple effect - also subtitling and
other types of interlingual transfer evidenced in this and similar books.
The EU is admittedly controversial in politics and is often considered bureau­
cratic. On the other hand, the decisions reached in democratic agreement among
the EU's twelve member states affect the everyday life of more than 300 million
people: the EU represents by far the most intimate multinational peacetime (and
peaceful) cooperation in human history.
To language workers, the EU is important in professional contexts, for its size,
for its emphasis on quality (however elusive) in language work, and for its
pivotal role in forcing language professionals, such as interpreters and translators
to discuss their work. These statements need some explanation for the uninitiated
reader:
Operating with nine official languages, the EU has the most extensive language
services of any major organisation in the world, and, serving the central needs
for communication between the member states, they form the backbone of the
cooperation.
This language-based polical cooperation has created a need for professionalism
in language work. This need for professionals is eminently obvious to the deci-
 Cay Dollerup and Annette Lindegaard, Denmark 7

sion-makers who use the language products every day at EU meetings and the
EU institutions therefore employ thousands of language professionals. Their very
number creates needs for new translators and improvement in quality. Their pub­
lic visibility has become very high. Unlike the language workers outside institu­
tionalised settings, these highly-paid professionals are also perforce obliged to
discuss concrete translation work with in-house experts and delegates, that is with
clients and users. In terms of organisation, they often work in teams and therefore
discuss translational products with their colleagues and departmental 'revisors':
translation and interpreting are no longer the exclusive provinces of gifted
individuals in ivory towers with little or no societal input, working on a person-
to-person basis for power-brokers, making up personal files of proper translations
for specific clients, or polishing literary masterpieces.
In tangible financial and political ways, EU has set the scene for contemporary
international discussions on translation. It has also, we suggest, a paradigmatic
character for future international cooperation in terms of the practice, manage­
ment and training of language professionals, not only in Western Europe, but all
over the world. The diffusion of that experience among teachers and researchers,
general discussions of the situation in other parts of the world, and presentation
of practical problems and theoretical considerations of central concern to teaching
will further interlingual and cross-cultural cooperation.

Cay Dollerup and Annette Lindegaard

 ACKNOWLEDGEMENTS
The organisers of the conference wish to thank The Danish Research Council for the Human­
ities for support tendered towards the plenaries.
We also wish to record our gratitude to the 'International Office' (The Danish Ministry for
Education) and to 'The Danish Democracy Fund' (The Danish Foreign Ministry) for grants to
scholars who could not otherwise have attended the conference.
We are grateful to the Department of English (University of Copenhagen) for financial help
for the final work on the present book.
Many people helped us edit it. It is unfair to single out any, but nevertheless we wish to ex­
press our deepest thanks to Mr Henrik Gottlieb for casting an ever vigilant eye on our efforts, to
Ms Marion Fewell, to Dr Viggo Hj0rnager Pedersen, and to Mr Robert Storace.

The pictures from Denmark which grace this volume are reproduced with the kind permission
of Mr Poul Andersen and Politikens Forlag.
LANGUAGE AND CULTURE
IN COOPERATION
HELINGIORS.HAVN
See fra don nordre Side.
 INTERPRETING AT THE EUROPEAN COMMISSION

Christian Heynold, European Commission

I represent an international organisation which uses, so to speak, the product

of teaching and training of interpreting, that is, fully fledged conference inter­
preters, and my article deals with interpreting, oral communication, at the Euro­
pean Commission. Translation is written communication, and is handled by an
altogether different Directorate General at the Commission.
The three elements of the subtitle of this volume, "Insights, Aims and Visions",
serve as signposts for my discussion.

History
First, then, some insights: our Service is the largest conference and interpreting
service in the world - it is a Joint Service, so it caters for several EC institutions
not just the Council of Ministers and the Commission, and renders its services
at meetings globally, not only in countries in the European Union.
Our service has grown from a small group of interpreters back in the 1950's,
to some 400 permanent staff now. It started with the Coal and Steel Community
where the President and founding father of the European Community - now better
known as the European Union - Frenchman Jean Monnet, worked with a German
Vice President, Franz Etzel. They had no common language. So right from the
start, there was a genuine need for interpreting.
In 1958, the first President of the Commission, W. Hallstein asked Mrs. van
Hoof-Haferkamp to set up an interpreting service with 11 interpreters.
At that time, we had 6 member states and four languages: French, German,
Dutch and Italian. The Treaties establishing the European Economic Communities
and EURATOM were drawn up in those four languages in texts which were e-
qually binding. At the very beginning in 1958, the Council of Ministers adopted
a regulation (regulation N°l) stating that all four languages should be official and
working languages.
In 1973, Denmark, the United Kingdom and Ireland joined the Union, adding
Danish and English, so that there were now 6 working languages.
In 1981 Greece gained entry, and then there were seven. The newest members,
Spain and Portugal brought the number up to the present nine working languages
for the 12 member countries in 1986.
It requires 27 interpreters with a total of 72 language combinations to cover
12 Teaching Translation and Interpreting 2

nine active and passive languages at a meeting.

We organise some 50 meetings a day now inside and outside Brussels, but not
all, I hasten to add, with 9 languages and 27 interpreters.

Languages and organisation

So far the number of languages has tended to increase with new member states.
Since this ties up with democracy and the freedom to express oneself in one's
own language, outsiders easily get the impression that this process can go on
indefintely. This, however, is not the case. Rather we must find out how we can
combine democracy (defined as the right of expression) with efficiency in terms
of the smooth running of international meetings.
Referring to the subtitle again - one of our aims is to be efficient and prag­
matic. We want to provide the high-quality interpreting which is our hallmark,
not just to provide interpreters for the sake of it.
In this context, it is worth seeing how another international organisation has
tackled the language problem. With 180 member states, the UN manages fine
with 6 languages. It dealt with the language problem quite early. In the 1970's,
when it passed from 5 to 6 languages, it commissioned a report. This paper, the
King report "on the implications of additional languages in the UN", states:
Such an addition, because of the non availability of staff handling all the language combinations
involved, could be achieved only at the expense of quality (e.g. as a result of the general use
of the relay method of interpretation ...) and the inspectors venture to doubt whether member
states would willingly accept any general lowering of standards. (p. 44, §144)
The UN, of course, is different from the European Union, for, unlike the Union,
the UN is not passing binding laws. Directly or indirectly, European Union laws,
regulations and directives affect all citizens in the member states and must
therefore necessarily be translated into all nine languages.
Here, however, we are concerned with interpreting. It goes without saying that
civil servants and experts who prepare directives and regulations must understand
and be understood in their meetings. This means that if, say, a Greek delegate
does not master any other language, there will be interpreting into Greek. If the
delegate can follow the discussion in another language but but does not master
it actively, his needs are covered as follows: there is no interpreting into Greek,
but interpreting from Greek into one or of the more generally understood lan­
guages such as English, French and German. We call this an asymmetric inter­
preting system.
This example is a key to our work: we cover genuine needs for most meetings,
especially at the Commission.
It is different in the Council of Ministers, where the 12 member states are
always represented. At the ministerial level we always have to provide inter-
 Christian Heynold, European Commission 13

preting from all the nine into all nine languages since all participants demand the
right to speak and to understand discussions in their own, officially recognised
language. Even there, the situation is more relaxed at times. Ministers want to
have more direct contact, for instance at lunch and dinner, so those who can,
switch to other languages. On these occasions there are no complaints about loss
of rights and identity. Problems are often solved at informal gatherings, where
protocol is waived and Ministers are placed according to language abilities with
a minimum of interpreters around. In fact, we do have dining rooms equipped
with booths in Brussels and Luxembourg. Even when they are used, the number
of languages involved is reduced, for nine booths around the table are not con­
ducive to a relaxed and informal atmosphere! Helmut Schmidt and Valery
Giscard d'Estaing enjoyed a special relationship, not least because they communi­
cated easily in English and they always sat together. Conversely there must
always be an interpreter present when Helmut Kohl and John Major meet.
All told, most nationalities generally prefer direct contact, no matter how good
the interpreters. Interpreting should be used only when needed. If it can be a-
voided, it should be avoided. This is a matter of common sense: a poor inter­
preter is worse than no interpreter, for the simple reason that poor interpreters
may make people believe that their rendition is adequate. Interpreters must be ex­
cellent in order to fulfill their difficult task. It is also important that interpreters
are motivated - just like all other professionals - and feel that they are needed.
It is demoralising to sit in a booth, redundant, because delegates asked for inter­
preting as a matter of principle but do not use the product.

Other considerations
There is also a financial side to it. As President Delors put it, "We pay a high
price for Europe." Interpreting is expensive and should therefore be used pro­
perly. The cost of operating the interpreting of one meeting from and into all
nine language is about 15,000 ECU a day.
So, all told, there are several reasons why the Union must look ahead and ask
if it can go on adding more languages.
This is a political question, and I am no politician but an official who must
follow political instructions. Our service reports directly to President Delors. All
we can do is draw attention to cost and feasibility if we continue to expand the
services.
In this context, it is pertinent to point out that it has always been a problem to
find qualified candidates for the profession. In terms of human resources, we are
also faced with a shortage. I mentioned that we had around 400 interpreters on
the permanent staff. Yet we must take in free-lancers all the time, for on the a-
14 Teaching Translation and Interpreting 2

verage we have about 750 interpreters working in 50 meetings, which means that
some 45% of the interpreters employed at any given time are free-lancers.

The in-house training programme: the 'stage'

Background and impact
Good interpreters do not grow on the trees. We discovered this about 25 years
ago at the eve of the first enlargement of the EC. No university could offer us,
for example, qualified interpreters with Danish, so we decided to set up our own
postgraduate programme.
This is a six-month in-house training course, a 'stage' in our terminology. The
trainees are university graduates in any university discipline with a solid lin­
guistic background, the ability to absorb and analyse an argument and a talent for
communication. By now 50% of our staff have been trained by us. However, we
are not a school. We started our training scheme out of necessity, and we are not
bound by any treaty to continue to do so forever.
Thanks to this programme we have accumulated a certain know-how over the
years, and we are ready to offer our experience to others, inside and outside the
European Union.
Thus, we have not only trained interpreters for the EC-languages but also - in
the framework of cooperation agreements - for the Republic of China, Hong
Kong, Turkey, and for Central and Eastern Europe within the TEMPUS pro­
gramme. We are passing on our know-how and experience to the Czech Re­
public, Poland, Hungary, Albania and recently also Vietnam.
Let us take the Czech example: so far 6 teacher interpreters have come to our
service for study visits of 1 to 3 months to sit-in on training sessions and tests
in order to observe our methods. This will enable them to set up postgraduate in­
terpreting courses at home. The main point is that our programme should have
a ripple effect.

Selection for the 'stage'

Our programme starts with a selection or, as we call it, an 'aptitude test'. We
are in contact with university career advisers all over Europe to find suitable can­
didates. Interested parties apply - we have processed more than 30,000 applica­
tions in the last 30 years, held over 450 aptitude tests, and eventually recruited
a total of 400 of the applicants.
The selection procedure is highly competitive and rightly so.
First, candidates must have the right qualifications, and less than 20 per cent
of the applicants foot that bill. We are looking for a quick mind, a talent for
communication and for the ability to speak fluently and naturally in the mother
 Christian Heynold, European Commission 15

tongue. In addition to the mother tongue, successful applicants must have a solid
knowledge of at least 3 more European Union languages. That may sound cynical
- but with only 6 months training time, we must be demanding on language
knowledge.
At the aptitude test, the panel is made up of senior interpreter trainers. The can­
didates are convened in groups of about ten. We give them short, uncomplicated
speeches with a line of reasoning or with logical series of points which can be
memorised easily without taking notes. Candidates must then render these
speeches into their mother tongue. Of course, alert candidates will note advice
and criticism directed both at themselves and at the others. On the basis of a few
hours including general questions, the panel will have a fairly good idea of who
has the knack for the profession. It is inevitable that even more will fall by the
wayside in the course of the training because they find it too demanding and too
stressful. But the job is' stressful and some cannot cope.

Contents
The training programme, then, lasts only six months. Admittedly, that is a very
short time for learning the skills. Yet we believe, and the facts bear us out, that
six months are enough for candidates with the aptitude, the background know­
ledge, the flair and the will.
The course itself has three parts.
Roughly speaking the students are organised in working groups, each of them
headed by an experienced in-house interpreter trainer who supervises the group
and does classes about twice a week. In addition, one day is set aside for a
session where all trainees work with a large panel of experienced interpreters.
The first part coincides with the first month and consists of "memory exer­
cises", that is, listening, concentration and analysis in order that the trainees learn
to follow a logical argument, see the speech in blocks, as I describe it, and re­
constitute it by retaining the essentials of the argument in consecutive renditions.
Although some students are uneasy about it, they must not take notes, for it is
necessary to make them grasp and retain ideas, not words. When they get the
point of a lecture, they can then give a natural account of it in their own words.
After the first month we introduce note-taking and with it a new set of head­
aches, namely a loss of the listening ability students have acquired, and all too
much attention to scribbling and deciphering of notes. We do not teach a par­
ticular note-taking system. We prefer to give guidelines and let trainees develop
individual systems. As experienced interpreters we can guide, advise and correct
but the rest is up to the individual.
At the end of the second month, the first eliminatory test is held.
16 Teaching Translation and Interpreting 2

We then return to the basic guiding principles: listening, analysis and delivery
as we introduce simultaneous interpreting. After another two months of training
in simultaneous and (continued) consecutive interpreting, comes another elimi-
natory test. And, at the end of the course, after 6 months, is the final recruitment
test. At this point, trainees must perform well enough for us to say: "I can work
with this colleague at a meeting."
Given our aim of functioning efficiently, you will appreciate that the learning
process has barely begun. New interpreters will learn from senior colleagues eve­
ry day. It is rather like passing a driving test in the country and then learning to
manage in rush hour traffic in a capital city.

Comments
The interpreters trained this way have to work for us for at least two years -that
is what we ask in return for the training. They are free to become civil servants
in our service if they pass an "open competition" ('concours'); or they may go
on to other jobs after the two-year term: Somebody with a wide academic back­
ground has better chances of profiting from professional mobility than somebody
familiar with only one field.
Our 'stage' is based in Brussels where we have a large corps of experienced
staff interpreters.
The unique nature of the Brussels course is that the training is provided on the
spot by the future employer and that trainees are taught in their future working
environment. They sit in on meetings during the last 2 months of the course to
practice in dummy booths (disconnected booths) in the presence of their trainers.
Within the EU, we have also run our course or part of it in member states. For
example, we have had 'stages' in the European University Institute in Florence,
Italy, where we provided the interpreter-trainers and the students benefited from
the multilingual environment with students and staff at the Institute as speakers.
The results were mutually beneficial, for while the trainees got authentic lectures
for excercises, the speakers learned how to deliver speeches for interpreting,
which was important since many would later participate in multilingual meetings.
There have also been "stages" in Lisbon, Portugal, and at the Diplomatic
School in Madrid, Spain.
In short: we are open to and welcome anyone who wishes to see how we train
interpreters and compare our experience with their own.

Visions of the future

What about the visions of tomorrow's world as far as Europe's languages are
concerned.
 Christian Heynold, Commission of the European Community 17

concerned.
I already outlined some problems facing us and the realities we must cope with.
And although there is little that can be said with certainty, there are still some
indications as to what the future may bring.
These indications include
1) the conclusions of the European Council in Lisbon of June 1992. It says
specifically: "To ensure effective communication in meetings, pragmatic solutions
will have to be found by each of the institutions."
2) the fact that so far negotiations with states applying for membership are con­
ducted in English. And
3) the fact that any solution must take into account financial implications. Let
me exemplify by using the Scandinavian languages. If Norway and Sweden
become members of the Community on a par with Denmark, there is no doubt
that Norwegian and Swedish will become official languages and that all legal
documents must be translated into these languages. But in interpreting the situ­
ation is different: many Scandinavians do understand one anothers' languages
and may well prefer to have one Scandinavian working language in interpret­
ing. I refer to the interview with the Swedish Prime Minister Carl Bildt (B0rsen
28 May 1993). He is in favour of just one Scandinavian working language in
the EC. In the "Nordic Council" the Scandinavians work together very effici­
ently without interpreters. And the Danish linguist Allan Karker prophesies that
the common Scandinavian linguistic heritage will come out strengthened in
European Community cooperation (1993:75). A pan-Scandinavian booth would
be a relatively cheap solution for the Commission.
However this may be, the moment of truth will come, probably at the next inter­
governmental conference scheduled for 1996. It will comprise a review of the
European Institutions, and discuss European opening to Central and Eastern
Europe, including the management of a Community of 15, 20 or 25 members.
For our service a Community with 15 languages is a 'nightmare scenario'. This
would imply 210 language combinations as opposed to the present 72; this re­
quires at least 45 interpreters in 15 booths at each meeting. No matter how com­
petent the interpreters are, this would involve a multiplication of relay-interpret­
ing, where interpreters must interpret the renditions of interpreter colleagues
instead of the original because they do not understand the source language. This
will pose a quality problem, and, as you recall, quality is one of our prime con­
cerns. It also involves managerial problems, not to mention the architectural one
of fitting so many booths into a conference room.
Facing this scenario we must stop playing the ostrich burying its head in the
sand hoping that the problem will go away. It will not. It will take courage, polit-
18 Teaching Translation and Interpreting 2

ical will, and political agreement to reach a pragmatic solution. Tentative solu­
tions are, for instance, the introduction of
1. one single or a limited number of working languages, or,
2. a central relay language, that is, a system where all speeches were rendered
only into one language, for instance English, and then interpreted into all the
other official languages.
With the introduction of the latter system, English could play the same pivotal
role as Russian did in the former Soviet Union where all speeches in executive
bodies could be delivered in the delegates' own languages, to be interpreted di­
rectly into Russian, and from there interpreted into all the other Soviet working
languages.
Alternatively, an artificial man-made language like Esperanto could be intro-
duced, or even - why not - we might return to Latin or Greek. It sounds far-
fetched, perhaps, but anything is possible.
The problem is there, and the central challenge is to combine democracy with
efficiency.
 LANGUAGE STATUS AND TRANSLATION STUDIES:
A NIGERIAN PERSPECTIVE

Evaristus O. Anyaehie, Abia State University, Uturu, Nigeria

Nigerian language policy

The core of the Nigerian language policy has not changed since independence
in 1960. English has continued to be the first official language. Hausa, Igbo and
Yoruba, dominant languages of the former northern, eastern and western regions
of Nigeria, have continued to be federal official languages. Today, with the cre­
ation of 30 local states, these languages, spoken by about 50 million Nigerians,
are still used for federal official and media broadcasts. The rest of the 250 lan­
guages, spoken by the other half of the population, have state or local govern­
ment status depending on the number of languages in the state or local govern­
ment area and their relative demographic importance.
This language policy has its corollary in the federal education policy which
stipulates that pupils at the elementary and secondary schools learn one of the
three major vernacular languages that is not their mother tongue. This policy is
based on the theory that mother tongue education, particularly at the early stages
of the child's development, guarantees better conceptualisation of phenomena and
ensures a more solid base for nurturing the technological culture needed for na­
tional development. But where the child's mother tongue does not have a privi­
leged state or local government status, this policy fails to attain its primary
objective and the child may be constrained to speak English and as many as three
indigenous languages.
The language grading system in Nigerian translation schools, manifestly based
on generally accepted psychological and environmental theories of language ac­
quisition and learning, does not quite reflect this national policy. Educated
Nigerian English, distinct from the many forms of pidgin or 'broken' English, has
understandably continued to assume the status of 'A' language (L 1 . The first
foreign language is considered as 'B' language (L2). The mother tongue is classed
' C (L3) without any distinction.
In this article, we will take a cursory look at the shaping of the Nigerian lan­
guage policy since independence, the resulting shifts in status of the languages
spoken in Nigeria and their impact on translation practice and studies. We will
also review the A-B-C language grading system within the psychological and en­
vironmental theoretical framework on which it seems to be based and assess its
20 Teaching Translation and Interpreting 2

application to the individual Nigerian's bilingual experience. From these obser­

vations and reflections, we will attempt to make proposals for the planning of
translation studies in a multilingual culture.

Language status and national policy interpretations

In the spirit of the budding Organisation of African Unity, the 1962 Addis-
Ababa conference of African ministers of education considered it desirable to
make every African speak English and French fluently and thus enhance the de­
velopment of one large African family without language barriers and, presum­
ably, without the intervention of translators and interpreters (UNESCO 1962). So,
energetic efforts were made in Nigeria to step up the teaching of French. Since
every Nigerian was expected to speak French and since the vernacular languages
had little or no social status, professional translation involving European as well
as local languages was not taken seriously.
But, by 1976, when another generation of African ministers of education met
in Lagos to the beat of the Festival of African Culture, the heat of the intra-Afri­
can English-French bilingualism had almost gone (Damida 1977). The new focus
was on "democratisation", "cultural identity" and "development". In Nigeria, new
educational provisions were made to promote the teaching of vernacular lan­
guages in primary and secondary schools. Universities created or upgraded privil­
eged linguistics and vernacular language departments that soon churned out local
language graduates. Vernacular language experts under the umbrella of the Lin­
guistics Association of Nigeria coined vernacular metalanguage so as to avoid the
use of English for teaching indigenous languages. There was a corresponding up­
surge in the production of vernacular language reading texts, plays and films.
This shift in status of vernacular and foreign languages created the first real need
for an elite group of professionals who would facilitate trans-national communi­
cation in English and French.
This need was eloquently expressed in the 1982 Dumbleton report on Language
services in West Africa (Dumbleton 1982) which focused on translation and con­
ference interpreting services in English and French for the Economic Community
of West African States. Inaugurated in 1975, this community now has an esti­
mated 200 million inhabitants living in 16 sovereign nations. The Dumbleton re­
port stressed the need for training and upgrading of language professionals to
serve some 60 officially recognised international or intergovernmental organis­
ations including member states. There were for instance only 40 interpreters in
the sub-region. Many of the international organisations located in the sub-region
also had their translations done elsewhere.
Two pertinent remarks in Dumbleton's immensely realistic report deserve men-
 Evaristus 0. Anyaehie, Nigeria 21

tion. The first is an observation that "students entering translation schools at the
pre-university level do not have sufficient knowledge of their foreign language
or even of their mother tongue or vehicular language" (Dumbleton 1982: 11). The
second is a recommendation that translators and interpreters be trained at the
postgraduate level "to work into their first language from two foreign languages"
(Dumbleton 1982: 12). The paradox in these two remarks is that a proposal pri­
marily concerned with improving transnational communication in English and
French should recommend the use of vernacular working languages in translation
studies programmes. Dumbleton's views reflected awareness of the prevailing
situation in Africa's multilingual cultures. They could also be explained from the
perspective of the individual bilingual experience, using the theoretical frame­
work on which the A-B-C grading system seems to be based.

Language status and individual bilingual experience

Piaget's claim that language acquisition in the monolingual child is tied to
sequenced cognitive development (Inhelder and Piaget 1966) is generally ac­
cepted. Also accepted are Chomsky's (1965) and Lenneberg's (1971) claims that
beyond some age limit (between 11 and 14 years) no language can be learned in
the same sequenced and dialectic manner characteristic of first language acqui­
sition. These two claims largely constitute the ontological basis for the unques­
tioned distinction made between "acquiring" a first language and "learning" sub­
sequent languages.
Being the result of an acquisition process, the educated native speaker's lan­
guage competence has generally been adopted as the model of 'perfect' mastery
of a language towards which the second language learner should strive. One
weighty argument advanced for this attitude is that language acquisition is
marked by an "emotional attachment" to the various cognitive developmental
stages. It is with the acquired language that one would most spontaneously
"think", "feel" and even "dream". This language would be the most likely to be
used when one is highly stressed or is obliged to mobilise all cognitive capacities
for full perception and interpretation of phenomena. One such instance could be
the interpreting of highly condensed discourse for which Seleskovitch (1981: 41)
recommends the mother tongue as 'A' language into which interpreting should
be done. Dumbleton's report seems to adopt this view.
Theoretically, based on the above and on the assumption that monolingual and
bilingual developments relate to the mind in the same manner, a Nigerian who
begins learning English as a second language at about the same time as the
mother tongue would probably apply his sequenced cognitive capacities equally
to the mastery of the two languages. Provided that the languages have similar ex-
22 Teaching Translation and Interpreting 2

posures to family, school, work and other environments, the educated Nigerian
would develop the same emotional attachment to the two languages and could
pass for a perfect bilingual as defined both by Ervin and Osgood (1954) and
Lambert (1958). If this purely hypothetical scenario could be used to explain the
'A' grading of educated Nigerian English, it should equally be used to categorise
as 'A' an acquired vernacular language used in about the same contexts as
English.
Past the threshold age bracket suggested by Chomsky and Lenneberg, environ­
mental exposure would seem to determine the mastery of any other language to
a level functionally equivalent to 'acquired' language competence. This would
explain the attribution of the etiquette 'A' to the "preferred" language of an adult
immigrant whose mother tongue is generally less solicited in a foreign land and
could therefore go for a 'B' status. This environmental consideration does not ex­
plain the 'B' grading of the second European language - French - placed before
any of the Nigerian vernacular mother tongues all classed 'C'.
The above language development scenarios reveal a non consistent weighing
of psychological and environmental parameters in the A-B-C grading of the Ni­
gerian translator's working languages. With the vernacular languages acquiring
more social and technical status, their frequency of use has increased, particularly
among speakers of the three majority languages which enjoy the status of sole
official vernacular language in many monolingual states. Chief executives in the
public and private sectors often seize the first relaxed opportunity to discuss in
the mother tongue with equally educated colleagues, switching to English and
other languages when they have to express notions that are not yet localised in
the mother tongue. This observation is probably not scientifically based enough
to sustain the claim that, in situations where the same notion is localised in Eng­
lish and the mother tongue, the latter would provide a faster and more effective
medium for its mental processing.

A lesson for the planning of translation studies

For years to come, and for many reasons, it is not likely that any vernacular
language will replace English as the first official tongue of Nigeria. The national
language policy has continued to be discrete on the issue of vernacular languages.
Material resources for developing these languages are limited. Research efforts
are inadequately sponsored and co-ordinated. Inter-lingual status rivalry could be
escalated with a return to civilian democratic government. But, with their
growing use for official, business, technical and various social communications,
those vernacular languages that enjoy unchallenged official status in many mono­
lingual states may eventually claim the dominant role in the educated Nigerian's
 Evaristus O. Anyaehie, Nigeria 23

language configuration, thus giving English a 'B' or ' C status.

This development need not compromise the effective use of English for intra-
Nigerian and international communication and for translation purposes. If ade­
quately "acquired" or "learnt", a 'B' or ' C English could be more effectively
used than an officially guaranteed 'A' English that is communicatively deficient
This English will naturally continue to assume a more complex Nigerian aspect
as each of the resurrecting vernacular languages leaves marks on its sounds, lexis
and even syntax. Its all-time adequacy for international translation and inter­
preting services may therefore not be taken for granted in translation schools
which should offer highly weighted courses aimed at exposing the translator or
interpreter trainee to other dialects of English world-wide.
In this regard, French, seen as a foreign language, is more likely to maintain
its foreign and 'international' features among the few Nigerians who speak it,
particularly if it is not learnt at the Nigerian borders from those who are doing
with French what the non educated Nigerian is doing with English. But this
French may have to be a ' C , 'D' or even 'E' language since, going by the pro­
visions of the Nigerian language policy, the study of French at the university may
well begin after the acquisition of two or more vernacular languages and English.
With this understanding, a Nigerian minority language speaker could still be a
good English-French interpreter, despite the many vernacular languages he speaks
very well.
With the above observations on dynamic language policy provisions and chang­
ing social and functional status of languages in contact, what seems to be needed
is a more realistic planning of translation studies in Nigeria. Ihenacho's (1991)
call for the promotion of a translation culture among Nigerians even at the pri­
mary and secondary schools suggests a long term solution to the problems of
translation studies in a multilingual Nigeria. According to him, the Nigerian child
could be trained to appreciate the structures of the languages he has been des­
tined to speak with real life context in view. A child so prepared would move
more easily from tongue to tongue, scaling the hurdles of vernacular languages
separating him from English, French and other foreign languages. It is this same
culture that would enable the translator-trainee to move more freely from one
dialect of English or French to the other in response to a given communication
situation.
The Language Centre of Abia State University, in Nigeria, has integrated this
vision in its translation studies programme. Alongside English and French, the
indigenous mother tongues: Hausa, Igbo and Yoruba are working languages for
the 24-month postgraduate professional MA degree in Translation Studies with
options in Translation, Conference Interpreting and Terminology.
24 Teaching Translation and Interpreting 2

The training of vernacular language professionals responds to the increasing

need for translators and interpreters to serve over seventy per cent of Nigerians
who, unable to communicate effectively in English, are often denied their min­
imum language rights. In this regard, the Language Centre seeks to enlighten the
Nigerian private sector and such government departments as the judiciary, the
media houses, the agricultural extension and health services on the need for com­
petent vernacular language translators and interpreters.
The above training scheme has stimulated a number of research projects in
comparative linguistics; language teaching methodology; harmonisation of ver­
nacular language alphabets, orthographies and metalanguage; terminology. Re­
search findings by translator trainees and teachers are expected to add to what we
know about the translation process and could be mobilised for propagating the
translation culture so vital in a multilingual nation. Broadly conceived, these
research findings could be of interest to other researchers within and outside
Nigeria since what may be regarded as "peculiarities" of African language situ­
ations could be pointers to our limits in understanding the universal phenomena
of human language.
 TRANSLATION: A SYMBIOSIS OF CULTURES

Niranjan Mohanty, Berhampur University, India

Introduction
This article suggests that translation, in essence, is not only a bi-lingual activ­
ity, but, at the same time, a bi-cultural activity. It further tries to show that
through his act of translation, the translator generates a symbiosis between the
source culture and the target culture.
The article limits itself to literary translation, and, for convenience and ease,
I have restricted my examples to the translation of Oriya and Bengali into Eng­
lish, which is the second language for most Indians.

Translation, culture and language

It is hard to do justice to all translation scholars who have pointed out the cul­
tural dimension in translation work, but from an Indian point of view, it is
reasonable to mention particularly Translation across cultures', a special issue
edited by Gideon Toury of the Indian Journal of Applied Linguistics. In this
issue, printed as a separate volume in 1989, the cultural dimension is discussed
by the editor,1 Mary Snell-Hornby2 and Vladimir Ivir.3 Translation studies have
expanded their scope and dimension so as to include culture studies. Devoting
my article to such a perspective of culture, I propose to view translation as a
process that consciously and quasi-consciously constitutes a meeting point where
cultures are mutually enriched. Of course, this is not the same as propounding the
view that the source culture is fully absorbed into the target culture. Yet I believe
that the degrees of untranslatability can be minimized by a determination to bring
the source culture to the target culture through translational practices.
Like religion, language defines one's identity and measures the immensity of
a society or community or country. In multiple ways it enshrines the deposits of
a particular culture whence the language draws its sustenance. The co-ordinates
of a culture are revealed and reflected through language. Apart from being a ve­
hicle of communication, language is thus a transmitter and repository of cultural
signifiers. It is an identity intensifier involving the social, cultural, and religious
issues with which a nation remains eminently preoccupied. Edward Sapir claims
that "language is a guide to social reality." (1956: 69) The Soviet semiotician Juri
Lotman views language as a "modelling system" and literature and art as a
"secondary modelling system":
26 Teaching Translation and Interpreting 2

No language can exist unless it is steeped in the context of culture; and no culture can exist
which does not have a center, the structure of natural language. (Lotman and Uspensky 1978:
212)
D.P.Pattanayak, an eminent Indian linguist observes:
Language is both an expression of culture as well as a vehicle for cultural transmission. It is
both cause and an index of social and cultural change. (Pattanayak 1981: 155)

The languages of India

India is a multi-lingual, multi-cultural subcontinent. The enormity of its lin­
guistic and cultural diversity is baffling not only to Westerners but also to Indians
themselves. This diversity calls for forging a link between one language and an­
other, between the sub-culture of one region and that of another. Indian people
have been familiar with a three-tier system of language learning and acquisition
comprising (a) the regional language or the mother tongue, (b) the national lan­
guage, Hindi, and (c) the official language or the acquired language, English. In
spite of such diversities in terms of language, translation and translation studies
have remained marginalised phenomena without necessary professional
orientation.
According to a 1991 census there are 1,652 mother tongues in India. All 26
states have their regional or state languages. Fourteen languages are used in mass
communication, higher education and administration. In such a complicated lin­
guistic context, translation should have assumed a dominant role. Unfortunately
translation has not achieved such prominence.
People of India are all more or less acquainted with the national language,
Hindi. Hindi is the language of Northern India. In Southern India most people
dislike and at times detest Hindi. So Hindi has not been a palatable language in
the South. People prefer English to Hindi. The people in the South still believe
that Hindi as a national language has been imposed on them. Yet, for the sake
of national integrity and unity, the people of the South have become less voci­
ferous in their complaint. Hindi is the language in which television programmes
are transmitted, All India Radio news is broadcast, movies are circulated through­
out the country, and the national political leaders deliver their speeches to the
public. Thus the role of television, the All India Radio, and films, in making the
national language familiar, is deep-seated. In multiple ways people are exposed
to Hindi, and accordingly even a villager in Orissa, whose mother-tongue is
Oriya, does not feel that it is urgent to translate Hindi into Oriya. But when it
comes to the transmission of regional films, translation of the dialogues becomes
essential. It is interesting to observe that the subtitles are not in Hindi - the
national language, - but in English.
Therefore, translation from one regional language into another is less frequent
 Niranjan Mohanty: Translation: a symbiosis of cultures 27

than the translation from one regional language into English. The need for trans­
lation is scarcely felt by the common man. R.N. Srivastava observes:
It is worth emphasizing that, in spite of an intricate system of multilingualism, speakers of the
Indian speech community do not feel any serious difficulty in speech interaction. This is
because there is a continuous chain from the most illiterate variety of local village dialect to
the highly specialized role of English as an associated language, with the reciprocal intel­
ligibility between the hierarchically ordered adjacent areas. (Srivastava 69)
But on the other hand, for an educated man working from a literary perspective
rather than a oral perspective, translation becomes a necessity. The Indian
Academy of Letters (Sahitya Akademi) honours Indian writers and poets in Eng­
lish with awards every year. National Book Trust of India plays a significant role
in publishing books in English translation from the regional languages. Thus Eng­
lish or Indian English has become a veritable part of this subcontinent's cultural
heritage. Therefore English has a meaningful translational activity in India.
M.P.Rege, the editor of New Quest, rightly maintains in an editorial:
English continues to be the dominant language in central administration, courts of law, industry
and commerce, technology, medicine, universities, research institutes, that is in all areas in
which the rising generation will have to make their careers. This situation has to be deplored
as unnatural, but, perhaps it would be unfair to blame those who, rather than make a
determined effort to end it, help to continue it by acquiescing, for it is a situation we have all
inherited as colonial legacy. Without in any way underestimating the inherent capabilities of
Indian languages, one has to recognise that, as things are at present, English is the only
available linguistic medium for continuing activities which require a certain level of conceptual
sophistication, clear and precise formulations of statements and policies and speedy and
successful intercommunication between participants. (1990: 67)
Although it is a second language, English plays a dominant role. When
translated into English, a regional language text reaches a larger audience in
India. It also has the opportunity of reaching out to all the English-speaking
countries of the world. However, if the same text were translated into another
regional language, the symbiosis of the source sub-culture with the target sub-
culture would take place at a faster rate than that of its translation into English.
Let us consider here L, to be the source language or the mother tongue of an
Indian. Let L2 be all other regional languages, and L3 be Hindi and L4 be English.
From the point of view of learning L2 = L3 = L4 for an Indian. For an Indian
whose mother tongue is L1? L2 and L3 remain as alien as L4, although L2 and L3
are of Indian origin. From the school level, and for that matter even from the
familial level, one is more exposed to L4 or English than one is exposed to L2 or
L3. Therefore, translating a text from L1 into L2 or L3 creates a very delicate and
intricate problem.
Translation from L1 into L2 or L3 poses a considerable difficulty in the sense
that it is very difficult to find people who are really competent in L1 and L2 and
L3, so as to seriously pursue translational practices (say, for example, from Oriya
to Bengali or Hindi). Even if a text is transated from Oriya into Bengali, the
28 Teaching Translation and Interpreting 2

scope of its readership becomes extremely limited. So for a larger audience or

readership, the text in Oriya has to be translated into all other regional languages.
What makes the situation complicated is that in the absence of translators capable
of direct translation, one must resort to using an English version of the Oriya text
for, for instance, Bengali and Hindi translations. With such translational practices
the translated text in L2 or L3 is doubly divorced from the source language text.
This process of translation poses a danger in the sense that nuances of the ori­
ginal text might escape the translator. So the authenticity of the translation in L2
or L3 would naturally depend on the authenticity of the translation into L4,
namely English. But then the problem that arises is how to measure and examine
the authenticity of the translated text (either L2 or L3 or L4). Under these circum­
stances, I am compelled to believe that the authenticity of a particular text in
translation is likely to be severely and heavily distorted. In the Indian context, I
would, therefore, suggest that the translation should be done from L1 into L2 or
L3 rather than of L1 through L4 or L2 to L3.

The role of translation

The next significant aspect that I would like to emphasise is the role of the
translator. There is no doubt at all that the translator takes great pains to create
a new text for the target culture. The translator becomes an agent for effecting
a symbiosis of the source culture and target culture at the linguistic level. The
overall symbiosis of cultures would take place on the metalinguistic level and
would depend on the nature of accessibility of the source culture into the folds
of the target culture. This process is further rendered possible by what Walter
Benjamin calls the translator's willingness to "defamiliarise" the source language.
Benjamin argues that
Good translation does not seek to dispel the foreignness of languages but on the contrary allows
its native language to be affected, expanded and defamiliarized by the foreign tongue.
Therefore, the translator engages in the task of transforming the totality of his native language
by liberating it from the servitude of sense and sending it adrift to pursue its own course
according to the laws of fidelity in the freedom of linguistic flux. (Benjamin 1968: 80)
Thus, on the one hand, the translator expands the scope of the source language
by setting it 'adrift in the freedom of linguistic flux': on the other, he defam-
iliarises the same language in the context of the target culture. I would like to
add a very important point here. I believe that it is not the scope of the target
language that is expanded but the scope of the deposits of source culture from
which the translation originated that is enlarged. In other words, in translation
two activities happen simultaneously: one of defamiliarisation of the source lan­
guage and one of familiarisation of the source culture into the target culture. If
the values, attitudes and relationships which constitute the source culture are
 Niranjan Mohanty: Translation: a symbiosis of cultures 29

reflected thouroughly in the source language, and if the translation is executed

with excellence and perfection, I am sure that the initial symbiosis of two cul­
tures at the linguistic level would lead to the same process in the societal level.
In order to make my point clear I cite here two examples. In the course of my
translation into English of prayer songs and poems of Salabega, a 17th century
Oriya devotional poet, I was sure that Salabega, whose fame as a poet never
spread beyond Orissa, would now be known outside his native language and that
the Orissan sub-culture could at least be made available to other non-Oriya States
in India and to America and Europe. Salabega, the son of a Brahmin widow and
of Muslim Lalbeg, the subahdar of Bengal Bihar, Orissa, was wounded in the
battlefield. The wound did not heal despite royal treatment. He sought advice
from his Hindu mother, who advised him to surrender himself totally at the feet
of the Lord Krishna. As the son of a Muslim, he was initially reluctant, but,
because of the intensity of pain, he had no other choice. On the nineth day of his
prayer he dreamt that Lord Krishna sat beside him and gave him bibhuti (sacred
powder). Salabega used the bibhuti on his wound. The Lord disappeared. Early
in the morning, when he woke up, he discovered that he had recovered. He ran
up to his mother to narrate the miracle of the dream. From that moment he
became a staunch devotee of Lord Krishna. In the course of time he came to Puri
to see Lord Jagannatha in the famous temple. After seeing Lord Jagannatha he
was convinced that there was no difference between Lord Krishna and Lord
Jagannatha. He spent the rest of his lifetime composing poems in praise of Lord
Jagannatha. Salabega's poetry celebrates the greatness of Lord Krishna and Lord
Jagannatha as the saviours of mankind. He humanised the divine incarnations.
Salabega's poetry celebrates the myths and miracles which were associated with
these two lords and unconsciously registered the profundity of Oriya people's
faith in these deities. This miraculous but invisible bondage of an individual to
his culture and religion can indeed be depicted in English. Lord Krishna's im­
mortal love affair with his aunt Radha has been described with such impassioned
intensity that this can be felt even in English. Thus, I believe that the English
rendering of Salabega's Oriya poems would not only acquaint the reader with
Orissa's history but also introduce one to the cultural co-ordinates of Orissa. The
idealised form of man-woman or lover-ladylove relationship has been captured
through the eternal love affair between Radha and Krishna. What Salabega does
in his poetry in order to immortalise them is simply to bring these lovers to the
human level and try to infuse into them all that is essentially human. This
relationship does not abnegate the body for the consummation of love, yet the
body has been transcended through love. This relationship, however paradoxical
it might appear, constitutes the co-ordinates of Indian culture in terms of man-
30 Teaching Translation and Interpreting 2

woman relationship.
In the following poem Salabega deals with the superiority of the spirit over
matter, soul and body. He believed that, at the utterance of Lord's name, the
mortals could hope to achieve redemption.

Knowing all,
and listening to everything,
you never become wise
you never seem to have learnt anything.

Close your eyes only once,

and the whole world sinks,
withering away from you.
Yet you do wail for others,
weep showers for others in vain.

Your kith and kin

only render you a smiling company
to your wealth and prosperity.
When you sleep beyond the seasons
it is only Govinda who leads your way
to the sea of silence.

The golden parrot sits and quietly sings

in the earthen cage of your body.
When the bond of relationships withers away,
it flees, leaving the cage bare
on the harsh earth.

Bare is the river's mouth.

Why do you build embankments
on shallow nothingness?
Why do you shed tears in vain
in front of your son
when you are certain
of your exit?

Whatever you embraced as yours

whatever you fenced as your own
is nothing more than an open field.
Surely, you would leave behind all you had,
all you had won, all you did preserve,
even the weeds of years
clutching and clinging to your body.

Somehow the day passes

with the concoction of joy and sorrows,
and the night in resting and sleeping.
Never could you pray, never did you utter
the names of the Lord.4
 Niranjan Mohanty: Translation: a symbiosis of cultures 31

This poem perceptibly depicts the essence of Hindu philosophy - a crucial part
of Indian culture. So I suggest that in the English rendering the symbiosis of
Indian and English cultures has taken place. The poem in English translation
becomes accessible to an English culture that has enshrined within itself the futil­
ity of materialistic progress against spirituality.
Every language has its natural cadence and rhythm. It is difficult to translate
this cadence and rhythm. Yet I believe in translation a new cadence is created
that partly defamiliarises the cadence of the source language and partly creates
a new cadence in the target language. The translator, therefore, has to develop his
love for these cadences. Or, as Donald Hall phrases it:
One's own common speech, one's culture, one's society, one's common life, one's uncommon
psyche add their own waters,... Cadence translates only rarely and partially, good translations
attend to image and overall structure in cadences managed in the language translated into. (The
Weather for Poetry 65-66)
The Oriya cadence cannot be re-created in any other language. But in trans­
lation a new cadence is created in the target language. The reader's experience
is deeply associated with this cadence. In other words, part of the cadence
becomes familiarised with part of the target language. Thus through cadence,
symbiosis also takes place at the linguistic level.
William Radice, translator of Tagore's poems, for whom Bengali was either L2
or L3, termed translation a 'marriage'.
In a way, translation is like a marriage. Two people - the original poet and the translator - bring
their two natures together to create a joint enterprise that is subtly different from their
individual... Well we know in life marriage can never be quite perfect, but most of us believe
it is still worth attempting. (Radice 1986: 34)
I am convinced that translation not only initiates the process of absorption of
two minds, but also renders possible the marriage of two cultures, where the
source culture is bound to become infused into the target culture. Yet there are
difficulties in effecting symbiosis of cultures at the linguistic level. In Oriya lan­
guage, mana, includes both mind and will, heart and soul. But in translating it
into English I must choose either 'mind' or 'will' or 'desire'. In the case of the
devotional poets of Orissa, mana referred more to the heart and less to the mind.
In translating the word into English what one is likely to miss is the dhvani - the
essence of poetry, according to classical Sanskrit poetics. Similarly the Oriya
word abhiman and the Bengali word maan pose problems for translators. No
English equivalents are available. The meaning of these words includes many e-
motions and feelings. It is a mixture of mild anger and wild love: a concoction
of subtle anger and violent sense of possessiveness. Vexation, anger, unhappiness
and disgust may constitute a psychic state. But the word abhiman is not the ex­
pression of psyche alone - it exposes the condition of a heart that has been
slighted or wounded. When such words occur, translators have to invent an illu-
32 Teaching Translation and Interpreting 2

minating situation or to provide footnotes and detailed expository explanations

of the words cited. For example, I faced considerable difficulty in translating
"Rim Jhim Rim Jhim Barasa ratire" into English. If I translate the phrase as "In
a night of rain" or "It is night and the rains drip, drop, drip, drop", I feel I miss
the intended effect or what Walter Benjamin calls the "intento" of the original:
The language of a translation must give voice to the 'intento' of the original not as reproduction
but as harmony, as a supplement to the language in which it expresses itself, as its own kind
of intento. (Benjamin 1968: 79-80)
The target language should not only serve as the transmitter of information. It
should serve as the repository of dhvani,
The dhvani in classical Sanskrit poetics corresponds with the total effect ema­
nating from the harmonious orchestration of form and meaning. In other words,
dhvani represents the sophisticated relationship that exists between form and
meaning in the poem. K. Krishnamoorthy sums up the concept represented by
dhvani:
Dhvani or the soul of poetry is the suggested overflow of the poet's thoughts and feelings in
all their pulsating intensity and discovered by the reader at the other end. Poetry thus transcends
the confines of time and space and becomes an abiding treasure for humanity. Dhvani, whatever
its form - since its forms are indeed infinitely rich - enriches poetic expression and becomes
the only viable norm for all aesthetic value-judgment... Logically speaking, Dhvani is what
comes to be grasped by the sensitive reader after the stated beauties have slid away from the
margin of his attention. It begins where the obvious meanings end as it were. (Krishnamoorthy
1985: 121-22)
The classical Sanskrit grammarians headed by Bhartrhari made it clear that
dhvani was the equivalent of sabdahrahman or logos in the philosophy of
Sanskrit grammar. It is the first cause. It retains its all-comprehensive and meta­
linguistic significance. Anandavardhan of Kashmir writing his Dhvanyaloka or
'The light of dhvani' in the 9th century A.D., asserted that dhvani, when structu­
rally analysed, was all-comprehensive and pervasive of all micro-element in
poetry. Thus dhvani included word and meaning, sound and sense, name and
form, and, in absolute metalinguistic terms, God and his glory. My primary in­
tention in digressing from the symbiotic nature of translational process is simply
to examine whether translation can transplant the dhvani of the source language
in the target language. I earnestly believe that such a transplantation is possible
provided that the translator treats translation not as an exercise but as a mediation
in which not only two languages but also two cultures support each other. Only
when such a transplantation becomes possible, one realizes that translation effects
the symbiosis of culture deposits more positively and profitably. Recently I trans­
lated a Bengali poem entitled "Maa" or "Mother" by one of my poet-friends,
Debi Roy. Let us see if the dhvani of the poem in the source language is
reflected in the translation.
 Niranjan Mohanty: Translation: a symbiosis of cultures 33

THE BENGALI VERSION

Maa-
Koler opår khola Rāmāyán. Udash chokhē bhāri chasma.

Maa -
Chokh ekbar Ramāyånēr pātāy, aar ekbār
oi doore, talmålo-payer damal-natni,
khola get, na beriē jaē!

Maa -
badoti - bēsh bado, giechhē bideshē, chithi o'
asheni mashadhikkāl!

Maa -
Karta giechhen swargē...ekatha keno udē asē?
Chokhē ki bali? Dristio jhāpsā, samner jākichhu
sab aspåstå, dhoan, - ēr bhitar natni doudē ēsē
bookē jhampae...

Maa -
chokher jål gadiē padē Rāmāyánēr halood pātāy!5

(Desk 17 October 1992)

ENGLISH RENDERING

Mother:
The Ramayan lies open on her lap;
her eyes, beneath the glass, grave.
Mother:
her eyes flutter once on the pages of Ramayan
and then on the unruly grand-daughter,
dragging her faltering steps to the gate,
open yonder.

Mother:
the eldest son has gone abroad;
more than a month lapsed,
but no letters!

Mother:
her man has already left for the heavenly abode...
why does this thought fly back?
Dust in the eyes?
Her vision dim; all is opaque, smoke-like...
And in the meanwhile, the grand-daughter
comes leaping to her chest
and begins to dance.
Mother:
her tears trickle down to the yellow pages of the Ramayanl
34 Teaching Translation and Interpreting 2

I have attempted to retain both the structural unity of the original poem and its
spirit. The poem reveals a familial experience. Mother's loneliness has been laid
bare. But the reference to Ramayan, the great Indian epic, serves as the co-
ordinate of culture. Even without a footnote on the Ramayan, the translated poem
is capable of evoking the plight of the lonely mother in the absence of her hus­
band. This loneliness is partially relieved by the presence of the granddaughter.
Torn between the memories of her husband and the present company of the
grand-daughter, her eyes no longer hold back her tears. The images in the poem
are sharp, well-defined and to the point in evoking the mother's emotion. I
believe I have managed to retain the dhvani of the original. In order to clarify the
significance of the Ramayan, a footnote can be added. Ramayan is an epic that
depicts the victory of good over evil. At the same time it shows the intimacy, the
relationship between Rama, the hero and Sita, the heroine. Rama accepted ban­
ishment for fourteen years just as he was about to become the king of Ayodhya.
Sita merrily accompanied Rama to the forest. This togetherness was temporary,
for Ravana, the evil incarnate in order to avenge the insult of his sister Sur-
panakha, stole away Sita from her cottage. Rama grew lonely. Yet he and his
brother Laxman could defeat Ravana, the king of Lanka, and could rescue Sita.
The Rama-Sita relationship is an ideal: both have experienced loneliness.
Ramayan is a religious book and hence old men and women read it to redeem
themselves from the pangs of loneliness and to achieve salvation. This faith is
peculiarly Indian. Thus, the incorporation of Ramayan into the poem makes Deby
Roi's poem significant. I am sure that a footnote would make a reader in the
target language and target culture fully aware of the significance of the Ramayan
as a cultural coordinate of the source culture. I am hopeful that a symbiosis of
cultures could come full circle by this method of translation.

Teaching translation
Keeping in mind the significance of translation as an instrument effecting sym­
biosis of cultures, I have conceived a model of teaching. Translators or trainee
translators would take a 'bridging course' in which the teaching of the cultures
involved is Central, and more than the mere teaching of language. The purpose
of such a course would be to enable the translator to identify the areas of
translatability of the source language into the target language. He would try to
discover the areas into which he can construct a bridge, not only between two
languages but also between two cultures. The languages, the language structure,
the grammar and the phonetics in both must be taught. The translator must have
acccess to the language history and literary history of both languages. Access to
dialects, and familiarity with registers of both languages would enable the trans-
 Niranjan Mohanty: Translation: a symbiosis of cultures 35

lator to handle any language situation. The translator should be trained to identify
the characteristics of each culture, especially those that stand out in relation to
the 'contrastive culture', the culture-specific features which we might term the
'intensifiers of cultures', in addition to the peculiarities of the language situation.
A bridging course along these lines could to be introduced in three phases. In the
first phase the translators would be trained to acquire competence of source and
target language. In the second phase they would be taught to acquire knowledge
of the cultural intensifiers, so as to enable them to identify the points of conver­
gence and divergence. In the third phase the translators should learn to identify
culture intensifiers. These three phases are summarised as follows:

Phase I
Acquisition/Teaching of linguistic and cultural competence in

Source Language + Source Culture

Target Language + Target Culture

Phase II
Identification of cultural identifiers by contrastive analysis between

Source Language + Target Language

Target Language + Target Culture

Phase III
Exemplification of culture intensifiers

a) National linguistic peculiarities:

Is the culture monolingual, bilingual, multilingual?
Is there a dominant second language? What does this imply for language and culture?

b) What are the media for teaching in schools and higher education, in terms of language, in
terms of teaching methods, norms, and so on?

c) School:
What types of school are there ?
And what is the relationship between students and teachers?
36 Teaching Translation and Interpreting 2

d) Nature of the Family:

Is it patriarchal or matriarchal? is it based on concepts of extended families or on a nucleus
family?
What are the gender and generation relationships:
Father - mother
Father - son
Father - daughter
Mother - son
Mother - daughter
Grandparents - grand children
Father/mother-in-law/ - son/daughter in-law

f) What is the literacy rate? How does this affect language, language policy, and culture?

g) Religion: minority or dominant?

h) Sources of national and private (family) income:

agriculture, industry, civil service, etc.

i) Type of government

j) History: colonised, coloniser, independent

h) Literature: history of literature, and the nature of popular and dominant genres.

i) Myths: traditional, vs. created

In my discussion I have illustrated how factors of this type must be taken into
account in translation work. It is true that I have not referred to all such factors,
but this would have been a Herculean task. I hope I have illustrated that any of
these points may be relevant to specific translations. Therefore, I sincerely
believe that teaching these three phases would be of benefit to translational
practice. This way of teaching would not only make the translators acquire the
competence of source language and target language, but would also enable them
to have a comprehensive knowledge of both cultures so as to accelerate the
process of symbiosis of cultures. I am hopeful that, in the years to come, re­
searchers and theoreticians would make attempts to invent new teaching methods
so that translation will become a scientific activity with a sound professional ori­
entation. Even after so many theories, so many new methods of teaching trans­
lation, I am sceptical about the harmony between the theory and practice of trans­
lation, unless theoretical thinking is integrated in teaching.
 Niranjan Mohanty: Translation: a symbiosis of cultures 37

Notes
1. Gideon Toury. 1989. Integrating the cultural Dimension in Translation Studies: an Introduction.
In: Toury, Gideon (ed). 1989. Translation across Cultures. 1-8.
2. Mary Snell-Hornby. 1989. Translation as Cross-cultural Event. In: Toury, Gideon (ed). 1989.
Translation across Cultures. 91-105.
3. Vladimir Ivir. 1989. Procedures and Strategies for the Translation of Cultures. In: Toury,
Gideon (ed). 1989. Translation across Cultures. 35-46.
4. Salabega. White Whispers. Unpublished manuscript. Translated from Oriya by Niranjan
Mohanty. 24-25.
5. Debi Roy. 1992. Maa. Desh, October 17.
 CULTURAL BARRIERS -
TACKLING THE DIFFERENCES
 TRANSLATING AFRICAN LITERATURE
FROM FRENCH INTO ENGLISH

Moses Nunyi Nintai, Cameroon

This article discusses factors which should, in my opinion, be taken into

account when translators are to be trained to work with two languages neither of
which is their mother tongue. This is the case in many places in Africa, such as
my own country, Cameroon, where French and English are the primary languages
of published literature.1 Translation of African literature between European lan­
guages, notably French and English, has been carried out since the middle of this
century, but little has been done formally to train the translators who disseminate
this fast growing literature to the rest of the world.

The current situation

Even today most written African literature is not published in the national lan­
guages, but in European languages of colonization (English, French, Portuguese,
and Spanish). Nevertheless, only a few of these works have been translated
between these European languages. Indeed, Ade Ojo estimates that translation
"has not succeeded in affecting up to five per cent of African works" in European
languages.2 Few Africans are interested in or do literary translation; the vast
majority of trained African translators work for government services, international
organizations, and private companies. Most of the translations of African literary
works from French into English have been done by university scholars and critics
(specialized in African literature) who also engage in literary translation as a
sideline.3
Most literary translations are published (and often requested) by European and
American companies, and only a few in Africa.4 Publication is dominated by
prose fiction; this may reflect the taste and preferences of readers, but it is clearly
also due to the fact that most African literature is prose fiction. Publishers are
cautious about drama and poetry which are less popular. Consequently, poems
by one or several poets are usually selected and translated for anthologies.5
The interests, ideology and cultural values of the large readership in Europe and
North America continue to play a decisive role in the selection of African literary
works translated. Moreover, publishers are often guided by market forces,
publishing only translations they feel would satisfy their target audience and be
economically profitable.
42 Teaching Translation and Interpreting 2

Objectives of training
The vast majority of literary translators have not received any formal training.
Since there seem to be fundamental problems about such training, many insti­
tutions have been reluctant to introduce such courses, but in view of the shortage
of qualified African literary translators, certain institutions are, nevertheless,
attempting to provide relevant training.6 The definition of the objectives, nature
and pedagogics of such training is, however, problematic.
Many practising translators question the possibility of and need for teaching
translation or imparting the competence and skills that would enable a person to
translate texts, especially literary texts. Arguments range from insistence that the
ability to translate is a gift (and so cannot be formalized and transmitted), that
the translation process is too complex to be reduced to a mechanical operation
(hence the failure of machine translation), to the assertion that translation skills
can be acquired and improved only through practice and experience. Even if one
admits that literary translation requires a certain "creative gift", there are grounds
to posit that training could enable those who are not gifted to acquire and im­
prove the necessary skills. That is why, I believe, there are several schools of
translators today.
In my view, a training course in the translation of African literature should aim
at imparting the required knowledge and skills, laying particular emphasis on the
features distinguishing Europhone African literature from the corresponding
European ones.7 Translating African literature from French into English would
thus require competence in the two languages involved, ability to analyze and
transfer source text message, and an adequate knowledge of African literature.
The following is my vision of such a course.

Contents of training
Literary translator training should comprise theoretical as well as practical
courses. In African literature, besides focusing on fundamental problems common
to literary translation from French into English, students would need to be intro­
duced to research and studies carried out on the translation of sub-Saharan
literature between English and French by scholars such as Ade Ojo, Brenda
Packman, Charles Nama and Timothy Asobele. Practical translation exercises and
analysis of existing translations should offer students the opportunity to translate
sample texts and see how others have solved certain translation problems.
Theoretical insights could enable students to become aware of their task, avail­
able translation options, and of factors involved in decisions and choices. Thus
rather than formulating translation rules and principles, theoretical courses could
focus on describing and explaining the implications of the distinctive features of
 Moses Nunyi Nintai, Cameroon 43

Francophone African literature for translation. For example, students must realize
that French is often used in sub-Saharan literature to express the world view and
an extralinguistic reality peculiar to those parts of Africa, and that the resulting
"double language" requires that the translator be also familiar with the African
language which influenced the French in question. Furthermore, students must
take into account not only the two cultures of the source text (French and
African), but also the target text cultures (English, African, and non-African
reader's, as the case may be). In order to convey the content, students should be
taught how to analyze texts within their African historical, political, social, and
cultural contexts. In some works, information about the authors and their position
within the literary tradition may further enhance comprehension; for instance,
information about the author, Camara Laye, will shed much light on L'enfant
noir in which some events can be adequately interpreted only in the light of
Laye's youth in Guinea.8 In references to African culture and environment,
students should examine how to render the names of local objects, dishes, dress,
drinks, etc. Other questions to be discussed include: How can the translator
convey the two cultural dimensions of the source text? How can African-based
proverbs, imagery, dialogue, and other rhetorical devices be rendered? How can
one render language switches between French and "français petit nègre"? How
can a translation be made accessible to a heterogeneous audience of Africans and
non-Africans? Will the uninformed reader need more information in order to
understand and appreciate the translation? If so, how can such information be
provided? (Within the text, in footnotes, in a glossary or in an introduction?) In
trying to answer these questions, the following comment by Ade Ojo deserves
consideration:
Not only is [the translator] to be faced with the African version of the European language that
he is to translate from but he has to do a very thorough study of the socio-cultural backgrounds
against which the [source text] is written and where the [target text] will be read. The translated
version of the [source text] must therefore have a tinge of Africanness; it must also possess the
style of the original text and express very appropriately the mind of the writer. (1986: 296)
The practical exercises should be organized so that they complement the theor­
etical courses. This implies that selected texts aim at illustrating aspects discussed
in theory and enable students to explain the translation decisions they eventually
make, or comment on existing translations. For example, discussions could focus
on how different translators have coped with similar problems, and the impact
of the various choices: while translators generally maintain culture-bound terms
and expressions, there are instances where words for local dishes, dress, drinks,
etc. are mistranslated either because of inadequate analysis of the source text or
because of unfamiliarity with the extralinguistic context as we find in John
Reed's rendition of "baton de manioc" (from Ferdinand Oyono. Une vie de boy,
44 Teaching Translation and Interpreting 2

p. 7) as "cassava sticks" (Houseboy, p. 3), and in Peter Green's translation of

"noix de cola" (from Mongo Beti's Mission terminée, p. 59) by "chewing gum"
{Mission to Kala, p. 64), respectively.9 This exercise could also be extended to
the translation of culture-bound proverbs and other oral literary devices, the ways
in which translation choices suggest the target readership, and other aspects.
Existing translations can serve to show translator creativity in literary translation.
Students should, of course, be encouraged to read Anglophone African literary
works by prominent writers such as Chinua Achebe, Ngugi wa Thiong'o, and
Wole Soyinka; this will expose students to creative writing that would be useful
if the "Africanness" of the source text needs to be reflected in translation.
Furthermore, samples from untranslated works can be examined and translated,
either in class or as homework; students' choices could be discussed, proposing
other options and determining which of them should be adopted in the exercise
and why.
Students could also be introduced to the profession of literary translator, for
instance, by having prominent translators and publishers of Francophone African
literary works talk to students about their difficulties in the field and how they
cope with them. Admittedly, in pedagogical environments, students are often
likely to make every effort to account for the literary qualities of the source text
and respect the original author's lexical and syntactic choices. However, in the
real world, the literary translator will have to deal with the publisher and the
target audience. Students need to be aware of the role and influence of the
publisher in the final translation of a literary text; they need to know that the
publisher will often influence the overall translation strategy, and even "edit" or
"tailor" translations to the taste and expectations of the target readers and for
other commercial reasons.
Moreover, students must also be acquainted with some of the potential func­
tions that the translated African literary text will have to fulfil: for instance,
making an African work in French available to Anglophone readers; conveying
the cultural setting and values in the work; producing a fluent or transparent
translation that will make enjoyable reading for a given audience; highlighting
the literary qualities of a particular work. These functions are different from those
of the classroom; and students who eventually become professional translators
should be made aware of them and how they can best be achieved.

Conclusion
The above views are not intended as rules or the only way of teaching the
translation of Francophone African literature into English; they represent only my
conception of how to teach literary translation. Teaching methods and perspect-
 Moses Nunyi Nintai, Cameroon 45

ives necessarily depend on, and are best adapted to, the specific circumstances
and needs of teachers and students; teachers will select methods and approaches
to suit their goals and preferences. Obviously, teachers will sometimes rely on
their intuition and competence to solve certain problems.
Despite the danger of training too many literary translators for a market often
marked by very low demand, relevant courses could be introduced in universities
in Africa for students interested in the profession on even a part-time basis; in
fact, most of those trained will, in all likelihood, come to work with publishing
houses or on a free-lance basis alongside other main jobs. Nevertheless, the im­
portance of consultation and cooperation among teachers cannot be overem­
phasized. Consequently, seminars and conferences to discuss pedagogical
problems and exchange experiences would go a long way towards promoting the
teaching of literary translation. Teachers must be encouraged to participate in
research projects, and exchange programmes between schools and universities
will expose students to different approaches and experiences. I am convinced that
with the introduction of appropriate courses, especially in African universities,
and better teaching techniques, interest in the translation of African literature
would be greatly enhanced.

Notes
1. Cameroon is officially bilingual (French and English). The French-speaking provinces of the
country account for about 80% of the population. Since much of the written literature is expressed
in French, translators could contribute to the development of Cameroonian literature by making
Francophone literary works available to Anglophone readers in Cameroon and other countries
through translation.
2. The Role of the Translator of African Written Literature in Inter-cultural Consciousness and
Relationships. Meta 31 (1986), 298.
3. These translators include Africans (Modup6 Bod6-Thomas, Simon Mpondo, Olga Simpson, etc.)
and non-Africans (Clive Wake, Dorothy Blair, John Reed, James Kirkup, Richard Bjornson, etc.).
Although Africans are becoming increasingly involved in the translation of African literature, the
field is still dominated by non-Africans, especially Britons and Americans.
4. These companies include Heinemann (African Writers Series), Longman, and Macmillan in
Britain; Présence Africaine, Hatier, Peuples Noirs,P.J. Oswald, and Gallimard in France; Three
Continents Press in the United States. In Africa, translations have been published by New Horn
Press in Nigeria, Les Nouvelles Editions Africaines in Senegal, and regional offices of Heinemann
in Ibadan (for West Africa), Nairobi (for East Africa), and Gaborone in Botswana (for Central
and Southern Africa).
5. For instance, Clive Wake has compiled French African Verse with English Translation (1972)
and Melvin Dixon has translated a collection of Senghor's poems under the title Leopold Sedar
Senghor: The Collected Poetry (1991).
6. Such is the case, for instance, at the Advanced School of Translators and Interpreters (ASTI),
University of Buea, Cameroon, where students are encouraged to translate excerpts from African
literary works, especially between English and French, in partial fulfilment of the requirements
of the postgraduate diploma in translation.
7. These features have been examined, in relation to English and French, in my article "African
Literature in European Languages: Major features and Implications for Translation", in
46 Teaching Translation and Interpreting 2

Proceedings of the 13th FIT World Congress. London: ITI. 1993. I, 564-572.
8. The discussion refers to Camara Laye. L'enfant noir. 1953. Paris: Plon; translated by James
Kirkup & Jones Gottlieb. The African Child. 1959. London: Collins.
9. The novels are Mongo Beti. 1957. Mission terminee. Paris: Buchet/Chastel; translated as
Mission to Kala by Peter Green. 1958. London: Muller. And Ferdinand Oyono. 1956. Une vie de
boy. Paris: Julliard; translated by John Reed. 1966. Houseboy. London: Heinemann.
 SUPRA-LINGUAL ASPECTS OF LITERARY TRANSLATION

Manouchehr Haghighi, Allameh Tabataba-ii University, Iran

This article deals with culture-bound problems in literary translation. My angle

will be Persian and the focus will be on English and Farsi. When I, after having
taught translation of political and scientific texts, turned to teaching literary
translation I suddenly encountered a whole new set of problems. I quickly real­
ized that it required more than a mastery of the two languages to make a felic­
itous translation. The main problems in translating English literary texts into
Persian and vice versa, are the cultural, social and moral barriers which separate
the European and Iranian cultures. Therefore I tried explaining to my students the
cultural differences between Europe and Iran, but found that this was very diffi­
cult. In my opinion, European values, norms, attitudes and perceptions are rooted
many centuries back, specifically in the Renaissance, and today these are blended
into the texture of Western society and taken for granted. But as my students had
a non-European background, these multifarious supra-lingual elements very often
led them to complete lack of understanding or a distorted view of the ideas
expressed in European literature. Some of the students' responses indicated that
they regarded these ideas as something of another world or even another planet
light-years away from Earth.
One of the things I discovered was that the closer we got to modern times, the
more insurmountable seemed the cultural and social barriers. For instance, the
students had less trouble translating Chaucer's The Canterbury Tales than ren­
dering Shakespeare and Marlowe into Farsi. I assumed this was due to the fact
that the latter two were influenced by the humanistic ideas of the Renaissance.
They saw man as a unique individual whose conditions (his relationship to so­
ciety, nature and God) should be reflected and discussed in their works. And
though language certainly adds to the beauty of their works, it is not the main
issue of their texts. For example, when we worked on Hamlet's monologue "To
be or not to be..." I had to make a choice: either I would teach the students how
to make a literal translation of the text without analyzing the cause of Hamlet's
grievance in the light of humanism and the social and cultural norms of
Shakespeare's time or I would go into extensive discussions on Shakespeare's
time, humanism, the Renaissance concept of the ideal man and so on. I first
opted for the literal translations, but soon discovered that these could not convey
48 Teaching Translation and Interpreting 2

the message - not only were they unfair to the writers, but they also gave me
moral qualms. The burden on my conscience was so heavy that I sometimes
feared that the offended ghosts of the writers would one day come to haunt me.
Moreover, students' responses often prompted me into discussions, as when they
asked "why doesn't Hamlet kill his uncle as soon as possible?" or when they
thought that when Hamlet talks about "to be or not to be" he is merely reflecting
on life and death. So I was forced to abandon the idea of teaching students how
to make a purely literal translation. And then came all the barriers, conceptual
ambiguities and misunderstandings that usually hinder a meaningful translation
and the gap between the students and the authors seemed so wide that they would
never be able to meet.
When we had finished working with Shakespeare and Marlowe, I assumed we
would have less problems when we were going to try our hand at translating 20th
century poets. I had erroneously thought that since the language of the modern
poets is superficially simpler and since they are, after all, our contemporaries, it
would be easier to understand their works and thus to translate them. But bitter
facts soon dawned upon me: the ideas and concepts behind the deceptively
simpler language were even more complicated than those of the Renaissance
poets. New dimensions had been added to them and the gap between these poets
and my students was wider than ever.
To clarify my point I will mention the difficulties my students had translating
T.S. Eliot and Samuel Beckett. Eliot's The Waste Land was conceptually enig­
matic to them, mainly because they could not grasp his theme of 'spiritual
emptiness'. Nor could they understand the loneliness and the useless waiting of
the characters in Waiting for Godot They could not understand that by letting the
characters exchange so few words, Beckett wanted to demonstrate that man has
nothing more to say. And I cannot not blame my students, for their lack of
understanding is due to culture-bound factors which tie up closely with the
Iranian versus the Western world. They have not got the Europeans' experience
of the First and the Second World Wars. They have not experienced the dimin­
ishing role of religion to the same extent that the Europeans have; they have been
less influenced by materialistically-oriented economic, political and social the­
ories than people in the Western World; they have not felt the pressures of mass
society and conformism! These and similar developments have shaped the frame
of mind of Western man and has influenced his literature. No wonder Iranian
students cannot grasp the concepts in modern Western literature. They have been
brought up in a traditional society where a fixed and never-changing view on
man, nature, society and God has been inculcated incessantly. A society which,
for different reasons, has not been through a process similar to that of the
 Manouchehr Haghighi, Iran 49

European Renaissance. This difference in cultural and historical backgrounds has

brought about mental differences, too. The task is then to reconcile the two con­
trasting cultures and their fundamentally different conceptual frameworks when
translating from one language into the other.
In my opinion the two cultures and two mentalities can never be fully recon­
ciled. We may be able to come to some sort of reconciliation by providing
students and translators with background information, but this will never be suf­
ficient to make up for hundreds of years of cultural and social differences, arising
from various periods of development such as the Renaissance, the Enlightenment,
the French Revolution, the industrialization of Europe, etc. These developments
have not only changed the material and physical life of man, but have also trans­
formed his perception of the world. Undoubtedly, in this process of faster and
faster transformation, poets and writers (who are usually seen as the antennae of
society) have been even more susceptible to the changes than the ordinary man,
and have therefore been more easily and profoundly influenced by them. Thus
the gap between Western and Oriental societies has widened. While people in the
West were occupied with scientific and technological advancement, people in the
Orient were still preoccupied with their traditions. They live in different worlds.
These cultural differences have also led to linguistic differences. For example,
the word 'loneliness' in a play like Beckett's Waiting for Godot has a metaphys­
ical connotation and refers to man's universal loneliness, his being left in a void
without any supreme power to turn to. To most of my students it simply means
being alone in the sense of not being in the company of other people. A number
of new concepts have entered the Western languages mainly in the wake of the
profound changes affecting Western civilization in the 19th and 20th centuries:
growing materialism, the declining role of Christianity, the findings of psychol­
ogists and sociologists and the increasing pressures of mass society. They include
concepts such as 'alienation', 'self-alienation', 'anguish', 'disillusionment' and
'spiritual emptiness'. We do have Farsi equivalents, but these are purely literal
and not conceptual. They can replace the English words, but they do not commu­
nicate the ideas. To the Iranian mentality words covering metaphysical experi­
ences seem odd. Not that the Iranians do not have metaphysical experiences, but
they describe them differently and see them from another perspective.
In conclusion, I would like to stress that translators must have background in­
formation about the writers whose works they are translating and about the ideas
they are discussing. This information may not be of great help to the translators,
but they are better off with it than if they are left completely in the dark. In my
opinion, some literary works are almost untranslatable (at least into Persian)
because they are so heavily charged with concepts that are unfamiliar both to the
50 Teaching Translation and Interpreting 2

translator and the reader. But in spite of all barriers and misinterpretations, I
think that we should go on translating literature, because this is one of the best
means of bringing about reconciliation between cultures. And although this takes
time we should not give up hope.
 CROSS-CULTURAL AWARENESS: FOCUSING ON OTHERNESS

Meta Grosman, University of Ljubljana, Ljubljana, Slovenia

It is not difference that makes for divisiveness; rather it is the lack of appreciation for diversity
that interferes with successful cross-cultural communication.
von Raffler-Engel

Introduction
This article discusses cross-cultural awareness as an awareness of the otherness
and differentness of others, or rather, of foreign cultures in all their complexity.
This presupposes a capacity for noticing, and, consequently, for understanding
and tolerating the otherness of foreign cultures and their literary products. If
cross-cultural awareness is cultivated, it can prevent the automatic tendency to
perceive the other and the different in terms of the known and the familiar,
whereby foreign texts are divested of their very otherness. With literature's
special capacity for drawing us into other worlds, literary translation offers the
unique possibility of re-creating imaginatively an alien society with its different
problems and solutions. Cross-cultural awareness promotes open-mindedness be­
yond one's own cultural border, contributing to a better understanding between
people. Conceived in this way, it seems a prerequisite for all successful com­
munication in a world rapidly turning into a multicultural village. Cross-cultural
awareness is particularly important to those involved in the promotion of foreign
literature, whether literary translators, critics and editors of literature in trans­
lation, teachers and students of foreign literatures, or those reading merely for
pleasure.
Cross-cultural awareness thus constitutes an indispensable body of knowledge
about the possibilities and relevance of differences between cultures and litera­
tures which must be integrated into the training of students of translation. The
aim of this article is to point out the need for a better understanding of cross-cul­
tural reading, and to show how existent literary translations can contribute to
heightened cross-cultural awareness in order to reduce undue assimilation in liter­
ary translation.
If we want to develop the ability to perceive and appreciate the otherness of
foreign literary texts in a cross-cultural context, we must first have some under­
standing of how and why we tend to assimilate when we read. In its basic, initial
stage, cross-cultural assimilation is but a form of the usual reader assimilation of
the text, resulting from an effort to make sense, to force the text into acceptabil-
52 Teaching Translation and Interpreting 2

ity. The readers' appropriation of the meaning of the text according to their needs
or desires and according to their culture-specific critical assumptions and pre­
dispositions, has come to be regarded as an inevitable consequence of the process
of active meaning-production.

The translator as reader

The significance of the translator's own reading of the source text, as first des­
cribed and analysed by Beaugrande in 1978, is by now well established. Both the
statement that translators are readers who translate their reading experience to
other readers (Oittinen 1992), and the analysis of the process of translating in
terms of scenes and frames (Vermeer and Witte 1990), emphasise the fact that
no translator can claim impersonal access to the textual meaning, nor is it poss­
ible to translate in a social vacuum. Contrary to other individual readings, the
translator's own reading will not remain private but will become that meaning of
the source text which will be passed on as its translation to the readers in the
target culture. The translation thus simultaneously closes off the source text and
opens new possibilities for its interpretation. This fact compels the translator to
a particularly attentive reading which presupposes a capacity to reflect on one's
own interpretation and its culture bound particularities. In such reflections the
translator can be assisted considerably by a knowledge of the process of cross-
cultural reading, which should, accordingly, be taught in courses on literary
translation.

Studies of cross-cultural reading

Some tendencies in cross-cultural reader responses were demonstrated as early
as 1932 by Bartlett, who clearly showed how readers assimilated unfamiliar
information in culturally alien texts to their own culture at all the levels of
reading. On the immediate textual level they omitted material which appeared
irrelevant, or turned unfamiliar textual items into more familiar counterparts. On
the level of story construction, they persistently rationalised the whole story and
its details and virtually divested it of all puzzling elements in order to render it
more acceptable, comfortable and straightforward. To make it fit their own ideas
of how stories should be, and so produce what seemed to them a more coherent
whole, they attributed special importance to certain incidents and reordered all
other incidents around them.
Since Bartlett, several scholars have examined reading in a cross-cultural con­
text. Kintsch and Green (1978) studied the role of culture specific story-schemata
in the comprehension of stories. Harris, Schoen and Hensley (1992) established
experimentally the impact of cultural scripts and other associated cultural knowl-
 Meta Grosman, Slovenia 53

edge on story recall, whereas Kang (1992) described some effects of culture-spe­
cific background knowledge and subsequent reading inferences in the process of
cross-cultural comprehension and interpretation. Halász (1991) demonstrated
cross-cultural differences between the readings of American and Hungarian
readers. In their study of the impact of text variables on personal resonance,
László and Larsen (1991) have shown that cultural proximity of texts generates
a large proportion of personally experienced, culturally rich, and vividly
remembered events.
Cross-cultural differences are above all noticeable in the field of story
construction that reveals a strong tendency to rewrite virtually the whole story in
accordance with the reader's own culture-specific expectations (Vargo, forth­
coming; Grosman, forthcoming). This tendency is especially strong in the final
thematization at the post-reading stage, at which the reader tries to make sense
of the whole.
In the latter field Krusche (1990) has provided an illuminating analysis of the
cross-cultural variation in students' thematization of Kafka's Das Urteil de­
pending on their different culture-bound concepts of human relations. His com­
plex study reveals considerable consistency in the thematization of the text within
nationally homogeneous groups of readers and great differences among readers
of different nationalities. Each group of readers - European, Algerian, Indian,
Japanese and Chinese - formed its own thematization of this complex text in ac­
cordance with their various concepts of family structure, of the role of the father,
his authority, and similar elements.
New interdisciplinarily conceived projects which involve readers from several
and rather diverse cultures promise to bring a more complex and comprehensive
picture of similarities and differences in the response to literature in a cross-
cultural context (Dollerup, Reventlow and Rosenberg Hansen 1993).

Case studies
Assimilation and shifts occurring in the immediate encounter between the
reader and the text are more difficult to trace simply because they are harder to
recognise and impossible to recall at the post-reading level. However, it is in the
immediate processing of the text that contrastive studies of original and trans­
lation can help to reveal shifts and loss of meaning in a cross-cultural context,
and can accordingly be used in order to sensitise the students of translation to
such loss and its consequences. These naturally vary a great deal since they de­
pend on the quality of the text, the particular relationship between the source and
target culture, and the individual translator. When choosing the translations to
make the students of translation aware of the pitfalls of cross-cultural reading, all
54 Teaching Translation and Interpreting 2

these factors have to be taken into account.

Some texts presuppose a well-defined knowledge of culturally specific pheno­
mena, such as a specific social structure, on the part of the reader or rather the
translator. In case translators do not possess the knowledge of such an underlying
social structure, or are unaware of its fundamental differentness and the import­
ance of such differentness for the appreciation of the text, they are likely, without
realizing it, to assimilate the text to the familiar social structure of their own
different culture and thus lose the otherness of the original, reducing its inter­
pretative potential.
We can observe this process at work in the Slovene translation of Jane
Austen's Pride and Prejudice, in which we find several examples of unnecessary
assimilation, invisible to readers acquainted with the translation only.1
The original is firmly based on gentlemanliness, which transpires through the
deeds and linguistic behaviour of all the characters and thus defines them,
whereas the Slovene translation offers an inconsistent rendering of the social
structure by muddled translations of polite address, titles and references to gentle­
manly behaviour.
Taking for granted the reader's knowledge of the position of the gentry and the
underlying concept of gentlemanliness, Austen uses the word 'gentleman' and its
derivatives carefully and consistently in Pride and Prejudice. Unaware of the im­
portance of these words and the underlying concept, the translator does not
render them with the same consistency. The Slovene translation sometimes re­
tains 'gentlemanlike' and 'gentleman' - the use of this loanword is permissible
according to Slovene dictionaries - whereas on other occasions, without any
special detectable reason, it renders these words by "gospod" and its derivative
"gosposki". Being the usual translation for 'Mr' preceding the family name,
"gospod" also refers to all male characters presented by surname and therefore
fails to provide an adequate textual clue for the reader's construction of the social
status of the characters involved. On the contrary, in some references, as for in­
stance in Elizabeth's description of herself as 'a gentleman's daughter' (Austen:
366), though trying to explicate the text by adding "rod" [= 'origin'] to "gos­
poski", the translation only makes a Slovene reader wonder about Elizabeth's
standing. In this way the intrinsic social satire is often lost while the text acquires
some of the characteristics of a trivial romance.
Careful examination of examples of such assimilation and the losses of
meaning they occasion, can help students to see the differences between the
translation and the original, and teach them the necessity of an attentive and open
reading in cross-cultural contexts.
Examples of various forms of assimilation of character, due to the translator's
 Meta Grosman, Slovenia 55

inattention to particular character features, are also quite frequent in translations

which are closer to our time and world than Pride and Prejudice. Apparently
minor changes in the description of characters can mediate distorted presentations
of them. This is illustrated by, for instance, the Italian translation of David
Lodge's novel Small World (1985), entitled Il professore va al congresso (1990).2
Professor Zapp's comment on the changed circumstances of academic life, in par­
ticular the disappearance of library work thanks to the introduction of photo­
copying, which induces him to "work mostly at home or on planes", ends with
a matter-of-fact statement: "I seldom go into the university except to teach my
courses" (p. 44) in the English original. This statement seems to be in perfect
accord with the entire paragraph, its ironical overtone, and with Zapp's person­
ality. Conversely, the Italian translation of this sentence: "Io vado all' university
raramente, e soltanto per tenere i miei corsi" (p. 62) ascribes to Zapp a common-
place attitude of certain Italian professors, which is not in keeping with American
notions of responsible professorship and not at all in accord with the Zapp image
of the original. The German translation "An der Universitat bin ich eigentlich nur
zu meinen Seminaren" (p. 52) comes much closer to the original. The Italian
translation cannot uphold the respectability of university professorship for Italian
readers, who take this profession much less seriously than other nationalities. Ob­
viously sharing the Italian attitude, the translator renders university titles incon­
sistently, also because 'il professore' is used more often and has a rather broad
meaning in Italian.
Texts in which the reader's perception and ideas of the characters depend to
a large extent on the characters' direct speech can present special cases of un-
translatability. Thus, for instance, all the attempts to translate Mark Twain's
Huckleberry Finn, with its four distinct culture-specific dialects, are bound to fail
in trying to present Huck and Jim in the same way as in the original. Every at­
tempt to render Jim's words, the most direct and revealing repository of his unas­
sertive, loving, irrational, passionate, and reliable humanity, is of necessity in­
adequate, because no other language has an equivalent sociolect or can even dis­
tantly imitate Black-American English. Thus all translations have to find sub­
stitutes, often inadequate variants of slang, which fall short of mediating Jim's
original dialect. Translations of Huckleberry Finn have never attained the popu­
larity of the original because it is impossible to recreate the central characters
through their speech.
The Slovene translation, for instance, renders the entire text in standard
Slovene, thus reducing this novel to a mere adventure story, intended for children
aged 12 to 14. Though the differences between the original and the translation
seem quite obvious at first sight - and a back translation is a shock to everybody
56 Teaching Translation and Interpreting 2

who has read the original - it is far from easy to analyse the profound
differentness of the original. The analysis must move beyond a contrastive scru­
tiny and this calls for cultural studies and for analysis of all the culture-specific
features lost in translation.

Discussion
The reader's ignorance of the specific socio-cultural context of foreign literary
texts and of their textual repertoire usually leads to misperception and assimila­
tion of the different and unknown instead of an enriching awareness of the un­
familiar and incomprehensible. That is why it is easier to uncover assimilation
at work in a translation than to notice one's own tendency to assimilate the un­
familiar textual elements, for the simple reason that assimilation is part and
parcel of the reading process. Contrastive analysis, however, provides an efficient
tool for illuminating the lost otherness of texts, especially in cases where as­
similation is unnecessary. The process of illuminating the differences between
source and target culture is of paramount importance in enabling students of
translation to form a more complete notion of the text, since the differences can
be analysed only after they have been made visible. Once the differentness of the
text has been identified as such, it can be related to and compared with the
student's own culture. The comparison with one's own culture also leads to
seeing the differentness of the other culture and so opens the ways of learning
about it, or rather perceiving its presence in a foreign text.
Illuminating the different also promotes a more complex appreciation of a text
in which students can simultaneously view it from two perspectives, namely their
own perspective as well as that of the other and the different. The two perspect­
ives can have a dialogical relationship and yet remain separate, thus combining
the effort to enrich the reading and to enhance the awareness of the otherness of
the other without losing one's own perspective. Without illuminating, the culture-
specific differences may remain unavailable for inspection and thus subject to
suppression by assimilation or misperception. Illuminated differences are easier
to inspect and to manage adequately.
By making students aware of subtle translational shifts which are invisible from
the usual post-reading perspective, contrastive analysis of the original and the
translation can help students of translation to make finer cross-cultural distinc­
tions. Having discovered the differentness and having learnt to tolerate it, they
will not be disturbed by it. They also come to know the precarious nature of their
own reality and their reader expectations. By virtue of this thorough under­
standing of the foreign text they are in a better position to decide more com­
petently what to retain, what to adapt, and how to handle the foreign and
 Meta Grosman, Slovenia 57

unknown in relation to their own culture. Cross-cultural awareness acquired in

this way can enable the students to view each translation - including their own -
as a context-bound reading, reflecting the peculiarities of its time and culture.

Notes
1. Austen, Jane. 1968. Prevzetnost in pristranost. Translated by Majda Stanovnik. Ljubljana:
Cankarjeva zalozba.
2. The references are to David Lodge. 1985a. Small World. Harmondsworth: Penguin. The Italian
translation is: David Lodge. 1990. Il professore va al congresso. Milano: Bompiani. And the
German translation is: David Lodge. 1985b. Schnitzeljagd. Munich: Paul List Verlag.
I would like to thank Professor Tjasa Miklic for calling my attention to the translations of David
Lodge's novels.
0sterlars Kirke
 TRANSLATION
AS A PROCESS OF LINGUISTIC AND CULTURAL ADAPTATION

Christiane Nord
University of Heidelberg, Germany, and University of Vienna, Austria

Introduction
In this article, I would like to elaborate on the relationship between translation
and adaptation from a functional point of view in order to show that a strict de­
limitation of translation proper as against adaptation does not lead us anywhere
neither in the practice nor in the teaching of professional translation (including
the translation of literary texts). Since professionality is what we are (or should
be) training our students for, this new perspective is bound to have consequences
for the training methods used at the academic institutions for teaching translation
and interpreting.
To illustrate my approach, I have chosen brief examples, particularly book
titles, which can be considered as paradigmatic texts, exhibiting 'in a nutshell'
all the functional aspects which are distributed over many pages in other texts or
text types (Nord 1993). Therefore, they are particularly apt for classroom dis­
cussions on functional translation problems.

Equivalence and adaptation

The concept of 'translation proper' hinges on equivalence, and equivalence was
introduced into translation studies parting from the standpoint of linguistics in the
early sixties by Nida (1964) and Catford (1965) in the English-speaking area, and
adopted, later on, as 'Aquivalenz' by Kade (1968) in the former GDR and Wilss
(1977) and Koller (1979) in the Federal Republic. As Mary Snell-Hornby has
made clear (Snell-Hornby 1988: 15), equivalence is not 'equivalent' to Aqui-
valenz (nor to equivalence or equivalencia, for that matter); therefore I shall
focus on the German-speaking authors and base the following considerations on
the concept of equivalence in the German sense.
Purely linguistic equivalence between elements of the language system, which
was the main concern of the "stylistique comparée" in the late fifties (see Vinay
and Darbelnet 1958, for example), very soon proved inadequate as a criterion for
the translation of more complex structures, such as texts. Thus, pragmatic aspects
were included in the model. In order to be 'equivalent' to the source text, a target
text then has to fulfil various requirements with respect to each level of textu-
60 Teaching Translation and Interpreting 2

ality, which I resume as follows:

a) As far as pragmatics are concerned, an equivalent target text has to have the same function
or communicative effect as the source text, being addressed to the 'same' group of recipients.
b) With respect to the linguistic dimension, an equivalent target text has to imitate or 'mirror'
the stylistic features of the source text.
c) As regards the semantic dimension, an equivalent target text has to convey the same
'meaning' or 'message' as the source text.
In his recent book on translation skills, Wilss (1992) wonders why precisely
the defenders of the functional approach want to 'do away' with equivalence. In
his view,
source text and target text can only be related to each other by some form or other of a
microtextual and macrotextual equivalence relationship, which is given even in those cases
where a text is not actually translated but - e.g. in the sphere of literature - "transplanted"
(Wilss 1992: 197. My translation and italics).
Maybe it is just this "some form or other" that has made functionalists look for
an alternative concept. Koller (1993) is even less explicit. He presents the follow­
ing example (without stating context or source):
A. Finland Finnland Finlande
Kartor Cartes Karten Maps
Karta over Finland och Helsingfors
Map of Finland and Helsinki
Karte über Finnland und Helsinki
Carte de Finlande et d'Helsinki

B. Det finns i Finland.

Finland - naturally.
Finnland - das Erlebnis.
Finlande - naturellement vôtre.
Koller claims that the fundamental difference between the versions in A and
B is "evident". In A, he says, there is a "translational (that is, equivalence) re­
lationship" between the Swedish originals and the English, German and French
versions, whereas in B, the English, German and French versions are "rather in­
dependent text productions" which do not have much in common with the orig­
inal (Koller 1993: 54. My translation).
I do not think that this is evident at all. In A, I detect a formal linguistic equi­
valence between Swedish over and German über (which, by the way, is not ac­
ceptable in connection with Karte) and between Karta, Map, Carte and Karte
(where in connection with the name of a town Stadtplan or plan would have been
correct in the German and the French version, respectively).
In B, the anaphoric pronoun det indicates that the sentence is part of a text and/
or a combination of text and photographs and that it is probably referring to the
natural beauty of the Finnish countryside. The repetition of the syllable fin(n)
produces a particular metrical effect which makes the phrase sound like a slogan.
The 'equivalent' translations, that is, the corresponding formulations in the target
 Christiane Nord, Germany and Austria 61

language system (This is what you find in Finnland, Das gibt es in Finnland or
Cela se trouve en Finlande) would definitely not achieve such an effect. There­
fore, the translators tried to find another solution: The English and the French
versions 'play' on the ambiguity of natural and naturel, whereas the German ver­
sion uses a rhythmical device which suggests an identification of Finnland and
ein Erlebnis ("a unique experience"). From a functional point of view, these ver­
sions are 'translations' all right: instead of the linguistic form, they 'translate' the
(intended) function of the slogan, adapting the form to the conventions of the re­
spective target culture in order to make sure that the target recipients will be able
to recognize the functional markers of the text type.
As it seems, this kind of adaptation is not accepted in Roller's equivalence
model although it actually leads to a functionally equivalent target text which
would perfectly conform to Nida's postulate of "dynamic equivalence" (1964).
Thus, Koller even gives up the progress achieved by the adoption of pragma-lin­
guistic aspects in translation theory. He allows "text-producing" procedures (in­
stead of text-'re'producing procedures) only ad hoc and in special cases, where
they are limited to a particular text element and intended "to transmit implicit
source text values or to increase text comprehensibility for the target text reader-
ship" (Koller 1993: 53. My translation). It is interesting to note that these pro­
cedures are usually more readily accepted in non-literary than in literary trans­
lations where a 'faithful' rendering of the formal or surface characteristics of the
source text is preferred.
The problem is that these conditions do not cover the wide range of adjust­
ments which may become necessary in any translation process, nor do they ac­
count for those cases where implicit source text values should remain implicit or
where comprehensibility is not the aim of the translation at all. There are texts,
for example, where it is precisely the lack of (general) comprehensibility which
determines text function, as in hermetic poetry or in certain legal texts. Moreover,
this restricted view on adaptation as something incompatible with translation
leads to 'untranslatability' in quite a number of cases.
Let us look at an example again:
Alan Bates: Fair Stood the Wind from France.
Peter Howell: Wuthering Depths.
These titles contain intertextual allusions to a line from Shakespeare's drama
Henry V and to the title of Emily Brontë's novel Wuthering Heights. Intertextu-
ality in its various functions can only 'work' within a cultural or literary system
(see Nord 1993: 200). Even when the works in question have been translated into
a target language or belong to the so-called "world literature", an allusion to a
translated text would never be able to produce the 'same' communicative effect
62 Teaching Translation and Interpreting 2

as a quotation from a work that belongs to the literary canon of the readers' own
culture. Therefore, these titles - as well as most plays on words - are "untranslat­
able" within the framework of a strict equivalence model.

Functionality and adaptation

A functional approach would solve these (and other) problems both in terms
of theory and translator training. Looking at the two titles mentioned above, we
will ask what kind of function the intertextual allusion is intended to achieve. Is
it meant to inform the readers about the contents or the stylistic features of the
book? (These would be referential functions.) Is it intended to appeal to the
readers' aesthetic sensitivity, attract them to buy (and read) the book and/or guide
their interpretation of the text? (These would be three different varieties of the
appellative function.) At any rate, the two titles are designed to distinguish the
books from any other existing ones. (This is the distinctive function which is
common to book titles and proper names.) After finding out the (intended or
possible) functions of the source title, the translator would have to decide which
of these functions, and in what hierarchical order, could or should be aimed at
by the formulation of the target title. And this hierarchy of (intended) target func­
tions sets the guidelines for the translation process.
In the light of these considerations, the translation of Aldous Huxley's Brave
New World (another Shakespeare quotation) by Le meilleur des mondes (a quo­
tation from Voltaire) is no 'better' or 'more equivalent' than the translation of
Marcel Proust's À la recherche du temps perdu (which is no quotation) by
Remembrance of Things Past (Shakespeare again) simply because it gives a tar­
get-literature quotation for a source-literature quotation. Any translation criticism
would have to take into account that perhaps the poetic function of the Proust
title is more easily achieved in the target culture by alluding to a poetic text than
by merely reproducing the semantic content or the syntactic structure of the
original. And these considerations can easily be applied to any other problems
arising in the translation of other text types which are of more frequent use in
translator training.
In this vein, the framework of a functional approach would therefore allow any
transfer procedure which leads to a functional target text, that is, cultural adapt­
ation, paraphrase, expansion, reduction, modulation, transposition, substitution,
loanword, caique, literal translation or even omission (see also Hermans 1991:
166). It is the aim of the translation, the skopos as Vermeer (1978) terms it (see
also Reiss and Vermeer 1984) or the "application or use the translation is
intended to have in the context of the target situation" (Roberts 1992: 7) which
determines the transfer methods used in the translation process. It is interesting,
 Christiane Nord, Germany and Austria 63

by the way, that this idea has been taken up by Hewson and Martin (1991), al­
though the authors never refer to any of the German skopos theoreticians.
The skopos rule is a very general rule which does not account for specific con­
ventions prevalent in a particular culture community. It might even be para­
phrased as "The end justifies the means", and this would indeed mean that the
translator is free to choose any translation skopos for a particular source text. The
principle of loyalty, which I have introduced into the functional approach (Nord
1991 and 1992), sets limits to the variety of possible translation skopoi, obliging
the translator to consider the author's communicative intention(s) and the readers'
expectations towards a text marked as a 'translation'. As a principle, loyalty is
not specified in the general model; what it really means for the translator in a
particular translation task is specified by the culture-specific translational con­
ventions prevailing in the culture-communities involved. Translational convent­
ions which ask for 'literal translation' have to be taken into account as seriously
as translational conventions which allow an adaptation of some or all text di­
mensions to target-culture standards.
'Taking into account' does not mean, however, that the translator has to follow
the conventions in any possible case; conventions can be counteracted or even
changed. But the translator cannot just 'override' the expectations of his or her
partners in the cooperative activities implied in any intercultural text transfer.

Source-text functions and target-text functions

Parting from a functional analysis, the translator tries to find out the function(s)
which can or would be achieved in the source-culture situation by the source text
as a whole and/or by any separable in-texts (such as examples, metaphors and
similes, quotations, plays on words, etc.), comparing them to the function(s)
required for the target text. If there is any possibility of achieving the same
function(s) by the target text in its (prospective) target-culture situation, the
translator would be free to decide on the transfer procedures which may become
necessary, and adaptation is one of them. Let us take a book title as an example
again:
Simone de Beauvoir: Une mort très douce
German translation: Ein sanfter Tod;
English translation: A Very Easy Death.
Beside the phatic, the metatextual and the distinctive functions which are com­
mon to all titles (see Nord 1993), this book title seems to be intended to achieve
an emotive function by the elements tres douce, which is reinforced by the soft
dark sound produced by the phrase. The German translation is apt to achieve this
function precisely by avoiding both the translation of douce by its literal equi­
valent süβ (which has connotations of liveliness and happiness) and the repro-
64 Teaching Translation and Interpreting 2

duction of the adverb tres by its systematic equivalent sehr, which would indeed
lead to a sharp, hissing sound: *Ein sehr sanfter Tod or even *Ein sehr süβer
Tod.
The English version, however, substitutes the emotional function of douce by
the evaluative function of easy. Whether a death is easy or not can be judged
from outside, from the doctor's point of view, whereas douce describes the feel­
ings of the dying person. Thus, the English title sounds very matter-of-fact
(which is in part also due to the nominality of the phrase) and would probably
not achieve the same emotive function as the original. Another adjective and an
adjustment or adaptation of the syntax structure might have led to a functionally
more adequate title like, for example, *Dying softly/gently or perhaps *Dying is
sweet. This would also be in line with formal conventions: An empirical analysis
of approximately 1600 German and 1800 English book titles has shown that both
verb patterns and sentence patterns are much more frequent in English fiction
(1,5% and 6,5%, respectively) than in German fiction (0,5% and 3,5%) or French
fiction (0,3% and 4,8%) (Nord 1993: 60).
But text functions need not necessarily be the same for the source and the tar­
get text. For French readers, the Beauvoir title does not contain any explicit or
implicit reference to the fact that the book is a fictional text; for them, this
information is implied by the author's name, who is known as a writer of fiction.
For English readers, this may not be as evident. Therefore, it might be wise to
use a sentence pattern and not a verb pattern, because in English book titles, verb
patterns are more frequent in non-fiction (4,4%, for example: Living with Alz­
heimer's Disease) than in fiction (1,5%) (Nord 1993: 60).
As we see, there is a shift of functions in the target title in this case. That is,
the information about the genre which in the source culture is given by the
author's name is shifted to the title in the target-language formulation.
But there may also be cases where the target text has to achieve a function
which is not vital for the source text (or vice versa). Again, we can take titles as
a case in point If the author is a famous writer in the source culture, but not
known (as yet) in the target culture, the original title does not need to achieve an
appellative function, whereas the translated title would have to attract the pro­
spective readers' attention. This may be a motive for the references to an 'exotic'
setting (very popular as an appellative feature in German fiction: 5%) introduced
in the German translations of the following titles of two novels by the Cuban
author Alejo Carpentier.
Alejo Carpentier: Los pasos perdidos [literal translation: Lost Steps]; German version: Die
Fluent nach Manoa;
Alejo Carpentier: El acoso (literal translation: The Siege); German version: Finale auf Kuba
(see Nord 1993: 240).
I will not discuss here whether the suggested versions have been the best or the
 Christiane Nord> Germany and Austria 65

only way to achieve an appellative function in the target titles; what I would like
to stress, however, is the fact that adaptations often are the only way to ensure
that a translation 'works' in the target-culture situation it is produced for.

The functional approach in translation teaching

Choosing the functional approach for translation practice will have considerable
impact on translation teaching or translator training.
a) The linguistic features of any text are determined by the situation the text
is used in. In 'normal' intralingual communication, we know the situation in
which and for which we produce an utterance or a text. In the traditional trans­
lation class, however, teachers often ask the students to translate the source text
'as such', that is, without specifying the situation or purpose the translation is
needed for. Therefore, trainee translators commit grammatical mistakes even in
their own native language, which they never would have made in spontaneous
intralingual communication.
Experience shows that when the prospective communicative situation is clearly
defined, linguistic errors are committed less frequently. Therefore, a commission
or assignment which defines the intended function or functions of the target text
can be expected to reduce the number of linguistic errors or faults in students'
translations.
b) By contrasting the target situation (especially with regard to the prospective
recipients and the intended text function or functions) described in the translation
assignment with the functional analysis of the source text in its own communica­
tive situation, translation problems can be detected in advance (Nord 1992). This
procedure makes it easier for the trainee translator to develop translation stra­
tegies for the solution of a particular translation problem which are designed for
the translation of the whole text and not for individual units such as words or
phrases.
c) Translation strategies should follow a 'top-down' procedure:

(1) Text function - in - situation

(2) Cultural norms and conventions

(3) Linguistic structures

(4) Context

(5) Trans-
lator
66 Teaching Translation and Interpreting 2

This means: in a first step, a particular translation problem (for instance, a play
on words) is analysed with regard to its function in the text and in the target
situation (1). The analysis leads to a decision whether the translation has to be
adapted to target-culture norms and conventions or whether it should reproduce
source-culture conventions used in the source text (2). This decision sets limits
to the range of linguistic means to be used (3), from which the translator chooses
the one which fits into the specific context, such as text-type, register, style, etc.
(4). If there is still a choice between various means, the translator may decide
according to individual stylistic preferences (5).
The degree of difficulty of a translation task can then be judged according to
the form of the triangle. The lesser the constraints set by situation or convention,
the larger will be the margin for individual preference decisions. In this case, it
will be less easy to find plausible and intersubjective reasons for their func­
tionality and consistency, which makes the task more difficult for the trainee.
d) The functional approach is consistent: unlike the equivalence model, it does
not suggest different norms for literary and non-literary translation. Therefore, it
can be applied to the translation of every kind of text and between every pair of
languages and cultures. Thus, it is more appropriate as a framework for metho­
dological principles and strategies and would make translator training more ra­
tional and even more 'economical' as well as more independent of language and
culture specific peculiarities.
e) The functional approach is an 'inclusive' model: Unlike the equivalence
model, which considers certain forms of intercultural text transfer as 'non-trans­
lations', the functional approach includes all forms as long as they are functional
and based on the loyalty principle. Thus, even the production of (a particular
form of) equivalence in the target text can be one possible translation skopos,
which might form the last step in a teaching progression leading from easier to
more difficult tasks. In the initiating phases of teaching, however, it would be
wise to start by translating strongly conventionalized texts with clear functions,
such as instructions or tourist brochures, for which existing models and parallel
texts can be found in the target culture.
f) The functional model accounts for all translation tasks a translator will be
confronted with in the professional practice of translation. 'Realistic' translation
assignments, which prepare the students for what expects them in 'real life', pro­
vide a good motivation for teaching and learning. It is simply more fun to work
on texts which ask for professionality than on newspaper reports which are hardly
ever translated in practice. And fun spells learning success.
 Christiane Nord, Germany and Austria 67

Concluding remark
In conclusion, I would like to say that my own experience in using the func­
tional approach in translator training shows that the trainees commit less gram­
matical and pragmatic mistakes, they learn a lot about cultures and conventions,
and they seem to have quite a bit of fun doing all this!
 TRANSLATION AS A MEANS
FOR A BETTER UNDERSTANDING BETWEEN CULTURES?

Heidrun Witte, Universidad de Las Palmas de Gran Canaria, Spain

Theoretical bases
Modern functional translation theory has defined translation and interpreting
as complex communicative action, the main aim of which is to establish com­
munication between members of different cultures in accordance with a pre­
viously determined communicative target purpose or Skopos, as Vermeer termed
it (Vermeer 1983; also Reiss and Vermeer 1984).
Culture is understood as the sum total of a social community's behaviour
patterns, including the 'rules' of behaviour and its (material and immaterial)
'results' (Gohring 1978 and 1980).
Functional translation theory also distinguishes between paraculture (that is,
the culture of a society as a whole), diaculture (namely subgroups within a social
community), and idioculture (which is individual). The definition of culture refers
to a dynamic concept, so that the delimitation of any cultural unit depends in
each case on the perspective of the observer/interactant (group of observers/
interactants) and his/their respective purpose of analysis (see Vermeer 1992a).
In accordance with the functional approach, in the following the translator
(throughout this paper "translator" will refer to both translators and interpreters)
is considered an expert in cross-cultural communication and his bicultural com-
petence regarded as a basic prerequisite for his work.
To enable students to develop an adequate bicultural competence, translation
training must emphasise the different roles involved in the translation process.
This implies the necessity of stressing the ignorance most clients will have of the
foreign culture and their unawareness of the problems of intercultural communi­
cation in general.
The most comprehensive functional approaches in translation theory have so
far been developed by Hans J. Vermeer (1983) and Justa Holz-Mänttäri (1984).
The following discusses some of the basic arguments of their approaches, that
is "Scopos Theory" (Skopostheorie) and "Theory of Translational Action"
(Theorie vom Translatorischen Handeln), respectively, with regard to pedagogical
implications.
70 Teaching Translation and Interpreting 2

The concept of communication

Before discussing some of the problems arising from intercultural communica­
tion, we will examine the concept of communication as such.
Social psychology, communication studies and discourse analysis have demon­
strated that any communicative act necessarily involves various 'levels', on which
'communicative messages' are 'exchanged' (Gumperz 1982; see also Poyatos
1983).
Methodologically speaking, there are three different levels in communication
(Witte 1992a):
(1) the object level, which refers to the 'contents' of what is being communi­
cated;
(2) the communicative level, which refers to the intention/purpose behind the
communicative act; and
(3) the personal/interaction level by which we refer to the idea the interactants
gain of each other and to the relationship that evolves between them in the
course of the interaction. (This level is usually termed "metacommunic-
ative". See, for example, Kendon 1981; 1982.)
When people communicate, they tend to be conscious of levels 1 and 2: you
want to tell somebody something and - if prompted - you will usually be able to
explain why and for what purpose.
As far as level 3 is concerned, however, people are usually not aware that they
are constantly communicating to each other a certain 'image' of themselves and
of their interpretation of the interaction situation (including their interpretation of
the interaction partner).
EXAMPLE:
In "Western" business negotiations interactants tend to concentrate either on the subject matter
at hand or on the result to be achieved, such as the signing of a contract. They do not,
however, seem to be equally concerned with one another at the personal level, nor to be
conscious of how their behaviour is interpreted by the other party. Nevertheless, in the end,
each party will actually have acquired a (usually subconscious) idea of the other which may
range from sympathy to a feeling of "what a strange guy" (see e.g. Reuter, Schroder and
Tiittula 1989).

Naive intercultural communication

In intercultural communication the problem becomes more complex: culture
shock theory has taught us that people who do not have previous 'knowledge of
the foreign culture they deal with in specific intercultural situations, tend to
project their own cultural frame of reference onto the foreign culture. That is,
they interpret and evaluate foreign behaviour in accordance with their own cul­
tural rules of conduct, and act according to the behaviour patterns of their own
culture (e.g. Furnham and Bochner 1986; see also Poyatos 1983).
 Heidrun Witte, Spain 71

This 'projecting' attitude constitutes an inevitable, indeed necessary process in

cognition in general: we would not be able to assimilate new experiences if we
perceived them as completely new. We have to be able to relate them ('to com­
pare' them) to categories we already know (Detweiler 1980; see also Lakoff
1987).
In turn, these categories are learned in a process of socialization, through which
individuals gain active and passive knowledge of the behaviour conventions
("rules of conduct") operative in their respective social communities.
Once individuals have gone through this socialization process, they will find
it extremely difficult to 'free' themselves of the result: they will always 'see the
world' through the 'filter' of their own culture.
This is no plea for cultural relativism (understood as determinism) but merely
a reference to sociology's well-documented claim that the decisive influence
exerted on the individual by primary socialization cannot be repeated with any
comparable force in secondary socialization(s) (Berger and Luckmann 1989).
It is thus inevitable that we get to know other cultures 'on the basis of our
own (primary) culture: in intercultural contact, people will necessarily 'compare'
foreign cultural phenomena to their own cultural background (see Muller 1980;
and 1986).
However, the more one learns about another culture the less one will turn to
one's own culture for comparison. Instead, one will become increasingly more
able to use the foreign culture itself as a frame of reference, that is, perceive and
interpret it in terms of itself, although (methodologically speaking) one's own
culture will always remain the 'deepest' level of comparison (Witte 1987).
Therefore, the idea people form about a foreign culture will usually not cor­
respond to the image the culture would claim for itself. Nevertheless, such an
idea generally does exist in people's minds and will necessarily influence their
behaviour in concrete cross-cultural situations.
To sum up: in intercultural communication, the everday 'neglect' of the inter­
action level, as described above, becomes aggravated in that it is precisely on this
level where most unconscious culture-specific projection takes place.

The translator's bicultural competence

The translator thus deals with (at least) two persons from different cultures who
have preconceived ideas about each other, ideas which, in general, are not 'ade­
quate' in so far as they do not correspond to the respective foreign culture's
'self-image'.
Therefore, the translator's bicultural competence cannot merely consist of a
competence in the cultures he works with but must also comprise what we have
72 Teaching Translation and Interpreting 2

elsewhere called a "competence between them" (Witte 1987): The translator must
be able to judge/estimate (from his perspective) the clients' 'knowledge' (or 'lack
of knowledge') of one another's culture and to anticipate the impact this know­
ledge may have on behaviour patterns in the concrete cross-cultural situation.
It is only by interrelating the two cultures that the translator is in a position to
compensate for the clients' 'inappropriate' preconceptions and projections as well
as the 'inadequate' active behaviour patterns they may lead to.
We have already discussed how communication necessarily involves the "inter­
action" level. It follows that any professional concept of communication will have
to include not only the contents and intention of the message but also the
participants in the communicative situation. This means that the translator must
operate with a much 'broader' concept of communication than the 'restricted'
everyday notion held by his clients. If we claim that translation and interpreting
are to establish intercultural communication professionally, then translational
action must be based on professional concepts (Holz-Mänttäri 1984 and 1986).

Towards better understanding between cultures

"To achieve a better understanding between cultures" translators must, then,
take into account all three levels of communication and pay more attention to the
interaction level than it has received so far, in order to avoid culture-specific
projection as far as possible. In other words, they should strive to transmit an
image of the source culture to the target receptors that corresponds to the image
the source culture would claim for itself (The "image a culture claims for itself,
however, can be distinguished further as (1) its self-image; (2) the image of itself
with regard to other cultures, that is, the image it believes/wants to evoke in
another culture.)
The practical realization of any such attempt would first of all presuppose that
the client saw the relevance of the interaction level for satisfactory cross-cultural
communication. But since the client is not an expert in intercultural communica­
tion, he is usually not aware of its problems. Judged from the translator's
professional perspective, the client's purpose/scopos is normally rather 'restricted'
(Vermeer 1989).
In order to achieve successful cross-cultural communication as defined from
a professional viewpoint, the translator may therefore consider it necessary to
make the client modify his original purpose.
There are, of course, practical difficulties in any such attempt in our society
here and now. As has been pointed out by Honig (1992), the translation market
is still much more willing to accept bad translations than self-confident trans­
lators. Nevertheless, professional performance requires professional concepts.
 Heidrun Witte, Spain 73

Social recognition and appreciation of the translator's work cannot be achieved

unless the translator himself recognizes and insists upon his social responsibility
(Vermeer 1990; Robinson 1991).
So far, our discussion has been focused mainly on face-to-face interaction.
As far as written translation is concerned, the aim to transmit an image of the
source culture as close as possible to its self-image can certainly not be achieved
successfully by the still predominant translation strategy for literary texts.
The vast majority of literary texts today are translated according to what
Christiane Nord (1991) calls "exoticizing" translation strategies.
These "exoticizing" translations certainly meet an existing demand. However,
the way they are usually done does not transmit anything like the self-image of
the source cultures, but rather fosters existing stereotypes and prejudices among
the target public.
Since there is a culture-specific interrelation between behavioural phenomena
and the values attributed to them within a society, perceiving and interpreting be­
haviour also implies evaluating it. The mental image we form of a social pheno­
menon, is thus necessarily composed of denotative and connotative elements (see
Vermeer and Witte 1990; also Vermeer 1992b).
However, literary translation still favours linguistic surface structures, without
bearing in mind the value attributed to certain source-cultural behaviour patterns
in the target culture.
EXAMPLE:
In German translations of Spanish novels one frequently comes across Spanish housewives
having their morning chats in bars, workers drinking three glasses of wine at eleven in the
morning and little children running about in the streets long after midnight.
In Germany a bar is a rather obscure place where you go to drink, possibly even get drunk; a
German wine glass is three times the size of a Spanish one; German parents send their small
children to bed at seven or eight in the evening.
A German receptor without previous experience of Spanish culture will therefore be greatly
surprised (to say the least) by such phenomena and may well question Spanish domestic order,
working morals or educational responsibility. (See Stackelberg and Kroeber 1986 for a similar
argument; also Poyatos 1983.)
"Exoticization" does not necessarily lead to negative effects in the target cul­
ture. And "exoticization" may also be a legitimate translation scopos (that is, an
intentionally pursued strategy).
However, translation procedures which transfer behavioral phenomena from the
source to the target culture by merely reproducing linguistic surface structures
generally 'bring about' exoticizing effects more or less unconsciously (and, ac­
cordingly, should not be termed "strategies").
Such translations generally fail to anticipate the ground on which the target text
will fall, that is, they 'forget' about the target receptor and his culture-specific
background. This is why, in our view, such translations do not serve to develop
74 Teaching Translation and Interpreting 2

an "understanding" between cultures. They cannot be described as representing

responsible translational action either.

Teaching
In terms of pedagogical considerations there are two main points about the
problem of intercultural communication:
Firstly, students have to become aware of the potential difficulties of inter-
cultural communication.
Elsewhere (Witte 199ba), we have therefore insisted upon the necessity of es­
tablishing the subject "intercultural communication" as a special topic in trans­
lation training. Here, students should be introduced to the ways in which social­
ization and communication work within one culture, in order to learn, in a second
step, to see the difficulties likely to turn up in cross-cultural contact because of
existing behaviour differences.
Secondly, students have to be made aware of their clients' 'unawareness' of the
problems of intercultural communication and of their usually inadequate
cross-cultural knowledge.
Once students have learned something about their working cultures and have
developed a certain cross-cultural knowledge, they tend to take at least some
degree of cross-cultural sensitivity for granted in everyone. They usually find it
hard to imagine that the majority of their future clients will have no special
training in problems of intercultural communication and will therefore simply be
incapable of distancing themselves from their own cultural background.
In addition to an insufficient capacity to adopt the perspective of a 'naive'
receptor, students sometimes seem to confuse the concept of "taking into account
the target receptor" with the application of concrete translation strategies like
"simplifying the target text" or "adapting it to the target culture".
It is essential that students are made to understand that bearing in mind the
target situation does not in itself entail a specific translation strategy, but first
of all it means anticipating the possible effects different translation alternatives
may have upon the target receptor.
However, our stress on the relevance of culture differences must not lead to a
static idea of culture specificity with the students and should therefore be backed
up with discussion of the relativity of 'detecting' culture-specific traits. Although
the translator may acquire a relatively high degree of "biculturality", he will, like
everybody else, never be able to loosen himself completely from his primary
culture. To a certain degree, he will therefore always perceive foreign cultures
from his own 'culture-bound' perspective.
No attempt 'to reach' foreign cultures can be more than an approximation.
 Heidrun Witte, Spain 75

Students must be taught to regard this 'never-ending' endeavour of approaching

foreign cultures not as a vain attempt but as an integral part of their future
profession. Only then will they recognize the potential inherent in translation as
a means to achieve a better understanding between cultures.
 ADVERTISEMENTS IN TRANSLATION TRAINING

Gabriele Becker, Universidad de Las Palmas de Gran Canaria, Spain

Introduction
In this article, we shall discuss some results of a recently initiated project that
investigates not only the use of newspaper and magazine advertisements in trans­
lation training, but also the communication and translation strategies behind them.
Our objective is to analyse advertisements of different cultures in class and to
demonstrate the enormous importance cross-cultural knowledge has for the trans­
lator. We intend to sensitize students to the fact that advertisements may include
specific aspects of a source culture, which must be taken into consideration when
translating for a target culture.

General objectives
The general objectives are that students
L become acquainted with other cultures through advertisements, and
2. learn about translation-orientated text analysis and text production
To reach these objectives, we have subdivided our material into 3 categories:
A. Advertisements involving the "home" culture as self portrayal or "inner eye
vision". (German and Spanish examples)
B. Advertisements involving "a different/foreign" culture seen through the
"spectator's eye".
a) Advertisements in Spanish of a German product in Spain
b) Advertisements in German of a Spanish product in Germany
C. Advertisements of the same product in both languages and cultures, with the
same layout.

The material
The analysis of different advertisement material includes:
1. a linguistic and functional text analysis (following Nord's Model for Trans­
lation-Oriented Text Analysis (1991));
2. an analysis of non-verbal means;
3. an analysis of the interrelationship between illustration, text, product and cul­
tural frame;
4. a cross-cultural analysis of verbal and non-verbal means and communication
strategies; and
78 Teaching Translation and Interpreting 2

5. an analysis of underlying translation strategies.

Advertising and translation

Commercial advertising is a form of propaganda, in so far as "propaganda" is
defined as any attempt to influence the development or change of attitudes of an
audience (McDavid and Harari 1967: 371).
Advertisements are normally produced to sell or to announce something, mostly
by persuasive arguments, so that they influence attitudes and/or behaviour. When
somebody reads, listens to or watches advertising information, this commu­
nication is communicated to the receptor. Therefore, advertising forms part of hu­
man communication, by means of multimedial systems. Commercial advertising
is used in the mass-media, for instance, TV, films, radio, newspapers or ma­
gazines. Advertisements are thus received via visual, aural or written channels.
Receiving information involves decoding and conversion into already existing
information patterns.
In focusing on newspaper and magazine advertisements, we deal with verbal
written information and non-verbal illustrations and their interrelationship.
Analysing texts automatically involves language; and language is "a system of
social conventions specifying relationships between certain symbols (either verbal
ones or graphic ones) and certain ideas or concepts. As such, language is a cul-
tural product that influences the structure and organization of the individual's
thought. ... The greatest social significance of language is that it permits the
communication of ideas and experiences from one individual to another"
(McDavid and Harari 1967: 171).
This social, or rather "cultural" component of language and of any communic­
ative process, is most important for a discussion of advertising: advertising as a
means of communication is at the same time an instrument for transmitting
culture.
Following the text typology by Katharina Reiss (1983), advertisements can be
described as operative text types. Their main communicative function is to
appeal, trying to change the reader's opinion or behaviour, provoking impulses,
reactions and actions, producing a persuasive effect through language and illu­
strations.
Newmark argues that the vocative function of language is used in advertise­
ments "in the sense of calling upon the readership to act, think or feel, in fact to
"react" in the way intended by the text" (1988: 41).
If the primary communicative function of advertising is to appeal, the per­
suasive effect of language attempts to influence attitudes or behaviour, and if lan­
guage is considered a cultural product, then advertising may be said to reflect in
 Gabriele Becker, Spain 79

some way or other the cultural background of a community. And different com­
munities with different cultural backgrounds need different stimuli to provoke the
same reaction-response.
Consequently, if the sender has any special intention as to how the addressee
should receive and decode the information, the transmitter of the information
must take into account the social conventions of the addressee's community in
order to be successful.
When we speak of "different communities", we do not refer only to different
language communities: all over the world we can observe regional differences of
habits, customs or social behaviour, even within the same society. For example
one needs different advertising strategies and campaigns to promote the drinking
of tea in the north of Germany than one does in the south, independent of the
language.
Of course, it is even more important to take into account cross-cultural be­
haviour differences between communities with different language systems.
Recent translation theory has repeatedly emphasized the importance of cross-
cultural knowledge for the translator. Translation has to take into account verbal
and non-verbal communication strategies. As mediators between cultures, trans­
lators are considered representatives of different cultures, not only because of
their linguistic abilities, but also because of their knowledge of culture-specific
behaviour patterns in general.
Teachers often describe the difficulties they have in making students aware of
culture-specificity and its relevance for translation: to teach what we call "life
and civilization" is quite difficult and even more so if we think about which
concrete aspects we have to take into consideration in translation training.
Introducing cross-cultural analysis of advertisements as a pedagogical in­
strument in preparatory translation classes seems to be a method of drawing the
students' attention - through inductive procedure - to cross-cultural behaviour
differences and communication difficulties potentially resulting from them: if we
introduce advertisements from two different language communities through com­
parative studies, students must become aware of the importance of cross-cultural
knowledge.

Analyses of different types of advertisements

In the following discussion we present only one example of each type.

Advertisements involving the 'home' culture as self portrayal:

Example 1: DEUTSCHER WEIN - EINZIG UNTER DEN WEINEN
We introduce an advertisement in German about German wines. It looks like
80 Teaching Translation and Interpreting 2

this:

We ask the students to analyse the advertisement and comment on it as

follows:
a) What do you notice about the illustration? (information value, associ­
ations,etc.)
b) Analyse the text (function, slogan, external form, syntax, rhythm, etc.)
c) Analyse the relation between text, photograph and illustration.
We follow Baraduc's theory (1972: 105-115) that text and illustrations belong
to two different systems. The function of the illustrations/photographs is the re­
presentation of reality, and the function of the text is the structuring of that
reality. We therefore first emphasize that both photograph and illustration in the
advertisement presents us with something 'typically' German: namely German
architecture, mostly in the central and southern parts of Germany, as well as
German commercial life in the Middle Ages, represented by the small drawing.
In the analysis of the verbal part of this advertisement we identify the
 Gabriele Becker, Spain 81

following communication strategies:

a) the use of superlatives in the slogan,
b) the almost poetical syntax
c) the folk song titles describing the typical autumn atmosphere in Germany,
and
d) the German customs of keeping wines in stock against the long, cold winter
evenings.
In the interrelationship of text and the illustration/photographs, we recognize
a clear double appellative function: the rational, informative aspect is expressed
in the text and the emotive aspect is expressed through the illustration/photo­
graph.
This advertisement is thus an example of a self portrayal of German culture,
or "inner eye vision".
In more general terms, the didactic advertisement, which is pedagogical in its
own right, illustrates a distancing effect: the same advertisement acquires greater
informative value instead of keeping its appellative function.
When we use examples of this type in translation training, we cover two areas:
firstly, we introduce knowledge about a country and culture, and secondly, we
train students in translation-oriented text analysis and in advertising strategies in
general.
Examples of this type are also useful in translation training because students
become aware of the culture-specific aspects in advertisements.

Advertisements involving a different/foreign culture

The second step is to introduce students to information about cultures in adver­
tisements through "the spectator's eye vision", by looking at advertisements for
German products in Spain and for Spanish products in Germany.
a) First we look at German products and how they are introduced in Spain,
In addition to numerous advertisements which use stereotypes, such as
"German quality" (ZEISS, WMF-cutlery), or "qualified German engineering"
(OPEL), there are other examples which indirectly reflect German culture.
In Example 2: MEICA. LA TENTACIÓN DE LUTERO [Luther's temptation]
(next page), the Spanish addressee is taken back to the 15th century and the
world of Martin Luther in order to be introduced to several types of German
sausage. The text plays with words from Luther's life and refers to his character.
In the practical lessons with our students here we insist on an exhaustive text
analysis in order to make them aware of any detail so that they are forced to look
critically at the culture-specific aspects expressed in this advertisement.
82 Teaching Translation and Interpreting 2

b) Secondly, we analyse Spanish products and how they are advertised in

Germany, by introducing a series of examples, like the one on the opposite page.
Example 3: CARLOS PRIMERO. LA NOCHE - LA VIDA
It is primarily the picture which calls the addressee's attention to the adver­
tisement: it shows a bottle of brandy and a photograph of young Mediterranean
people enjoying themselves in the evening. In the text of this German advertise­
ment, we recognise Spanish words "la noche - la vida", and then there is a
stereotyped description of Spanish nights - throbbing, full of life and warmth.
 Gabriele Becker, Spain

Thus, examples of this type also transmit cultural aspects of a community to

the students, but as seen from abroad, from a spectator's perspective. They often
include the stereotyped views cultures have of one another.

Advertisements of the same product in both languages and cultures

Our pedagogical objectives in the previous examples have been to demonstrate
the existence of direct or indirect cultural information in advertisements and to
sensitize our students to these phenomena.
84 Teaching Translation and Interpreting 2

The subsequent examples of the same advertisements in different languages

focus on the analysis of strategies for text production. Later on, the knowledge
of text production strategies in advertising and the ability to analyse such stra­
tegies may become relevant for the elaboration of translation strategies.
As mentioned, mediators of operative texts must take into account the function
of the target text. This means that it has to be adapted in terms of its appeal to
the mentality and to the general socio-cultural conditions of the addressee in the
target culture.
Our objective in using advertisements in preparatory translation training is
therefore to make students aware of the necessity to take into account the socio-
cultural background of the target culture. Students have to learn that cross-
cultural transmission of operative texts implies adaptation to the socio-cultural
background of the target culture, including, possibly, the production of a new text
in the target language.
In order to reach this objective, we introduce several advertisements in both
languages. In pedagogical progression they are compared with regard to the re­
levant verbal and non-verbal elements.
Example 4: BOSCH: EIGENTLICH WOLLTE ER NUR EIN SCHAUKEL-
PFERDCHEN BAUEN/ TODO EMPEZ6 POR UN BALANCIN
 Gabriele Becker, Spain 85

We present our students with the original German source text. We first analyse
the pragmatic situation factors, that is to say:
- text function (operative text, advertisement, persuasive strategies)
- addressee (the average German citizen)
- intention of the transmitter (BOSCH's aim is to sell the product)
- relationship between text and illustration (the illustration takes up half the
advertisement space, and text and illustration are connected in a way that would
make either incomprehensible if presented in isolation).
Among the culture-specific aspects we emphasize the illustration which shows
a bald German citizen engaged in a "Do-It-Yourself' job in the open air. In Ger­
many this is quite normal; almost everybody does "Do-It-Yourself" jobs around
the house and is proud of it. There is a well equipped toolbox, a drill and saw
in every household.
The text insists on two very important aspects for the German addressee: the
technical innovations and the possibility of working like a professional (since the
German is an amateur).
The average Spanish customers have a totally different attitude and back­
ground: they are not accustomed to doing jobs around the house, since they nor­
mally buy finished products.
In the Spanish text the message is focused on "do it yourself", in order to con­
vince the addressees that this activity produces fun and creativity. The function
of the advertisement is not to enter into technical details - therefore we find
fewer than in the original version - but to motivate the Spanish addressees to
carry out handiwork and convince them that the easiest way is to do it with
Bosch.
By presenting our students with the version translated for the Spanish customer,
and by making the same analysis, we call their attention to the importance of
culture-specific aspects in translation of advertisements.
The examples of this third type help students develop an idea about how to
adapt advertisements to target cultures. They find out about detailed cross-cultural
translation strategies by means of comparison and can then apply their theoretical
knowledge in practice using advertisements.

Concluding remarks
In our discussion of various types of advertisements we have shown that ads
directly or indirectly communicate culture-specific aspects and that this must be
taken into account whenever we deal with advertising for different target cul­
tures.
We have tried to demonstrate that by virtue of this special communicative and
86 Teaching Translation and Interpreting 2

culture-specific character, advertisements may serve as an excellent instrument

for discussing the transmission of culture and, at the same time, constitute an
important pedagogical instrument in preparatory translation training.
TRANSLATION AND CLASS
 KARL POPPER IN THE TRANSLATION CLASS
Andrew Chesterman, University of Helsinki, Finland

One of the problems in translation teaching is how to build a mass of individ­

ual insights into a coherent whole. I have found the philosophy of Karl Popper
to be a real inspiration in this respect. It has provided me with a rich conceptual
framework which is theoretically satisfying and also eminently practical. This ar­
ticle introduces some of the central aspects of this framework.
First, though, a couple of words of introduction. Karl Popper was born in
Vienna in 1902. He studied music, and also worked as an apprentice to a cabinet-
maker for a time, before moving into mathematics and philosophy. He was crit­
ical of the logical positivism of the Vienna Circle, opposed both fascism and
Marxism, and became an enthusiastic Social Democrat. After a period teaching
philosophy in New Zealand, he emigrated to England in 1946, and was Professor
of Logic and Scientific Method at the London School of Economics for many
years. His major work has been in the philosophy of science, but his most recent
publications have been more concerned with cognition and the workings of the
mind. I shall be focussing mainly on some of the ideas in his Objective Knowl-
edge. An Evolutionary Approach (1972).

Problem-solving
Through much of Popper's work there runs a simple thread: the idea that
knowledge advances by means of problem-solving. In particular, the objective
knowledge of science does not start with data, or with theoretical axioms, but
with problems. Popper's view of the scientific method is illustrated in a schema
which he uses again and again in slightly different forms, applied to many areas
of life, from evolution to politics. The schema is:

PI —> TT -> EE -> P2

PI represents an initial problem, which may be of any kind whatsoever. TT

stands for Tentative Theory: this means the first hypothesis, the first attempt to
solve the problem. Given the fallibility of human nature, this tentative solution
is unlikely to be right, or even the best possible. In Popper's view it is immaterial
how you might arrive at this TT - imagination, guesswork, chance, rational de­
duction, whatever. For scientific knowledge, the crucial stage is not the nature of
90 Teaching Translation and Interpreting 2

this first hypothesis but how it is tested: EE stands for Error Elimination. This
refers to the process whereby the TT is tested and refined. At this stage it is the
responsibility of the scientist to submit his Tentative Theory to the most rigorous
testing possible. Since no theory is ever more than a hypothesis, no theory can
ever be proved true, but theories can be shown to be false or inadequate. A
theory that cannot be falsified may or may not be a good theory, but according
to Popper it cannot be an empirical one. Further, the goal of empirical science
is not 'truth' (because absolute truth can never be known) but rather maximal
verisimilitude, or truthlikeness. The result of the EE process is not, therefore, 'the
right answer' or 'a perfect theory', but in fact a new problem, P2. P2 might be
a refined or reformulated version of P1, a consequence of P1, a lack of fit
between TT and the data, something at a more general level that perhaps sub­
sumes P1, or whatever. P2 in turn is then subjected to testing etc., and so the
process goes on. Knowledge is never final.
In translation studies, P1 has taken many forms. At the most general level, P1
can be simply "what is translation/translating?" Many traditional answers to this
question have been metaphorical: translating is giving a performance of an
original; reproducing an original; betraying an original, and so on. Such answers
may bring us insight, but they are not empirical in Popper's terms, since they
cannot be falsified.
Other questions that function as P1 have been: what is the best way to trans­
late? what is equivalence? how can we program a machine to translate? what
counts as a translation, in a given culture at a given time? what goes on in the
translator's head? how can metaphors be translated? And many more. The
progress of translation studies can thus be seen as a vast network of intertwined
TTs that have been set up to answer a huge number of P1s, plus the EE that
takes place within the community of translators and scholars assessing the TTs,
plus the subsequent P2s. In teaching translation studies, or theoretical aspects of
translation generally, it is obviously vital to specify the P1 to which a particular
TT or EE attempt is related; for apparent disagreements between theoretical
views often turn out to be because these views are linked to different initial
problems.

Translation as theory
Perhaps more relevant to the practical translation class is the way that a trans­
lation can itself be seen as a Tentative Theory. After all, the initial problem faced
by a translator is: how shall I translate this text (or: this sentence/ this word/ this
idea ..,)? At multiple levels, all translations are proposed solutions to such a P1,
tentative theories. Different translators, different times, usually come up with
 Andrew Chesterman, Finland 91

different solutions. There is nothing "final" about a translation, insofar as a

translation is merely a theory like any other theory. And just as it makes no sense
(in Popper's terms) to claim 'perfection' for a scientific theory, so there is no
reason why a translation (qua theory) should be 'perfect'. This point of view
suggests that much of the traditional argument about equivalence has been noth­
ing but a confusion of red herrings.
Note that, in Popper's view, it does not matter how a particular translation (=
theory) is arrived at. This argument would run counter to the claim (e.g. by Wilss
and some others) that the evaluation of a translation must take into account the
translation process. On the contrary: in the view I am propagating, the interest
is not in how to get to a TT but in what happens when we have already got one.
(At this point it is mnemonically convenient to point out that TT may denote not
only Tentative Theory but also Target Text, or at least Tentative Target Text.)
What matters is the Error Elimination stage.
An approach along these lines is in fact advocated by Gile (e.g. 1992b, and in
the present volume (pp. 107-113)). In his model, for any element of the source
text, the translator first develops a hypothesis of what it means and how it might
be translated, and then tests this for plausibility and target language acceptability.
Testing involves a wide range of procedures, from documentary research, com­
patibility with background knowledge of the world, assumptions about back­
ground knowledge of the readers, and so on. Any errors or shortcomings are e-
liminated, and alternatives proposed, until the translator is satisfied.

Criticism and corroboration of a translation

The TT can thus represent, initially, the first draft of a translation. In Popper's
schema it then undergoes the process of Error Elimination, criticism, revision. In
the advance of objective knowledge in general, it is this stage that distinguishes
the scientific method from other methods; by analogy, it is this stage that shows
up the difference between amateur or trainee translators and professionals. For
professionals, this is a crucial and time-consuming stage. Draft versions are
checked against other possibilities, against other parallel texts, against reference
books, against other people's opinions, and so on. Account is taken of sociolin-
guistic factors, target language norms, readership response, readability require­
ments, etc. etc. The more one knows about how a translation can be assessed, the
more rigorous this Error Elimination process becomes. Hence, by the way, the
usefulness of purely theoretical knowledge for professional translators: it makes
for more demanding hypothesis-testing, suggests more pertinent questions to be
asked of the TT.
Notice two things here. First, in order to criticize a translation (a draft) it is
92 Teaching Translation and Interpreting 2

not sufficient simply to say what you think is wrong with it; Error Elimination
includes the necessity of replacing an inadequate item with one that you think is
better. Criticism thus includes the suggestion of improved versions, which them­
selves are then subject to the same critical process. And alternative versions
themselves must be justified, defended, corroborated.
Which leads to my second point here: corroboration. Popper uses this as a
technical term in his view of scientific methodology. To the extent that a Tenta­
tive Theory has been exposed to rigorous Error Elimination, and has survived this
testing process, to that extent the theory is said to be corroborated. That is, it is
not "proved true" or anything of that sort; but it is shown to be a pretty strong
theory, a well corroborated theory. The more rigorous the testing that a theory
survives, the higher is its degree of corroboration. In other words, the more feed-
back you can get about your translation, and the more feedback it survives, the
better it will be shown to be, in the sense that it will be well corroborated.
An obvious application of this idea is in the use of pair work or group work
in class, or exercises involving the repeated revision of a translation (one's own
or someone else's). A class might continue revising a translation until a version
is arrived at that is within the acceptability range of all members of the group -
and then this version could still be submitted to a parallel group. Like theories,
a translation is never 'finished'.

Translational competence
This view of the translation process has obvious implications for the way we
see translational competence. Here I would like to draw on a definition proposed
by Pym (1992: 175, referring to his own earlier work). Pym also argues that
translators "theorise", and in fact that the ability to theorise is an integral part of
translational competence. His formulation is extremely Popperian, and it may be
paraphrased as follows.
Translational competence comprises two abilities. (1) The ability to generate
a series of possible target texts for a given source text (where I think we can read
'text' as anything from an individual item or feature to a complete text). (2) The
ability to select from this series, "quickly and with justified (ethical) confidence,"
the particular version most suited to a given readership.
In other words, a translator must be first able to produce a range of Tentative
Theories, and the more the better, since the wider the range, the more likely the
possibility of ultimately selecting an optimal version. Secondly, a translator must
be able to criticize these TTs, to carry out an exhaustive Error Elimination
process; this requires not only knowledge of the readership, target text function,
desired relation with the source text, etc., but also knowledge about how to self-
 Andrew Chesterman, Finland 93

criticize; how to select relevant criteria to weigh different possibilities against one
another; how to take into account general translational principles prevailing in the
relevant cultures at the time; and so forth. In Pym's words, such a definition of
translational competence "recognises that there is a mode of implicit theorisation
within translational practice, since the generation of alternative TTs [target texts]
depends on a series of at least intuitively applied hypotheses" (1992: 175).
It is precisely at this point that theory joins hands with practice: in fact, there
can be no practice without theory. And there can be no better justification for the
need to encourage trainee translators to become aware of and familiar with purely
theoretical ideas. To quote Pym again:
Unsung theory ... may thus be seen as the constant shadow of what translators do every day;
it is what improves as student translators advance in their specific craft; it is the mostly unap­
preciated form of the confidence slowly accrued through the making of countless practical
decisions; it is what most competent translators know without knowing that they know it
(1992: 175-6).

World 3 and plastic control

One of Popper's most influential concepts is that of World 3. This is a notion
that he first developed as part of an epistemological argument dealing with sub­
jective and objective knowledge, but the concept has since seemed to take on an
independent life of its own, and not just in Popper's own work.
Popper distinguishes between three worlds. World 1 is the world of physical
objects or physical states, what we normally call the objective world. World 2 is
the world of states of consciousness, mental states, feelings and the like - tra­
ditionally, the subjective world. World 3 is the world of what Popper calls "the
objective contents of thought" (1972: 106). World 3 is the site of scientific
knowledge, the contents of libraries, scientific ideas, theories, ideas, arguments,
problems.
World 3 entities are distinct from World 1 ones. A book (as a World 1 entity)
may be destroyed - even all copies of it - but its ideas may survive. Of particular
interest here, however, is the way World 3 interacts with the other worlds. World
3 entities are, in the first place, products of World 2, that is, of human minds.
But (like the concept of World 3 itself) once these entities have entered World
3 they can exist there independently of World 2. Ideas pass from people to
people in conferences, for instance. But now comes the crux: World 3 entities
can subsequently affect both World 2 and World 1. A new idea you come across
may affect the way you think yourself, your own World 2; it may even affect the
way you act on your environment, on World 1.
The nature of this effect of World 3 is not deterministic: Popper is no deter-
minist. Yet he also argues against the extreme indeterminist postulate of absolute
94 Teaching Translation and Interpreting 2

freedom. His key concept here is "plastic control". This is understood as a kind
of flexible control plus feedback. Popper explains the idea thus (1972: 240-1):
the control of ourselves and of our actions by our theories and purposes is a plastic control.
We are not forced to submit ourselves to the control of our theories, for we can discuss them
critically, and we can reject them freely if we think they fall short of our regulative standards.
So the control is far from one-sided. Not only do our theories control us, but we can control
our theories (and even our standards): there is a kind of feed-back here. And if we submit to
our theories, then we do so freely, after deliberation; that is, after the critical discussion of
alternatives, and after freely choosing between the competing theories, in the light of that
critical discussion. (Italics original)
With respect to translation, and the teaching of translation, I think the implic­
ation is clear. "Theories" here are ideas about translation: the whole collection of
principles, norms, arguments, commonly agreed rules-of-thumb and the like that
constitute the set of expectancies about translation within a given culture at a
given time. All these exist in World 3. To the extent that we accept these World
3 entities, they do indeed exert a plastic control over our translatorial action. But
before we can accept them, before we can even weigh them up criti-cally, we
have to be aware of them. Hence, of course, the importance of teaching such
things overtly.
Hence, too, the usefulness of volumes like this one, which may even churn up
new World 3 entities. Thus we help to create our own future.

Linguistic ethics
Much has been written recently on translatorial ethics (see e.g. Robinson 1991;
Pym 1992). While Popper does not discuss the ethics of the translator as such,
his general views on ethics seem to me most applicable here.
Popper's view is based on a kind of upside-down utilitarianism. He criticizes
the Benthamite assumption that what matters is the greatest happiness of the
greatest number: why is it moral to increase the happiness of someone who is
doing all right anyway? More urgent than the maximization of happiness is the
minimization of suffering.
Note how here too Popper's schema is appropriate: minimizing pain is, as it
were, a form of Error Elimination. We do not start from some kind of ideal state
of bliss, any more than we start from some kind of absolute truth. True, such
ideals may function as regulative guiding lights, exerting their plastic control
from World 3; we may have our visions of Paradise lost. But we live in the real
world, where we can only grope tentatively forward by trial and error, eliminat­
ing errors where we can along the path to truthlikeness, and eliminating suffering
where we can along the path to a just society.
For translators, I suggest that the first application of such ethics is in the con­
ception of the translation process not as something straining for an ideal, but
 Andrew Chesterman, Finland 95

rather as one of minimizing error, what we might think of as "communicative

suffering". That is, the translator's task is to minimize misunderstanding. It is
this, not some mythical equivalence, that is the fundamental motivation for the
way a translator approaches the source and target texts, for any adjustments, re­
visions and the like that may be necessary. I take this to be the bottom line of
everything the translator does.
The corollary to this linguistic ethics is manifest in Popper's stress on clarity,
which for him is perhaps the most important standard of language.

Bootstraps
Popper also uses his schema to illustrate his understanding of the Darwinian
theory of evolution, which he links to a theory of the functions of language. Like
animal languages, human languages have functions of self-expression and sig­
nalling, but unlike the former they also have higher functions, in particular those
of description and argumentation. The language of these higher functions is pri­
marily exosomatic, in that descriptions and arguments are World 3 entities, and
it is this property that allows the development of rational criticism and hence of
reason and humanity as such. The rational criticism inherent in the proposing of
Tentative Theories and the process of Error Elimination becomes the tool by
which mankind evolves. Popper thus claims that his schema "gives a rational de­
scription of evolutionary emergence, and of our self-transcendence by means of
selection and rational criticism" (1972: 121. Italics original).
The schema is thus a model for education and personal growth, too, for one's
personal research and development programme, as it were. Translators learn from
their own previous translations, their own previous errors; they can learn from
exposure to new translations, from ideas about translation and its theory. They
can become better translators. In Popper's terms, they do this first by interacting
with others, by soliciting feedback etc., in critical communicative dialogue. But
secondly, they do this themselves, through their own inner self-criticism: note
Popper's self-transcendence.
In this way "we lift ourselves by our bootstraps" (1972: 121). Or, to quote one
of my own bootstraps (Chesterman 1993: 78): "we translate ourselves."
 THEORY AND PROFESSIONAL DEVELOPMENT:
OR ADMONISHING TRANSLATORS TO BE GOOD

Sergio Viaggio, United Nations

This article has been prompted by the following assertion in Anthony Pym's
otherwise formidably insightful Translation and Text Transfer: "Happily, general
translational practice has quite probably changed far less than theory" (1992a:
190). I beg profoundly to disagree: If translational practice had not lagged so far
behind theory, most translations would be much better, and fewer readers would
have so much trouble with their electronic gadgets. The aim of his and most
works on translatology ought to be, precisely, to influence the way translators
translate, and not, as in descriptive approaches, merely take stock of it. A lot has
been justly said about the deplorable chasm between translatology and translators;
that should be a reason for us to try and meet practitioners at their working tables
or, better, before they get to them: at the university, where the theoretical
rationale (a redundancy if there ever was one) of the practical do's and dont's is
to be learned.
Of course, much profit is to be derived from observing good translators and
studying good translations. But how do we determine who and which they are,
and what makes them so? Unless we postulate a provisional concept of what a
translation should be, there is no way we can determine how close to or how far
from it any particular effort falls. If, on the other hand, we do set such criteria,
we can then elicit general or specific principles, rules, methods, strategies that
prove systematically successful, or at least more successful than others. Once thus
elicited, they become inevitably those standards a translation should follow- even
if for the nonce, until superseded by new, more advanced ones; exactly as is the
case in any other discipline.
It is, I think, clear that the moment a theoretician comes up with a definition
of what a translation is, he is also, up to a point at least, defining what it is not,
and the degree of conformity to what it is acts automatically as a quality quan­
tifier. Unless we agree with scholars such as Gideon Toury, for whom a trans­
lation is whatever a given culture considers it to be1 (and even then, we can
safely state that in our culture the best translation of an owner's manual is the
more intelligible one), prescriptivism is inescapable. If there is no right, or at
least better way of translating, then we are all wasting time, breath and money:
there is nothing but language to teach.
98 Teaching Translation and Interpreting 2

What nearly all translators, from the best to the worst, seem to have in com­
mon is precisely a lack of any coherent, systematised, weighted conceptual
framework behind their practice. Observing them allows to glean more dont's
than do's. As Albert states:
Each translator (whose task is to create equivalence) will have an individualised equivalence-
concept... [His situation] is often the most uncomfortable: Even if he knew all the possible
equivalence-types which may come up while translating a specific source-language text, he is,
as an expert, fully aware of the fact that he will not be able to create total equivalence be­
tween the two texts: his 'freedom' only means that he has a choice as to which equivalence-
types to reject and which to keep. (Albert 1993: 13)
A translator, he adds,
always needs some kind of global conception (idea, strategy, philosophy) when translating any
text. There is some kind of philosophy in all acts of translating, even if the translator himself
is not aware of it or tries to deny it: the latter is often the case, for in my opinion it is more
often than not the lack of awareness of such a philosophy which makes poor translations
really bad (Albert 1993: 13. His italics, my bold type).
However, he laments further on, "no theory of translation will tell the trans­
lator which features are, or should be, more significant than others in a specific
text; there is no higher forum to which he can turn concerning the adequacy of
his decisions" (Albert 1993: 13).
Perhaps not so much in literary translation, but as far as pragmatic texts are
concerned, the situation with theory, to my mind, is not half as bad, as I shall
endeavour to show. To begin with, and if he knows his translatology, the trans­
lator is not that much more at the mercy of his personal intuition, competence
and luck than a physician faced with any specific case. Collective knowledge and
professional lore bring the qualified practitioner of any discipline safely further
and further into his realm. If translatology has been born at all, it is because
translation, in its broadest sense, has itself changed drastically over the last forty
years or so: It has ceased to be a mainly literary side activity of knowledgeable
or incompetent dilettanti to become a full-fledged, mainly technical profession.
The bulk of translations being now production-related, an overwhelming social
need has arisen for the specialist at translating. The sheer fact that we can now
make a living not just translating, but talking and writing about translation or
teaching it is yet a further proof of our amateur craft's coming of age into a pro­
fession, that is the conscious, socially conditioned and recognised practice of a
discipline. It is no coincidence that, in general, translatology is most developed
where translations are better; translations, on their part, happen to be better where
they are better paid, and better paid where they are more readily recognised as
essential aids to economic production, the work of necessarily highly qualified
practitioners. So much so, that the translator of literarily inane but politically
relevant or production- or trade-related texts stands to make much more money
 Sergio Viaggio, United Nations 99

than the heirs of Dryden and Gerard de Nerval.

That being said, the gap between collective professional lore and the lagging
or pioneering individual physician is probably minute compared with the ex­
tremes we witness in our field, where stragglers are so rife and far behind, and
pioneers so few and far ahead - again, a typical phenomenon at the inception of
any discipline or science. The bulk of our practitioners do not approach their task
scientifically; they do not even see translation as communication; many an inter­
preter, for instance, is not even aware of the difference between written and spon­
taneous oral speech. It is a shame, of course, but quite inevitable: We are at a
stage where scientific thought is beginning to creep into our ranks, where indi­
vidual ideas and insights have only started becoming a systematic body of knowl­
edge, where we actually read and criticise each other. We owe it to the pro­
fession and to the rest of our colleagues to be prescriptive, to tell them that ade­
quacy takes precedence over equivalence, and textual equivalence over local cor­
respondence; that there is no point in setting out to translate without a clear con­
cept of our own and the original's skopoi.
For instance, we understand now that there are as many possible valid trans­
lations of a given text as there are different uses for the target-language versions.
Nida's dynamic equivalence, Newmark's semantic vs. communicative approaches,
Seleskovitch's interpretive theory, Reiss and Vermeer's skopos theory, and Gutt's
application of relevance theory are but more and more refined ways of conceptu­
alising this crucial epistemological leap. All of them have been dialectical stages
in the evolution of the concept of what translation should accomplish and the
ways to attain it; and all of them have given, if not birth, at least a theoretical
foundation to - precisely - different ways of translating. Any serious and qualified
translator who has followed this evolution cannot have failed to refine his prac­
tice as well. But, as I have pointed out, this progress is still far from permeating
general translational practice worldwide: most translators go on without even
realising that the texts they are called upon to translate are functionally different;
thus an ad, a prescription, a notice, a contract, even a poem get the same literal
once-over.
Most of my professional life I have lived in New York, where everything
seems to be translated into Spanish. I can aver that the overwhelming majority
of the pragmatic translations of any kind I have encountered have ranged from
bad to gruesome, not only linguistically, but, above all, conceptually: they are the
product of an incompetently carried out misconceived assessment of what a trans­
lation is or, rather, should be. I know but very few of those translators; yet most
of those I do know - some of whom are very good linguists - have somehow or
other abandoned or failed at other professions and decided to put their non-speci-
100 Teaching Translation and Interpreting 2

alised linguistic knowledge to remunerative use:2 their craft has improved haphaz­
ardly - if at all - with time. Spanish translations at the UN, for instance, have
gotten at best marginally better over the last four decades, while there seems to
be no coherent approach among the six language sections as to how to translate
any single document (the French section, mindful of their watchful readership,
will tend to be cibliste and go communicative, while the Spanish, fearfully sourc-
ière, usually goes semantic, for example).3
Now, advances in translatology, the refinement and specification of the
concept of translation itself, its progressive delimitation from neighbouring
phenomena such as adaptation have been the result of a deeper understanding of
language, discourse, and communication. The descriptive part of translatology,
I submit, is mainly in the hands of linguistics, stylistics, poetics, discourse
analysis, communication theory, cognitive psychology, neurophysiology, and what
not - in short, of all those disciplines that describe for us what communication,
and more specifically lingual, and still more specifically mediated interlingual
communication is and how it works. On the basis of such evolving description,
it is our collective task, insofar as we can, to conceptualise translation and to
find, rate and make explicit the means to achieve it.
The time will certainly come when Nida, Seleskovitch, Vermeer, Gutt, and
Pym will be names analogous to those of Archimedes, Faraday, and Newton,
their theories partially borne out, partially refuted, surely superseded by theories
based on a more thorough and deeper knowledge of the world and the mind. Up
to a point, it is beginning to happen: skopos and relevance have placed dynamic
equivalence in a new, more proper context; pretty much as relativity did to uni­
versal gravitation. Both have themselves been criticised already: skopos by Pym
and Gutt, relevance by Tirkkonen-Condit (1992). The main difference between
translatology and the more established disciplines is that, exactly as with the
beginning of all other sciences, metaphysics and speculation have not as yet
taken their rightful place behind scientific observation and experimentation; but
that too is very much happening already, particularly with respect to simultaneous
interpreting.
Translation has often been described as intelligent reading followed by com­
petent writing; true, but not enough. If it were, any intelligent reader and com­
petent writer who knew his languages could translate professionally. Despite in­
terested and disinterested assertions to the contrary, it is not normally the case,
and even when it is, it is not altogether satisfactory. I have known several ex­
cellent writers who know their passive languages to the hilt, and whose translat­
ions of UN development projects leave a lot to be desired: their linguistic instinct
prevents them from excessive awkwardness or outright mistakes, but they never
 Sergio Viaggio, United Nations 101

seem really to grasp what translation is all about: their version of a camcorder
owner's manual would undoubtedly be better than most, but still too dependent
on an original which is, for all practical purposes, irrelevant. The boldness to let
go of the original, to deverbalise, to assess the skopoi of source- and target-text
requires something additional between intelligent reading and competent writing.
Translatology is there to provide it.
Let us hypothesise how a translator might have refined his practice as a re­
sponse to the development of theory: Take as an example precisely the owner's
manual of a camcorder to be translated from English into Spanish for the Latin
American and Spanish markets. The intuitive, uninitiated translator (one of our
beginner students, for instance) would try and produce a literal translation. All
he knows is English and Spanish; he has no tools other than his bilingual diction­
ary. A phrase such as "Press on the lid until it clicks into place"4 will give him
a real headache. At the threshold of translatology proper, our translator can al­
ready safely extrapolate Vinay and Dalbernet's (1957) insight into the basic dif­
ference in the way French and English go about making sense: English operates
on the plane of reality, French (and Spanish) - on that of understanding. Our
translator can then understand that, at the sheer linguistic level, a radically
different approach is required for rewriting. It is the sudden irruption of sound,
of reality, that gives him his pain; but if he chooses to switch to the level of
understanding, he could safely write "oprima la tapa hasta que calce" ["push the
lid until it fits into position"]. He can already think in terms of the linguistic/
stylistic acceptability of the target text. Translation theory has already begun
helping our translator map his way between the kind of language equivalences.
His translation has become idiomatic. But that is still not quite enough for a
successful target-language owner's manual. He must go on reading.
Next, Nida will lead him to think in terms not only of linguistic idiomaticity,
but that of cultural acceptability as well. The original goes "Keep an image of
your favorite batter in his moment of glory." Our translator has now stopped to
ponder the fact that, while baseball is extremely popular in Central America and
Cuba, the national game elsewhere in the Spanish-speaking world is football. So
he does not hesitate to add a football centre-forward; except that in football it is
not so much the player but the goal that matters, so he writes on "Conserve una
imagen de su bateador favorito en su momento de gloria o de ese formidable gol
de medio campo"5 ["Keep an image of your favourite batter in his moment of
glory or of that astonishing goal from mid-field"]. Our colleague has realised two
crucial points: one is that he may adapt, even rewrite; the other is that he must.
He has had his first insight into the true science of translation: translation works
not on languages but on texts; texts are not language specimens but communi-
102 Teaching Translation and Interpreting 2

cative intentions linguistically consigned; and in this kind of pragmatic text, tar­
get culture acceptability is of the essence. He may share my and many others'
doubts about whether the Bible really lends itself to the same approach as a cam­
corder manual, but he will feel sure a camcorder manual does.
Our hypothetical translator will next have the opportunity of reading Seleskov-
itch and Lederer. Aware of the distinction between meaning and sense, he starts
thinking like the user he is translating for. Sense, as grasped by the reader, is the
extra-linguistic product of the connection the reader establishes in his mind be­
tween the linguistic meaning of the utterance and what he knows about the world.
So shared knowledge of the world between addresser and addressee is crucial for
an utterance to be understood by the latter as intended by the former. Our trans­
lator sees that the fundamental thing is not so much that the text be idiomatic and
culturally acceptable, but that it help its reader use the device. He tries to place
himself in the reader' shoes. This allows him to detect better ways of having the
sense carried across to a Spanish speaker. He feels free to be more explicit than
the original or, more crucially, less so. He has also learnt that communication
works on the basis of the principle of synecdoche, so he strives to make sense
as plainly as possible with no unnecessary repetitions or moronic instructions
such as "In order to replace the cassette, open the lid marked cassette on the left
side of your HARAKIRI 6000 Super eight hand-held compact enhanced image
recorder. After opening the lid, replace the cassette, then close the lid once again
(see figure 2 below on this page)." He knowingly and boldly writes "Para
cambiar la cassete, abrir la tapa (figura 2)" [To replace the cassette open the lid
(figure 2)]. He will also have made an additional conscious effort to go beyond
searching for the right electronics terminology, and to try to understand how the
gadget works, so as to make sure his Spanish explanation is accurate - not just
linguistically, culturally and commercially, but also practically.
If our translator has access to Russian translatologists such as Schweitzer
(1973), he begins to look upon his task not so much as a search for linguistic and
other kinds of equivalence, but as the quest for adequacy. Adequacy with respect
to the ends pursued, rather than equivalence of the means employed; a point he
will be able to round up next in his literature, when he chances upon Reiss and
Vermeer, Neubert and Lvovskaja. He now asks himself why the text was written,
why it is going to be translated, and for whom. Conscious of the relevant factors
to be looked at in the situation, he can be sure that the originator of the trans­
lation wants basically one thing: to sell camcorders. Within that global skopos,
the owner's manual is meant to complement the product. Pleasantly presented,
idiomatic, clear and concise instructions are as important as an attractive packag­
ing and a futuristic design. Our ideal colleague will now do away with the base-
 Sergio Viaggio, United Nations 103

ball batter altogether, mindful that 99% of the camcorders will be sold in Mexico,
Venezuela, Colombia, Peru, Argentina and Spain, rather than in Cuba or Panama.
The disadvantage of leaving out, as it were, a few owners is more than out-
weighed by the concision, coherence and familiarity of a monocultural text.
Further in his reading, our translator encounters Gutt's adaptation of relevance
theory (1990 and 1991). He sees now the crucial difference between interpretive
and descriptive use. He will have finally corroborated a nagging suspicion that
entered his mind the very day he was advised against translating literally: to wit,
that the original, in this specific case, is for all practical purposes immaterial. The
English manual is but a model, a format, a sequence of text and images whose
only constants are, precisely, the images. So his real job is not to come up with
a translation as he initially conceived it, as a source-language dependent target-
language text - albeit a linguistically, culturally and practically adapted one. Now
he realises that his version must be first of all faithful a) to the originator's and
the users' economic skopoi, and b) to electronic extra-linguistic reality. Gutt will
tell him that what he is supposed to produce is not a translation at all, but a
parallel text. The original has systematically referred to an "engine." In the last
page there is a glossary, where it is explained that "throughout this manual, the
term engine refers to the camcorder." Our hypothetical colleague stops racking
his brain to find a plausible equivalent for "engine," writes "cámara" throughout
and does away with the imbecilic terminological gimmick, together with the glos­
sary entry. He takes Gutt's point that "the receptor language texts are intended
to achieve relevance in their own right, not in virtue of their interpretive resemb­
lance with some source language original" (1991: 57). In principle, then, no
formal conformity between source and target text need be expected, function
reigns supreme, and practical acceptability in the target culture carries the day.
Pym's analysis of translation as a reaction to transfer (1992a, 1992b) would
shed even newer light on the essence of the problem, helping our colleague better
understand his task, nay his responsibility as a translator. Before asking why the
text is to be translated, a new question springs now up in his mind: Why has this
text been transferred to a position where it is going to be translated? He already
knows why he blithely well-nigh disregards the original when translating a cam­
corder owner's manual: the original is hopelessly irrelevant. Nevertheless, our
colleague's last effort reproduced at the beginning a dense page blackened with
minute writing, where the reader is warned, among other things, that the device's
warranty is issued as mandated by article 4.2 of statute 4321(b) of the Laws of
the Commonwealth of Massachusetts, but that the provisions of such statute ap­
ply only partially in Virginia, are void in Colorado, and become somewhat illegal
in Delaware. Does it do any harm if, disregarding his financial reflexes, he fol-
104 Teaching Translation and Interpreting 2

lows his translatorial instinct and omits this purely and simply? On the contrary,
answers Pym: the status of the reader of the text transferred through time and/or
space, whether to be translated or not, can be participative, observational, or ex-
cluded. In the case of the camcorder owner's manual, the "translation" (as, by the
way, its original) is meant for participative readers only. All references to source-
country institutions, commercial practices, and dealers automatically turn anybody
not residing in the US into mere observers, and must be eliminated.
The, say, Colombian reader - unbeknownst to him, and to the originator of the
translation, but not any more to our translator - is but going to be confused, mis­
led or simply put off whenever he is excluded from participation. It is his device,
he bought it and paid good money for it, he does not give a hoot if a special tax
applies in Ohio, or if the warranty is not valid in California. He wants a manual
that will tell him clearly and concisely how to use his gadget. For him, there is
only one manual: his, in Spanish; whether or not it happens to be a translation
or an adaptation from another language is absolutely immaterial. He has already
bought the device. Not only does he not need to be reminded at every turn that
it is a "HARAKIRI 6000 Super eight hand-held compact enhanced image re­
corder" that he must learn to handle, but - in Spanish at least - this kind of ono-
mastic overkill is irritating and intrusive. (The initiator should also have had it
in mind, of course: He has already sold the camcorder. All he really wants now
is that his client should be happy and use his toy adeptly. He is not interested in
the least in having to abide by his warranty.)
Thus, our translator would have seen himself constantly develop as one leaving
most of his colleagues farther and farther behind. He would have dared increase
his fees (his translations have become, after all, shorter and shorter). And if he
is not yet a translator but a student leaving the university to try and make it as
a professional, he would find out in no time that many, perhaps most, of the orig­
inals he will be called upon to translate will be pragmatic texts such as this man­
ual. And when confronted with his first real-life translation, he will be already
better armed than most veteran self-made colleagues.
In both cases he will be totally aware that the original has, for all practical
purposes, no author - only originators and intermediaries. It will be his duty as
a true specialist at translating to talk them into accepting his expert view of text­
ual reality and securing from them maximum formal leeway. His chances of suc­
ceeding will basically depend on the fact that, besides providing a communicat­
ively competent text (the idiomatic and accurate product of a correctly assessed
skopos, whereby descriptive use prevails over interpretive use in order to ensure
the reader's participative status), he can verbalise the rationale for it, that is
explain the theory behind it - more or less the way a physician explains to a
 Sergio Viaggio, United Nations 105

patient why she should take the prescribed pill and not the one recommended by
her aunt. Only when a translator has secured (earned, that is) his client's trust and
respect as a specialist at mediated interlingual communication can his profes­
sional relationship with him be akin to that between the lawyer, the physician or
the architect and theirs. Our translator will realise that, given the indispensable
linguistic ability, it is his knowledge of the specifics of translation as a discipline
that will give him the edge over equally talented intuitive practitioners.
If only the rest of his competent colleagues had become aware of this, if only
they strove collectively to have it understood and accepted by the initiators, if
only they dared rewrite and adapt whenever it is called for, if only they did not
lag so far behind theory, how much more translators would be recognised as full-
fledged professionals ... and how much more money we would all be making!

Notes
1. For a deeper elaboration of the point, see Viaggio (1992a).
2. The French dichotomy is Ladmiral's (1990), the English one Newmark's (1981; 1988).
3. As quoted by Pym (1992b: 187-188), and Gutt (1991: 6-7).
4. All the following examples have been extracted from actual texts.
5. If our translator is, besides, a competent stylist, he will also change the second su to un.
THE PROCESS-ORIENTED APPROACH IN TRANSLATION TRAINING

Daniel Gile, INALCO & CEEI (ISIT), France

Product-oriented vs. process-oriented training approaches

Traditional translation training is based on translation assignments which are
corrected in class, with teachers criticizing or approving the students' choices and
presenting their own solutions. Although the method is operational, it is less than
optimal on two points:
- It focuses on the product rather than on the processes, which means that inferences for the
correct processes are to a large extent made by the students themselves, with little possibility
of control by the teachers. This is one reason why many authors (such as Delisle 1981;
Komissarov 1985; Gentile 1991; Larose 1992; Shlesinger 1992a) have advocated the use of
theory in training. It is, however, difficult to integrate theory into practice in a product-oriented
approach, the more so since many practitioners, including teachers, strongly oppose theory (as
noted by for instance G6mar 1983; Komissarov 1985; Pochhacker 1992a; Sager 1992; Viaggio
1992c).
- Very often, students reject the teacher's criticism and solutions because of diverging
linguistic norms and because they feel attacked. This slows down the learning process.
If teachers focus on the process, they can be less critical of the product and,
to a large extent, avoid such problems.
On the other hand, a process-oriented approach also has potential limitations:
- Because its focus is on the process, it may not be a powerful tool for product
fine-tuning.
- In order to be able to operate on the processes, teachers need to learn or
develop explicit process rules, and they are sometimes reluctant to do so.
- In order to use the rules efficiently, teachers need to be able to convince the
students that they are correct. This again requires theoretical preparation. It is
very difficult to come up with explicit rules that are simple enough to learn
(in a practical, professional course, theory should not take up too much time
and energy), and yet cover all the difficulties students may encounter.
Process-oriented training has been advocated by several teachers in recent
years (Martins 1992) and is becoming increasingly popular with translation
teachers, as will also appear from several articles in this volume. It has been
practiced in several forms, including a series of exercises developed by Delisle
(1980). This paper introduces the process-oriented training methodology devel­
oped at the Department of Japanese and Korean Studies of the Institut National
des Langues et Civilisations Orientales (INALCO) in Paris. It has also been used
in other contexts, including workshops for experienced translators.
108 Teaching Translation and Interpreting 2

A process-oriented translation training system

The basic philosophy of the system can be summed up as follows:
- During the process-oriented part of the course, trainees are considered as stu­
dents of translation methods rather than as producers of finished products.
Throughout this period, their target language texts essentially serve as a look­
ing glass revealing their methods, insofar as their problems are generally
symptoms of methodological weaknesses. Problems which can be attributed
to linguistic deficiencies are not dealt with during the process-oriented phase
of the course.
- Teachers take a normative attitude as far as the processes are concerned. As
regards the product, they put questions to the students whenever possible
rather than criticise them ("Why this choice?" "Did you consider alternatives?"
"If so, what made you choose this solution?" "Are you satisfied with this
solution as far as logic/clarity/language is concerned?").
- Processes are supported by theoretical models which explain and integrate
them. The most important one is the sequential model of translation, discussed
below (See also Gile 1992b).
- Problem diagnosis can be done partly by analyzing the product and partly by
putting questions to the students as explained above. Written problem reports
by the students are a very useful tool for diagnosis: When handing in trans­
lation assignments, students are also required to report in writing the problems
they encountered while doing the translation - difficulties in understanding a
particular sentence, in reformulating an idea, in finding the meaning of a
source language term, in finding a good target language equivalent, etc.

The sequential model of translation and error analysis

The model is shown on the opposite page. It was developed for pedagogical
purposes. It is not intended to be a full or accurate description of the translation
process, but as a framework beginners can use in order to optimize their produc­
tion and avoid the most common errors.
The model consists of a 'comprehension phase' and a 'reformulation phase'.
Translation starts with a 'translation unit' (which is not to be understood in a
strictly scientific sense, but as an intuitive entity consisting of a word or small
group of words that translators deal with at the micro-textual level - in our ex­
perience the intuitive nature of the definition has not caused problems for train­
ees). It is read. Its meaning is inferred from the text as a meaning hypothesis.
This hypothesis is then checked for plausibility on the basis of the translator's ex­
isting linguistic and extra-linguistic knowledge. If the first meaning hypothesis
is deemed plausible, the translator can move to the reformulation phase. If not,
Figure 1. The sequential model of translation.
110 Teaching Translation and Interpreting 2

the translator must find another hypothesis and check its plausibility, etc.
In the reformulation phase, the translator formulates a first target language text
for the translation unit. He then tests it for fidelity and for linguistic acceptability.
If results are not satisfactory, he writes a new target language text for the same
unit and tests it again, and so on.
Periodically, fidelity and acceptability tests are conducted for groups of trans­
lation units, as good results at the single translation unit level do not ensure good
results at the text level: sentence segments, full sentences and even paragraphs
may be skipped between individual translation units just as repetitions and other
stylistic incompatibilities as well as terminological drifts (a lack of consistency
in terminological usage) may emerge in a larger text segment.
At every step of the process, existing linguistic and extra-linguistic knowledge
must be used, and whenever necessary, additional knowledge must be sought.
One of the advantages of the model is that it conveniently labels necessary
components of correct translation work ('comprehension phase', 'reformulation
phase', 'comprehension loop', 'reformulation loop', 'plausibility test', etc), and
thus provides an easy way of indicating to students, both linguistically and graph­
ically, the location of the methodological weaknesses that have led to most errors
and infelicities (Gile 1992a).
For instance, a grammatical error most often indicates that the student did not
perform an acceptability test properly. Bad spelling in one occurrence of a word
and correct spelling in another definitely indicates that the problem lies in this
test. When the student's translation is illogical, it is probably due to the fact that
he did not do a 'plausibility test' in the 'comprehension phase', or a 'fidelity test'
in the 'reformulation phase'. Terminological errors can almost always be traced
back to an ill-conducted 'knowledge acquisition' operation. Terminological drifts
(a lack of consistency in terminological usage) can be ascribed to a missing
'acceptability test' on groups of translation units.
When finding a problematic word or statement in a translation, the teacher
asks the student whether this solution sounds logical, plausible, linguistically ac­
ceptable, and consistent with the rest of the text. If the student thinks it does and
gives a plausible explanation, the teacher can accept the answer on the grounds
that the procedure was correct, even if the student's solution is wrong by his
standards. The teacher can also make a mental (or written) note of recurrent
problems which will have to be dealt with in the product-oriented part of the
course. However, students usually answer spontaneously that the segments are not
logical, linguistically acceptable or consistent with the rest of the text, thus
making their own diagnosis of their faulty procedures. Having identified the
problem themselves, they are presumably more willing to take remedial action.
 Daniel Gile, France 111

Implementation and results

We started introducing the process-oriented approach in our scientific and
technical Japanese-into-French translation classes at INALCO about 10 years ago. .

For that purpose we developed a set of basic concepts and models (Gile 1990a;
1992b). Over this period we have made the following observations:
1. Psychologically, the process-oriented approach seems to generate less stress
than the product-oriented approach.
2. Students are interested in the models and rules that are presented to them,
and tend to accept them.
3. There are difficulties with problem-reporting, which the students tend to
forget during the first few assignments because it is new to them. However,
once they become accustomed to the idea, problem-reporting becomes a very
efficient tool. Not only does it help the teacher with error diagnosis, but it also
requires that the students carry out a methodological analysis and reflect upon
fundamental questions, which makes them more receptive to instruction on
issues such as: "To what extent is it admissible to deviate from the text?",
"What is the reliability of background documents if their terminology is vari­
able?", "How should one deal with source text segments which seem illogical
even after much analysis?"
We consider problem-reporting a strong component of the process-oriented
approach.
4. There is a fast and considerable reduction in the number of errors due to
faulty analysis of the source text and there were remarkably few logical con­
tradictions in the students' texts.
In our view, this is the most positive result of the process-oriented approach.
5. However, as far as difficult decisions are concerned, the process-oriented
approach seems to be of little help to the students.
This is probably due to the fact that rules presented to the students are too
general to be used for specific decisions where intuition and experience play
an important role. We do have a number of models, rules and methods for
areas which are not covered by the sequential model, but, inevitably, there are
some problems which they do not address adequately.
6. The process-oriented approach does not seem to improve the students' im­
plementation of additional knowledge acquisition.
The problem apparently lies in motivation. Although they know that additional
information is needed, students do not perform the necessary operation thor­
oughly enough, probably because of the time and effort required.
7. The process-oriented approach is moderately efficient with respect to the
linguistic quality of the output.
112 Teaching Translation and Interpreting 2

There are three reasons for this:

a. In the process-oriented approach, the teacher only questions the stu­
dents' norms, without the rectification which is found in a product-orient­
ed approach.
b. For lack of sufficiently strong motivation, even when they are aware
that the linguistic quality of their target text leaves something to be de­
sired, students tend to go through the 'reformulation loop' (writing a target
language version of the translation unit, then testing it, then correcting it,
testing it again, etc.) a few times and then stop, even if the result is not
satisfactory.
c. Testing for acceptability does not necessarily prevent a text from being
unacceptable. Distinct text writing tactics need to be taught for this
purpose.
The process-oriented approach described here therefore only helps students do
away with some of the linguistic weaknesses of their initial output.

Conclusion
The process-oriented approach has been very popular with students. It was es­
pecially appreciated by self-taught practising translators who attended the course
or workshop after several years of professional experience. It does seem rather
powerful as a methodological guiding approach in that it strongly and rapidly re­
duces product deficiencies attributable to incorrect translation methodology. How­
ever, it is definitely not a sufficient teaching tool for students whose motivation
is weak and for those whose linguistic norms in the target language are poor, in­
sofar as it does not teach how to write. Similarly, it does not provide solutions
for specific, difficult cases. We therefore feel that the approach is very useful in
the first part of a course, but that product-oriented teaching must follow.
We strongly recommend the process-oriented approach in two specific cases:
1. Courses for experienced translators, who want to improve their methods
rather than acquire basic expertise, which they already have. Such translators
are also particularly sensitive to product criticism. Therefore, a process-
oriented approach is more appropriate for them than a product-oriented
approach.
2. Courses involving source or target languages that the teacher does not
know. This occurs in various international cooperation programmes. In such
cases, the teacher is not in a position to inspect the product as such, but may
provide useful guidance through process-oriented methods.
 COMPREHENSION IN THE TRANSLATION PROCESS:
AN ANALYSIS OF THINK-ALOUD PROTOCOLS

Jeanne Dancette, Universite de Montréal, Canada

To understand a text is to actualize links that may or must be established be­

tween given elements in the textual structure and other elements pertaining to in-
tertextual and extratextual information. It is to build a meaningful and coherent
representation of the text. Such a representation does not operate at the semantic
level only but more at the "level at which linguistic and non-linguistic informa­
tion are mutually compatible" (Jackendoff 1985: 95), that is at the conceptual
level.
Paradoxically, translation - which is a linguistic operation concerned with se­
mantics as well as syntax and morphology in their relation to semantics and to
the rules of well-formedness of both the source and target texts - cannot occur
successfully without this meaningful and coherent conceptual construction.
This broad assumption, however, needs to be qualified:
a) Successful translation can occur to a certain extent without an elaborate
conceptual construction, that is with a limited understanding. This could corre­
spond to what Cunningham (1985: 235) calls "literal comprehension" as opposed
to "inferential comprehension" and "creative response", which he sees as higher
levels of comprehension. In the few occasional instances of isomorphy between
languages, literal translation and even sometimes mere transcoding are possible.
In such cases, literal comprehension is enough.
b) Conceptual construction as a cognitive process does not guarantee success­
ful translation, for the main reason that the conceptual links made by the trans­
lator may be wrong, albeit coherent.
We are concerned here with the notion of referring. If reference is defined as
a "relationship between expressions in a language and things in the real world
that these expressions refer to" (Jackendoff 1985: 29), then we are dealing with
the interaction between linguistic performance (decoding linguistic structures,
identifying their grammatical class and meaning) and cognitive conceptualization
(relating the meaning of these structures to a "text-world model", to use de
114 Teaching Translation and Interpreting 2

Beaugrande's terminology (1980: 77). I will show that in the event of a compre­
hension difficulty, a reader (translator) must be looking at the right paradigm in
order to succeed. In other terms, his decoding of a given structure is dependent
on his expectation (however rudimentary) of a hypothetical tentative text-world
representation.

The experiment
To illustrate this, I will present some preliminary results of an empirical study
using videotaped think-aloud protocols with a group of five translation students.
The students who volunteered for the study were chosen on a random basis;
two had two or three years professional experience, the others had hardly any ex­
perience at all; they were given five segments of text from news magazines; they
received the entire article but were told to translate only one or two sentences in
each article. The simulated pragmatic situation was that a French editor wanted
to publish the translation of the article. He noticed that a segment had been o-
mitted and asked the student to translate the missing passage.
The students were given four dictionaries: the Webster's New Collegiate
Dictionary, the English-French Robert-Collins, the Petit Robert and the Sylvain,
an English-French dictionary of accountancy and related sciences. These diction­
aries are the usual tools of translators in their normal work and were considered
to be sufficient for the task.
After a short briefing on the think-aloud method, the students were asked to
verbalize their thinking; they were recorded and video-taped. After the trans­
lation, they were given a questionnaire, the purpose of which was to establish
more precisely whether they knew the meaning of a few key words or concepts
used in the text. No time limit was set; they performed their task in one and a
half to two and a half hours.
All the texts presented problems that we assessed as average for professional
translators, but rather difficult for students of translation. I will present parts of
the protocols dealing with one such text.
A LEADING INDICATOR THAT'S LEADING IN THE RIGHT DIRECTION
Is business activity starting to stir? Chief Economist Irwin L. Kellner of Chemical Bank notes
that the Commodity Research Bureau's spot industrial-price index recently hit its highest level
in over a year and is more than 10% above its early February low. Although a lead strike in
Peru helped push the index higher, Kellner notes that 10 of its 13 components have posted
gains since February. "At the very least, the rise in the index suggests that the economy is
still expanding," says Kellner, "and it may just be the first sign of an acceleration in economic
growth. {Business Week October 5, 1992)

The students had to translate the passage in italics; the words in bold ("leads",
"components", "gains") indicate where errors were made:
 Jeanne Dancette, Canada 115

Lead: Three students missed the unit altogether, which at first sight is surpris­
ing considering that they had dictionaries; a fourth student translated by "produc­
tion" (grève de production) after a twisted, fallacious reasoning discussed below.
Components was mistranslated by "facteurs" in one case and by "usines"
['factories'] in another.
Gains (in posted gains). Two students translated this word by "profits" or
"bénéfices", thus shifting from the semantic field authorized by the text to
another unauthorized one. The student who chose "usines" for components chose
"bénéfices" for gains, which is at least coherent.
The passage also contained notional difficulties, that is difficulties depending
on 'encyclopedia-type' knowledge in a given field, namely economics.
LEAD STRIKE.- If one knows that spot prices mainly concern raw materials,
then it can be inferred that lead [led] belongs to the class of commodities or raw
materials, and does not refer to lead [li:d] as is mischievously suggested by the
title. Graph 1 identifies this link as an implication; it occurs at the pragmatic
level.

Graph 1

INDEX refers linguistically to "spot industrial price index" and to "compo­

nents", since an 'index' by definition involves averaging components. Graph 2
indicates a co-referential link occuring at the linguistic level.
116 Teaching Translation and Interpreting 2

Graph 2

GAINS in the phrase "posted gains" is the repetition of the topic; as shown in
Graph 3, it relates to "highest level" and "push the index higher". I term this link
'thematic reiteration'.

Graph 3

Another notional link could be established between "higher levels" or "gains"

in the "industrial price index" and the "expanding economy". However, this link
was not crucial for the translation.
Let us now turn to the students' approach to these lexical and notional diffi-
 Jeanne Dancette, Canada 117

culties. The analysis of their protocols and the answers to the questionnaire made
it possible to trace lexical errors back to conceptual or notional shortcomings.

Translator A: translation and protocol

Bien qu'une greve au Perou ait fait monter I'indcx encore plus haut, Kellner
fait remarquer que 10 de ses 13 composantes ont affiche une progression
certaine depuis fevrier.
Bien qu'unc partie de la progression de I'index soit due a
Bien que la progression de I'index soit en partie explicable par une greve que
le Pérou a recemment connue, Kellner fait remarquer souligne que dix des
treize composantes de cet index ont affiche une progression certaine depuis
fevrier,
Except for the rendering of lead and index, this translation is good both for its
lexical and phraseological precision and for its rhetorical structure. This student
is the only one who paid attention to the internal argumentation signalled by
although, as can be demonstrated by the progression from a literal translation:
bien qu'une greve au Perou... [crossed out] to a more elaborate one: bien que
la progression de l'index soit due [crossed out] soit en partie [added] explicable
par une greve...
I submit that with the omission of lead, it was impossible for the student in
question to build the conceptual connectivity of the passage; lacking a complete
and correct conceptual representation of the text, he could not "catch on" to lead.
The protocol ran as follows:
1 [Reads] OK, alors lire le texte ... aussi je vais lire ce qui est avant, le petit paragraphs
2 histoire de comprendre;... alors c'est I'index des prix industriels... [Reads] OK, although a
3 [li:d] strike in Peru ... helped push the index higher ... 10 of its 13 components ...
4 components de quoi? de I'index! [Reads]... posted gains since February ... hum, [li:d] strike
5 ou [led]? ... [Takes Webster] ... bon, y a rien, y a pas [li:d] strike dans le dictionnaire, on
6 va regarder le bilingue ... [Takes Collins]... [li:d] strike... [Reads] ... strike, strike, à strike
7 y a rien ... [li:d] ou [led]? non, rien a strike ... on va le traduire litteralement parce que ...
8 grève au Pérou ... a fait monter I'index encore plus haut ... Kellner note que ... fait
9 remarquer que ... nous a ... non, fait remarquer que 10 de ses 13 composantes ... de I'index
10 ... ont affiché une progression certaine, nette, depuis février ... Ça veut dire quelque chose
11 ça? [Reads] non! I'index, c'est I'index de quoi, en fait? des prix ... qui est mont6 ... ah, OK!
12 qui est monté par rapport au minimum qu'on avait connu en février ... une grève au Pérou
13 a fait monter I'index encore plus haut ... OK ... bien qu'une partie de la progression de
14 I'index soit due explicable ... bien qu'on puisse expliquer ... bien que la progression de
15 I'index soit en partie explicable par une greve ... [li:d] strike ... par une grève ... y a quelque
16 chose la-dedans ... [Takes Sylvain] pff, y a rien du tout dans ce truc-là! [li:d] strike ... ça
17 veut pas dire du plomb quand même! [li:d] strike ou [led] strike?
It is easily seen that in line 2 the student situates the theme industrial price
index leaving out spot. In lines 10-11 he notices that it does not make sense. So
he re-calls price index (line 11) and is temporarily satisfied with his self-explana­
tion, then he goes on translating and returns to lead strike (lines 16-17) and right
 118 Teaching Translation and Interpreting 2

until the end he knows that he missed out something important. He spent 7
minutes out of 15 on lead.
There is one link he did not make: the link with COMMODITY or SPOT
PRICE (see Graph 1), which was necessary to understand that lead is included
as a specific of the generic commodity and, as such, lead is a component of the
spot industrial price index. We can see (line 13) that he knows the meaning of
lead, but rejects this reference. From his questionnaire, it turned out that he knew
what an 'index' is in economics and also that lead is an important export article
in Peru. Moreover, he was well-versed in economics (as an MBA) and was an
experienced translator. Why did he fail?

Wrong paradigm
Misled by the title of the article, he was looking for an adjective modifying
strike, such as general strike or wildcat strike. I believe he was looking into the
wrong paradigm, because he could not use his thematic knowledge to the best;
he had no clear notion of commodity (he thought of consumer goods instead of
raw materials) and he did not know the meaning of spot (which he admitted
afterwards). So, we have an illustration of the impossibility of deciphering the
meaning of a text segment properly: our translator's syntactic decoding (modifier
vs. complement) is hampered by the shortcomings of his text-world representa-
tion; he had the wrong expectation about the grammatical slot of the word he is
looking for. His semantic representation of the text was too rudimentary to allow
him to accept the correct connection when he made it in line 13 ["lead, it cannot
be plomb"].

Translator B: translation and protocol

Bien qu 'une greve de production au Perou ait contribue a augmenter l'indice,
Kellner constate que 10 des 13 constituants ont inscrit des gains depuis le
mois de fevrier.
This protocol shows a different pattern - a very sketchy text-world representa­
tion - and illustrates how a good linguistic intuition can abort because of a poor
use of dictionaries.
1 [Reads] Although a lead [li:d] strike in Peru helped push the index higher, Kellner notes
2 that 10 of its 13 components have posted gains since February. Eh! que c'est compliqué!
3 lead [li:d] strike in ... des greves de ... [Takes Sylvain] index, l'indice, c'est ça index, indice
4 ... indice c'est ça; a lead [li:d] strike: lead, délai. OK, une grève de production, bien que des
5 greves de production aient aide* a hausser l'indice aient contribu6 a faire augmenter ... a
6 augmenter l'indice, Kellner constate que [Reads] components ... have posted gains ...
7 [Takes Collins] posted gains ... mmm ... que ses composants, que ses constituants [Takes
8 Sylvain] component [Takes Collins] component: composant, constituant... ça veut dire que
9 10 de ses 13 ... [Takes Collins] post... ont inscrit des gains ... Kellner constate que 10 des
10 13 constituants ... qu'est-ce que je vais dire pour posted gains ... [Takes Collins] j'ai
 Jeanne Dancette, Canada 119

11 l'impression de perdre la ménoire ... inscrire des gains ... inscrit des gains depuis février,
12 le mois de février.
Coming across lead strike (line 3), the student says: "strike in -, grève de -"
(as in greve des postes, du charbon). She expects to find the sector affected by
the strike. So far so good. She identifies the locus of the gap properly. She just
needs to look up in the dictionary under lead as a substantive. But she opens the
dictionary of accounting and accidentally finds lead time with the translation
equivalence delai de production. So she analyses as follows: The dictionary says
lead time equals delai de production, and I know that time means "dé1ai", then
lead [li:d] means "production"; so I can infer that lead strike means "grève de
production".
She was satisfied with this and felt no need to make more elaborate conceptual
links since the phrase greve de production fits. (It took her 12 seconds for that
decision; and she performed the task in five minutes and 30 seconds) Because
she translated the words, she believed that she understood it, which is not the
case. Her protocol shows that she established conceptual connections neither with
spot industrial price index, nor with commodity. Her entire reasoning occurred
within the scope of the segment to translate, and for that task, which was her
only concern, she relied heavily on dictionaries which she opened 7 times, look­
ing up index, lead, gains, post, component. Before commenting on that attitude,
I will present the third example and contrast the two.

Translator C: translation and protocol

Bien qu'une greve dans le secteur du plomb au Perou ait contribue a la
hausse de I'indiee, Kellner observe que 10 des 13 composants ont enregistre
des profits depuis le mois de fevrier.
1 [Reads] lead, qu'est-ce que e'est? ah! plomb... posted gains, afficher; il y en a qui prennent
2 leurs désirs pour des réalités ... Bien qu'une grève dans le secteur du plomb ... c'est bien,
3 oui! au P6rou ait contribue' ... aïe aïe aïe, de temps en temps j'ai des problêdmes ... [Takes
4 Collins] qu'est-ce que je regarde? [Index] ah oui! c'est bien ce qui me semblait ... ait ...
5 pushed the index higher ... comment on dit... ait contribue* a la hausse de l'indice, Kellner
6 observe que 10 des 13 ... composants ... composantes ... have posted gains ... ont affiché?
7 non, ont enregistré des gains ... de quoi? qu'est-ce que e'est que cet index ... Commodity
8 Research ... the industrial-price index ... indice des prix industriels ont affiché des profits...
9 probablement dans le secteur de ... des profits depuis le mois de février ... next!
This student is the only one who found "plomb" right away for lead. She runs
through the text (5 minutes in total) and translates with ease till she encounters
the word gains. Here, in lines 7-8, she establishes a connected network between
index, commodity research bureau and industrial-price index. However, as can
be judged from her answer on the questionnaire asking for the definition of
'index', she only has a vague idea of how an index works. So she builds a new
120 Teaching Translation and Interpreting 2

text-world representation, coherent but different from the text. Her reasoning
seems to go as follows: if the industrial prices are higher, then corporations can
make more profits; and she switches from the idea of larger profits for corpor­
ations to higher profits for the components. Her text-world representation is at
the level of the enterprise, not at the level of the raw materials whose price index
is an indicator of economic trends. So, she is in the wrong paradigm, in micro­
economics rather than macroeconomics.

Cognitive mapping vs linguistic constructivism

The last student's reasoning provides an example of 'mapping': she maps the
sentence onto a model of reality she is familiar with. Conversely the previous
student constructed a semantic representation using the form of the linguistic in­
put. So this is a case of cognitive mapping versus linguistic constructivism, as
discussed in McGonicle (1986: 143-146).

Concluding remarks
I have used these three students to discuss how semantics intersects with con­
ceptual representation; however, I remain perfectly aware that the validity of any
experimental study is limited to the scope of the experiment. I am presenting pre­
liminary results of a pilot experiment that will be further analyzed, expanded and
replicated.
Also, when dealing with protocols, it is necessary to be aware of the fact that
protocols, even when they are combined with questionnaires and interviews, are
only a posteriori and clumsy and possibly false justifications for a performance.
So experiments of this type cannot be used to verify hypotheses. They can how­
ever produce valuable data on what blocks and hampers the comprehension
process, and thus may contribute to a theory of natural languages processing as
well as improved teaching methods.
In that regard, translation that encompasses the multi-sentence contexts of dis­
course and verbal reasoning is particularly suitable for this kind of open-ended
experimenting. It shows that text comprehension is more than the interpretation
of isolated words and even sentences. Any study of discourse comprehension
must account for processes whereby language calls forth appropriate knowledge,
and processes whereby wording and grammatical construction alter this
interpretation.

Note
1. We want to thank our research assistant Stephen Dupont for gathering the experimental data
and the Social Science in the Humanities Research Council of Canada for funding this research.
 SYSTEMATIC FEEDBACK IN TEACHING TRANSLATION

Cay Dollerup, University of Copenhagen, Denmark

Translation classes and foreign language teaching

This article presents some methods and ways for systematically tackling a prob­
lem in translation teaching which is usually disregarded in the genteel society of
translation studies, namely the existence of a large interface, an overlapping area,
between foreign language learning and translation. As noted by Gideon Toury
"translation abounds in manifestations of interlanguage" (1979: 224). It is there­
fore no surprise that teachers are confronted with large numbers of foreign lan­
guage errors in translation classes, notably in the beginning. If the existence of
this overlapping is openly acknowledged, it follows that translation classes must
strive to minimise this interface between foreign language teaching and trans­
lation, and to shorten it in order to focus on translation as such. It is true that the
specific problems are not identical between all - or for that matter any two -
language pairs, but the general problem is there.

The Danish background

The feedback system I am going to discuss has been developed for Danish
university students of English in their first three years at university.
Students of English have chosen English of their own volition. Translation is,
however, mandatory, which means that classes will include students who consider
translation a waste of time as well as those who find it fun, and any of whom
may later come to work professionally with translation. In terms of language
proficiency, Danish undergraduates are good by international standards: they have
been taught English in school for at least eight years, and in addition, Danish
society is functionally bilingual in many respects thanks to a barrage of English
language material on television, radio, in entertainment, foreign trade, and
tourism.
Translation has always been an important part of foreign language study pro­
grammes in Denmark, for, in some measure or other, university graduates were
always expected to be able to translate. Yet the role of 'translation' is not well-
defined: formerly most graduates would become college ('gymnasium') teachers,
and to this day much 'translation' at college level is synonymous with grammar
drills, a practice copied from the teaching of the classical languages. On the other
hand, university graduates would also become editors, authors, scholars and pro-
122 Teaching Translation and Interpreting 2

fessional translators, and they, too, would have acquired their first schooling in
translation at university. The content in translation classes was always practical
work, but the ideology would differ according to the teacher's attitude. And to­
day, with more emphasis on theoretical approaches to translation, there are enor­
mous differences in teacher views on the objectives of translation classes.
The persistence of the grammar drill attitude can, in no small measure, be attri­
buted to the fact that English and Danish are closely related Indo-European lan­
guages which have basic syntactical and grammatical points in common. They
also share numerous words both from Indo-European origin (such as 'arm',
'house'/'hus') and from Danish introduced by Danish settlers in England in the
8th to the 11th centuries (e.g. 'live', 'sky', 'egg', 'they'). These linguistic facts
affect translation as well as views on translation: in many cases an interlinear
translation between Danish and English will make sense, although the actual
wording may jar.
Translation classes must, in my opinion, first and foremost come to grips with
typical 'interlanguage errors' in the Danish-English language opposition; this
must be done to further the main objective, namely to emphasise translation as
translation proper. Thanks to the proximity of the language pair, translation can
be based on careful textual study and a high degree of linguistic approximation
between source and target-language expressions. In translation classes for
beginners, I include tasks that illustrate fundamental translation problems and
which often lead to class discussion of major questions in translation studies. But,
by and large, the class format does not allow for taking up all general problems
in translation theory in a systematic way. This must be done in more advanced
classes and in other ways.
Classes may comprise more than forty students. Classes as large as this are
found nowhere else in the Danish educational system.
These sketchy comments serve to illustrate that my feedback is largely made
up to ensure teacher survival, a modicum of individualised feedback and an open
class discussion of points of communal interest. In some respect or other, all
Danish teachers develop their own way of surviving.

Some views behind the feedback system

In various ways, the system I am about to describe is affected by my view that
translation is a social activity, primarily as communication, but also in a larger
social context.
Communication, translation, especially between closely related languages, must
be a close semantic approximation to the source text, or, to put it in traditional
terms: it must have a high degree of fidelity. This view involves a taxonomy of
 Cay Dollerup, Denmark 123

translation as communication which covers the spectrum:

(a) an excellent translation,
(b) the minor inaccuracy where the original meaning is preserved in the target
language,
(c) a distortion, but no more so than the meaning of the original can be
grasped,
(d) an incomprehensible rendering. It may confuse, but will rarely lead astray,
(e) a self-contradictory rendering which is misunderstood,
(f) a rendering which reads fluently and makes perfect sense in the source
language but distorts the meaning of the original. (Dollerup, 1982)
Most freshman students's translations are easily bracketed as (b) to (d)s, that
is, as muddled and confused renditions, which are either understandable to the
(imaginary) target audience (b-c), or at the very least by the teacher (d).
Type f is the gravest error which can be committed in translation. However, it
is doubtful whether it can be stamped out. The best one can do is presumably to
call attention to it in classroom settings and hope to make students better at
avoiding it in professional work.
Even within these categories, there are variations in importance: a distortion
involving only a word will usually (but not invariably) be less serious than the
one affecting a sentence, and so on. Here, by way of illustration, are two ex­
amples of distortion (category f), from my classes:
EXAMPLE 1:
Source text: Da ægteparret fik t0mt skabene for en masse ragelse, fandt de obligationer for
150.000.
Target text: When the married couple had emptied the cupboards for a lot of junk, they found
bonds worth 150,000.
In this case the distortion is caused by a 'false friend' ('for') in the rendition
of the preposition: in Danish the couple emptied the cupboards of the junk. In
English they were remunerated with the junk. On the other hand, the erroneous
rendition only affects this one segment linguistically; in terms of content, the
error is subordinate in the context where the most pertinent point, identical in the
two languages, is that the couple acquired some valuable bonds.
In another example we get a speciously correct translation in Danish of the
English original.
EXAMPLE 2:
Source text: Dublin's modern progenitors were "Black Tom" Wentworth, Earl of Stafford, a
Yorkshireman, and James Butler, 2nd Earl of Ormond, an Anglo-Irish Protestant.
One Danish rendering goes: Det moderne Dublins fædre var Sorte Tom Wentworth, Jarlen af
Stafford, en mand fra Yorkshire, og James Butler, Den anden Jarl af Ormond, samt en engelsk-
irsk protestant.
This fluent rendering has, however, transformed two persons in English into six
124 Teaching Translation and Interpreting 2

in Danish. As this is the only information we have about the founding fathers of
modern Dublin, this distortion carries more weight than the previous one.
There is one further aspect which must be taken into account in an assessment
of translations. In a strict sense, all errors in translation violate the trust of
senders and addressees. The continuum outlined above takes into account the
gravity and implications of these violations, but it disregards the social dimen­
sion, for the immediate social implications of translation errors are different.
I believe that in real life most errors in the bottom category (f) are not detected.
Conversely, there are severe social reprisals for one particular subgroup of errors
in the second type (b), namely minor distortions which are fully understandable.
Errors in this category divide easily into two types: one type of errors is not ob­
vious to most non-natives (and perhaps many natives), such as unidiomatic collo­
cations and clumsy phrases, whereas the other subgroup, namely formal errors,
are spotted by every person with even the most superficial knowledge of the
target language. Misspellings and syntactical errors typically lead to the immedi­
ate loss of respect for the translator with both senders and addressees. These
latter subgroups may, of course, sometimes be identical with interlanguage errors:
with Danes, for instance, errors in concord in English are 'interlanguage errors'
because Danish has no distinction between the verb forms of the third person sin­
gular and other persons.
Errors of this type belong to the social dimension of translation: they have little
to do with textual proximity and semantic fidelity. Yet this societal parameter is
so important that it, too, must be taken into account when student performance
is assessed and it must be included in any feedback in translation teaching.

The physical framework

In my translation classes for beginners, students receive a booklet at the be­
ginning of term. The booklet contains a series of instructions, general advice,1 a
page of the notations which will be used, that is signs, symbols and abbreviations
used for correction, a copy of the feedback sheet, and all texts to be translated.
All texts are authentic and unedited texts, selected because they present real-life
translation problems. They are considerably more difficult than the texts given
at the finals. Most of them have been used previously in classes. The first texts
are very short, down to fifty words and they tend to be non-fiction. Gradually
they become longer - up to seven hundred words - and as we advance there will
be more literary texts. The average length is 300-350 words. Translation work
goes both ways, from Danish into English and from English into Danish. The
Danish-English directionality is dictated by the real-life fact that most English
translations from Danish are made by Danes as Danish is not known by many
 Cay Dollerup, Denmark 125

foreigners. The translation into Danish is based on the implicit understanding that
graduates must also master Danish.
The first major problem in class is that undergraduates will know 'translation'
from the 'gymnasium' (which corresponds to Die Oberstufe, the lycée, the two
first years at American Universities). It means that students have got the im­
pression that translation is a kind of disguised grammar drill, and, accordingly,
that the main point is to figure out where the teacher set the traps - one, two or
three - depending on the length of the sentence. This implies that the rest of the
translation of the source text - usually phrased in immaculate Danish - is to be
a word-by-word rendering. For this reason the first texts I use in classes are very
short to make students realise that this procedure does not work in real
translation.

The feedback
The feedback is given in class and consists of three components, namely cor­
rections in the translations which the students have handed in (shown on page
126); an oral discussion in class covering adequate as well as inadequate ren­
ditions (directed by a 'model', page 127); and a feedback form assessing
strengths and weaknesses with each student (page 128).

The feedback in the translation

The first type of feedback is found in the student's own translation. It is
returned individually to students at the beginning of the session, with the cor­
rection signs and symbols listed and explained in the booklet.
The point of departure is the student suggestions in the target language. This
means that the primary yardstick is the general linguistic competence of indi­
vidual students as evidenced in that particular translation; in other words, the
stick is held higher with good students than with poor ones.
In the same vein, teacher solutions suggested retain as much of the students'
phrasing as possible in order to allow for a translation adapted to each student's
personality, rather than the best way out. Incidentally, I rarely suggest alterna­
tives, or mention why a given solution is bad, unless the student's problem will
not be discussed in the class session at all, because it falls outside any pattern
which we have time to discuss.
Two student translations - shown on the next page - will serve to illustrate this
correctional part of the feedback system, although they will not allow of a de­
tailed discussion of the teacher responses.
With their corrected translations in hand, students gain a general idea of which
solutions were brilliant, which were acceptable, and of any specific objetions on
126 Teaching Translation and Interpreting 2

Two student translations with correction marks

The most remarkable thing about the church of Vestervig

is the size. It appears enormous in the landscape. It is
the second largestJut(ishandic) ashlar church. Only the
cathedral in Viborg is larger. As a result of a necessary
repairworkabout 1920, it may appear glossy, but it is definit-repairer???
ly(is)one of the most important monuments of the Middle Ages.

It is well worth a visit. Everything is inluenced by

tourist in cars : a fine parking ground is placed Just ag??? ?
across the church.

The thing you notice the most about Vestervig Church is its size.
It looks hugely in the landscape and after Viborg Cathedral it
is in fact the largest church built out of ashlar altogether. It
may seem a bit finicky- atheresult of a by the way much needed
renovation from around 1920, but it definitely belongs withamonyour
most important monuments from the Middle Ages and it is worth a
visit. They have planned everything for driving tourists as well-such?
a supercool parking lot right across from the church.
 Cay Dollerup, Denmark 127

Excerpt of model translation

Danish source text: Det mest pafaldende

Adequate solutions: What is most remarkable
The most remarkable feature about
striking
Inadequate solutions: no prop-word

Danish: ved Vestervig Kirke er st0rrelsen. Den syner kolossalt i

Adequate solutions: + Colossal in the landscape, it is.
the Vestervig Church is its sheer size. It dominates
the Church of Vestervig is its size. It stands out
Vestervig Church is the size of it. enormous the
It looks colossal in
vast
Inadequate: bulk seems
dimensions appears

It looks huge in the -

Danish: landskabet, og næst efter Viborg Domkirke er den faktisk

Adequate: Next to the Viborg Cathedral it is, in fact, (by far)
scenery (?). in fact, second to
landscape . It is, actually, next to the Viborg Cathedral,
Inadequate: countryside
- surroundings

Danish: den st0rste jyske kvaderstenskirke overhovedet. Den kan virke en

Adequate: the largest of all Jutish ashlar churches. may seem
the largest ashlar church in all Jutland. It seems
the largest Jutland (?) ashlar church. Some will find it
Jutish
Inadequate: at all It may/can look
altogether. It may come across
built of ashlar altogether.

Danish: smule slikket - et resultat af en i ørigt nødvendig istandsættelse...

finicky which is due to some urgent repair work
smart which is the outcome of a - necessary - restoration
a bit trim which is due to some urgent repair work
shined up which is the result (??) of a restoration, which was, incidentally,
sleek
licked which is a result of an otherwise much needed renovation
128 Teaching Translation and Interpreting 2

No marks = The feedback form

No problem for you, OR
Not checked in this translation

1. TEXT
omission
addition
insufficiently checked (Tense. Numbers. Other)

2. SPELLING
capital letters
words
compounds
split words
3. PUNCTUATION
relative clauses
object clauses
other (Longish discourse. Adverbials. Other)

4. WORDS/ WORD KNOWLEDGE

elementary
rare
idioms and phrases
constructs
plural vs singular forms
caiques
false friends
contaminations
equivalents
irregular verbs
change of word class
gender
5. SYNTAX/GRAMMAR
concord (subject - verb)
concord (other)
genitive
article (Indefinte. Definite. Form)
preposition
adverb, form
adverb, position
prop-word
tense ........'......
modal verbs
parallelisms
relations (Parataxis. Hypotaxis)

6. EXPRESSION
collocations
calquing
construction of sentence
idiomatic usage
style
precision
word order
7. OTHER COMMENTS
 Cay Dollerup, Denmark 129

my part. Students soon grasp the differentiation in the system: it moves from a
plus (+) for fine phrasing over a dot under words I object to, a wavy line for
unquestionable mistakes, to a straight one under real howlers, namely the formal
errors. Students therefore immediately know whether we are concerned with so­
phisticated points in syntax or something that should turn up once only in a uni­
versity undergraduate career.
Students also soon catch on to the fact that we are operating with fine shades
between the two extremes, namely the plus as opposed to the circular arrow sig­
nifying a 'distortion'. Since there are rarely indisputable 'teacher improvements',
most signs only serve to alert the students to the fact that they must keep track
of what is happening in class.

The oral analysis in class

The second component in the feedback is oral. It is a class discussion of
solutions.
In translation, there are usually several equally valid or at least adequate
renderings in terms of individual words, idioms, clauses and sentences. I consider
it important that students are introduced to a multiplicity of valid solutions in
translation work. For this purpose I make up the model shown on page 127,
which covers the opening sentences of the exercise we saw in the previous illus­
tration. The chart notes all student suggestions. In class all adequate - and (why
not admit it?) brilliant - solutions which have surfaced in student translations (or
mine) are presented first. Deviations between style and content in adequate tar­
get-language renditions are discussed. Attention then focuses on inadequate ren­
derings. Like the correct suggestions, they are read out one by one, and students
are asked to specify the inadequacy or distortion. This is done by demanding an
explanation of what the infelicitous renderings mean. The erroneous renditions
are discussed in class both for the benefit of those who did not write them and
for those who did: everybody hears an explanation as to why the suggestions
were not adequate, and, hopefully, a heightened awareness of language use.
There are a few points to be made clear: I never, ever, identify students who
have committed a specific 'error', but it often happens that students give them­
selves away, for instance by laughing when they realise what is wrong; and it is
rare for me discuss elementary howlers because they are corrected individually
in the written translations. The approach makes for lively discussions where
points in the translation may illuminate central principles and problems in trans­
lation, as well as confrontations between student views and mine, perhaps even
corrections of the latter. And, thirdly, since I find it very depressing not to laugh
at times, I will take in some tricky, or funny, errors made in previous classes.
130 Teaching Translation and Interpreting 2

The feedback form

The third component is the feedback form (shown on page 126). It is a system­
atic and individualised assessment so that all students receive an evaluation of
their translations.
Comprising a total of 42 problem areas in interlingual transfers between Danish
and English, loosely grouped as 'collation', 'spelling', 'syntax' and 'expression',
the feedback form covers most errors generated in translations between Danish
and English, and, in my opinion, they also move from formal errors and mani­
festations of interlanguage to points which are more pertinent for 'good' trans­
lation work. Each area is assessed on a five-scale differential which is filled out
for each student translation: I have a look at each translation, and, then, taking
into account the overall possibilities for committing errors of a given type, I in­
dicate all problems revealed in the translation. If, for instance, I know that it is
possible to commit five errors in concord, and I find two in one translation, this
is marked in the middle category: a warning to the student that this may be an
overall weakness. The feedback is thus individualised at this point. No two feed­
back forms are completed in the same way. Of course, no single translation exer­
cise will generate all points, nor indeed will individual students make the errors
to the same extent in every exercise. But when students compare these feedback
forms after, say, ten translations, they can identify their own weak spots and do
something about them.
My use of the evalutation sheet is not the same with beginners and advanced
students: initially the top differentials which cover formal points are used fairly
frequently, and it must also be stressed that I mark only errors. In the advanced
classes, it is the strengths and the stylistic points at the bottom which are marked.
Doubtless, it has been noted that I have not discussed the use of computer
spelling and grammar programmes in translation classes. I certainly accept their
existence, and appreciate that they will change the translator's workplace. Yet in
teaching, I believe we must operate on the premise that we are educating special­
ists who can also manage competently on their own without access to machine
translation. In that context, it is a weakness that the existing systems are either
cumbersome or too rigid in their approaches to allow for creativity in translation
work: this, on the other hand, is appropriately lauded in my classes.

Discussion
The procedures presented here are not the be-all and the end-all of translation
studies, for despite all the technicalities there is still subjectivity galore. I am not
infallible, and I do not catch all student errors (In the first example given, for
instance, there are a few slips: "touristy", "across from", and "beginning"). Stu-
 Cay Dollerup, Denmark 131

dents will ask me why I have not corrected errors I discuss in class in their trans­
lations. Also, it may turn out that in the course of my correction work, I change
my mind - true, in the recorded history of Dollerup classes, no more than one
grade, but nevertheless enough to make students aware that protests may bring
forth some public teacher contrition which can be used for individual self-repair
and face-saving. Similarly, students occasionally catch me out in a solution which
I have classified as wrong, and point out that it is correct, for instance, by inter­
pretations overlooked, in dictionary entries I missed in my work, or simply by
having a referendum in class on proper language usage. Or it may be that their
and my language usages differ, or perhaps even that we are dealing with seman­
tic shifts due to language change. In these cases my way out is to have a public
vote.
As far as the evaluation sheets are concerned, students will turn up to ask
probing questions about why I have not marked particular types of error which
they have noted were discussed in class - where the only proper answer is a
blush of shame. The model shown on page 128 is actually the eighth one. I occa­
sionally find weak points about it - points that I have not included, or that the
sheet covers irrelevant matter. But then the next batch of translations makes me
change my mind again and stick to the version I use.
These are minor details: what matters most in translation classes is that students
become aware of linguistic problems, formal as well as semantic ones, and use
this knowledge in subsequent translation work. It is more useful to heighten the
undergraduates' general linguistic sensibility than to correct one specific error,
so if I overlook something or judge too harshly, it is all done in the pursuit of
that honourable goal.
It is in the nature of things that all teaching must work, otherwise life as a
teacher would be hell. As far as I can make out, the feedback discussed does
work: my students appear to do away with the formal errors; they seem to fare
slightly better at exams, and some of them remember classes years later. But of
course this may be an illusion.
Finally, no two classes of students are the same: procedures, advice, and
teacher performance may change - or may have to change - with new classes. So
there is a subtle interplay between teacher personality, teacher feedback, student
personality, teacher and student idiolects and sociolects, class size and general
knowledge of the languages used and the translations done.
These differences are large even in a small country like Denmark. Therefore
it would be presumptuous to suggest that the procedures presented in this article
are applicable in classrooms other than mine. However, the bottom line is that
by an identification of the most prominent errors in translations between specific
132 Teaching Translation and Interpreting 2

language pairs, and by calling students' attention to them systematically, these

errors can be eradicated, with blood, sweat and tears. Accordingly, some of the
features and some of the ideas discussed here may inspire others operating with
other language pairs to do something, either along the same lines or in radically
different ways, to improve feedback in translation classes in different climes and
under foreign skies.

Notes
1. Among the pieces of advice is a general exhortation to translate texts other than those used in
the classroom and then to discuss these translations in study groups outside class (which usually
means at somebody's home). This makes for peer-correction which I find a very effective means
of doing away with elementary howlers.
 STUDENT-CENTRED CORRECTIONS OF TRANSLATIONS

Maria Julia Sainz, Universidad de la Republica, Uruguay

Background information
At the University of the Republic of Uruguay in South America I teach trans­
lation from English into Spanish. The Translation Course is one of the courses
taught at the University's School of Law. It is placed there because the degree
awarded is that of Public or Certified Translator, that is, it allows graduates to
translate and sign public documents, which in this way become official trans­
lations.
In 1885, the School of Law of the University of the Republic of Uruguay (the
only state university in the country) was authorized to issue the degree of Public
Translator as a way of regulating the work of translators. At the official level in
Uruguay, translations done by Public Translators have become mandatory since
the late XIX century for all documents either coming from other countries to be
submitted to State organizations or for documents issued by Uruguayan organ­
izations to be officially considered abroad, such as letters rogatory, powers of at­
torney, decrees, awards, affidavits, certificates, and financial, commercial, and
technical documents in general.
Article 28 of Act No. 12.997 dated 28 November 1961 defines the Public
Translator as "a University professional who has a liberal career and is a trustee
of public faith whose activity is mainly intellectual".
After many changes in the course of this century, the School of Law has had
a four-year course leading to the degree of Public Translator since 1976. It is the
only institution in Uruguay where such a course can be taken. Like all the other
courses at the University of the Republic of Uruguay, it is absolutely free of
charge. Students are admitted irrespective of age, sex, race and colour.
To enter the course, candidates must have finished their secondary education
(six years) in Uruguay and must sit for an entrance exam both in Spanish and in
the language(s) to be taken. This exam ascertains whether candidates have a suf­
ficiently adequate knowledge of those languages with which they intend to work.
The specific objective of the School is not to teach languages but to train
students to become high level specialists in a profession which uses languages
as its main tool. To pass the entrance exam in English, candidates are required
to have a level similar to that of the Cambridge Proficiency Exam. (See Sainz
134 Teaching Translation and Interpreting 2

1992: 69)
The language pairs offered by the School (English, French, Italian, Portuguese
and German) are always in combination with Spanish, and students are required
to work in both directions.
Courses run from March to October and the evaluation system consists of five
tests in the course of the academic year. With a work load of around 15 hours
per week, the subjects in the four years are:
First year: Foreign Language I
Spanish I
Applied Linguistics
Public Law
Second year: Foreign Language II
Spanish II
Theory and Methodology of Translation
Private Law I
Third year: Foreign Language III
Culture of the Foreign Language
Professional Practice I
Private Law II
Private Law III
Fourth year: Foreign Language IV
Professional Practice II
Foreign and Comparative Law
Language Workshop.
Because it is a part-time programme, courses are scheduled in the late after­
noon or early evening. In order to attend any given year, students must have
passed all the subjects covered in the preceding one.
The University does not offer any courses in interpreting, but rules and regu­
lations state that Public Translators may act as official interpreters in case of
marriages (when one of the spouses does not understand Spanish), wills, judicial
proceedings in which non-speakers of Spanish are involved, etc.

Evaluation system
I will start the discussion of my topic with a Socratic thought. Socrates said
that ideas exist but are scattered: all we have to do is find them. The idea I
would like to awake here is one of the topics I feel most concerned about, since
it has to do with the type of evaluation system that we, as teachers, choose. I am
not thinking of tests or end-of-year exams but of how these tests and exams as
well as any passage we ask our students to translate can produce useful feedback
for our courses at an on-going micro-evaluation level (as opposed to macro-
evaluation, carried out by authorities at the national or international level).
In their book Adult Learning Principles and Their Application to Programme
Planning, Brundage and MacKeracher (1980: 21-31) established some principles
 Maria Julia Sainz, Uruguay 135

which make us see the need for a student-centred approach to learning which can
be applied to our translation classes. These are some principles:
1. Adults learn best when they are involved in developing learning objectives
for themselves which are congruent with their current and idealized self-
concept.
2. The learner reacts to all experience as he perceives it, not as the teacher
presents it.
3. Adults are more concerned with whether they are changing in the direction
of their own idealized self-concept than whether they are meeting standards and
objectives set for them by others.
4. Adults do not learn when over-stimulated or when experiencing extreme
stress or anxiety.
5. Those adults who can process information through multiple channels and
have learnt 'how to learn' are the most productive learners.
The term "student-centred" has been fashionable in language teaching for some
time now. It can describe different ideas: a student-centred class; a student-
centred course; a student-centred approach, etc. I am going to discuss a student-
centred approach to correction of translations at our university in Uruguay.
Some students have never been asked to think about themselves as students nor
to answer questions about their ways or strategies of learning, furthermore, some
never thought they had any strategies of learning at all. They need to be made
aware of all this and to derive benefit from their past experience in the learning
process.
In Confucian terms "If you give a man a fish, you feed him for a day. If you
teach him how to fish, you feed him for a lifetime." I designed the questionnaire
in Illustration 1, on the next page, in order to sensitize my students to their own
position regarding the translation course they had chosen. I suggest this
questionnaire can pave the way to teach students how "to fish".
Teachers must make it clear that there are no right or wrong answers to these
questions and that the students' answers are going to be used only as feedback
and food for discussion later on. The purpose of the questions is also to give
students a wider insight into their own learning process and make them aware of
the rationale underlying class activities.
Some students find it difficult to give accurate answers (or at least the type of
answers that can be useful for teachers). Therefore, teachers can help them to cla­
rify their thoughts and avoid answers which are too generalised or vague.

Student-focus process
I designed a process which I called "student-focus process" to help both
teachers and students become aware of the learning process involved in cor­
recting translations. In other words, how to learn from mistakes.
 136 Teaching Translation and Interpreting 2

Illustration 1

Student's name:

QUESTIONNAIRE

1. WHAT DO YOU EXPECT FROM THIS TRANSLATION COURSE?

Learn more about language

Learn more about the culture/literature of the language
Keep in touch with a language I already know
Other. Specify

2. WHAT DO YOU THINK MAKES A GOOD TRANSLATION STUDENT?

3. WHAT DO YOU THINK IS THE BEST WAY TO LEARN TRANSLATION?

4. WHAT TYPE OF ACTIVITIES WOULD YOU LIKE TO DO IN YOUR

COURSE? WHY? For example:

FORMAL LEARNING INFORMAL LEARNING

... grammar exercises ... reading in the SL and TL
for voc, register, style,
etc.

... doing translations at home ... comparing official

translations

... pair-work translations ... watching a film / video +

subtitles

5. HOW DO YOU ORGANIZE YOUR LEARNING? Material you use:

(monolingual-bilingual dict.; grammar books; your own notes)

6. HOW WILL YOU ORGANIZE YOUR TIME FOR THIS COURSE?

 Maria Julia Sainz, Uruguay 137

This process is not only about what our students want (and need) but about the
kind of students we as teachers want (and need). It is not only about who our
students are but about who we are and about what level of teaching they are en­
titled to as learners. By this I mean whether we teachers adopt a human rights
based approach to translation teaching at the University when dealing with
learners' rights. Following Francisco Gomes De Matos' "Checklist for Teachers"
(1991: 256), this approach could also be called "student rights based". The list
goes like this:
Do my students have the right to make translation errors?
What kinds of errors?
Are they told about the typology of the errors referred to?
How empathic can I be, when evaluating translation accuracy and appropriateness?"
The process I have in mind is shown in Illustration 2:

Illustration 2: The student-focus process

SELF-
MONITORING

INTEGRATION

MONITORING

IMPLEMENTATION

DEVELOPMENT

The process comprises various stages. On the bottom line of the pyramid we
have the DEVELOPMENT stage intended to understand and anticipate students'
needs in order to respond to those needs more efficiently.
How do we build understanding and how do we retrieve feedback? I can see
how students performed at the entrance exam for the translation course, con­
sisting of translations from and into Spanish plus a 'Use of English'. This helps
me to build up knowledge about my students and anticipate what level my first
year class has. Although, as I said before, the level required to pass the entrance
exam in English is equivalent to the University of Cambridge Certificate of Pro­
ficiency, there are problems that appear in translations which are quite different
138 Teaching Translation and Interpreting 2

from those where the language is used for other purposes (for instance com­
position,précis writing, grammar exercises).
As teachers, we should establish common ground from the very beginning.
Students have the right to know the evaluation system we will be using to
evaluate their translation work throughout the year and we should be consistent
in its use. They have the right to know who is in charge of judging their work,
who is fully responsible for it, and who they have to address in case of further
questions. Let me give an example: I tell my students that the translations they
are handing in will be corrected exclusively by me (not by my assistant or by the
two of us) but only if this is really the case.
In order to make our correction clearer, it is practical to ask students to leave
a margin on the left-hand side of the paper on which they write their translation
and to skip every second line. This margin is divided into a column for Serious
Mistakes (S) and one for Minor Mistakes (M).
What strategies can be devised for correction so that it becomes a way of
learning instead of a source of fear, stress or punishment for students? It is not
my aim here to draw the line between what may be considered serious or minor
mistakes. Accuracy and appropriateness must be evaluated depending on the
teacher's aim for that particular translation passage. A given mistake can be con­
sidered serious or minor according to that aim.
When correcting translations, I underline the word, phrase or sentence where
I believe there is a mistake and I put a "1" in the margin of the line where a
minor or a serious mistake appears under the corresponding column. This is my
only way of saying there is something I do not "like", a non-aggressive way of
giving students feedback on their errors. The traditional method of re-writing the
correct version on the students' sheet is, to my mind, very disruptive, frustrating
and stressful for students who have made conscientious efforts to carry out their
assignments.
In the IMPLEMENTATION stage, students get the 'Correction Chart' shown
on the opposite page. The chart is divided into four columns: Mistakes/Possible
Correction/Source/Type of Mistake. Under 'Mistakes' students write the word,
phrase or sentence which was underlined as incorrect in their translation. Under
'Possible Correction' they try to produce an "error-free" version. They may do
so on their own since, quite often, by simply being made aware of their mistake,
they can produce a correct version.
However, it may happen that they are unable to do so and may have to resort
to their classmates. It is here that peer work and peer correction prove invaluable
tools. Peers may provide students with answers which they consider correct.
Unlike students in many other countries, Uruguayans tend to ask their peers for
 Maria Julia Sainz, Uruguay 139

Illustration 3: Correction Chart

Student's name:
Text File No.:
Date:

Mistakes Possible Source Type of

Correction mistake

an answer first, thus welcoming peer correction.

If no peer can help them, students may want to consult a reference source (that
is a dictionary, a glossary, their own notes, etc.). If there is still no satisfactory
answer, they may resort to their teachers.
The source of the answer for their correction is entered under the column
'Source' as: 'Myself; 'Peer'; 'Dictionary'; 'Teacher'.
The column 'Type of mistake', filled in by the students, can become a good
exercise to help students recognize what types of mistake they are making and
consequently eliminate them.
While students are filling in the 'Correction Chart' teachers can walk around
the class checking that students are filling in the chart and help them out if
necessary. This allows for a very personal and individual contact with each
student, an effort highly appreciated, especially if classes are somewhat large
(15+). This is the stage I call MONITORING. Teachers can monitor the process
in order to make adjustments as the course unfolds, on the basis of the inform­
ation they retrieve from the 'Correction Charts'.
Once students have gone through all their corrections in this way, they have a
clear picture of their main difficulties. Sometimes students do not know how to
make progress in their work simply because they cannot identify their own weak
points. If they find they have been able to correct most of their mistakes on their
140 Teaching Translation and Interpreting 2

own (many attributed to 'Myself in the 'Source' column), this may well show
that they are not meticulous enough but they know how to do it well and that
they should show more care in the future. If the word 'Peer' comes up too often,
it may mean that they are performing below the average. If the word 'Teacher'
outnumbers the rest, it may be that that particular text is beyond the students'
level. This is a most important piece of feedback which teachers can take into
account next time they give out a text to translate.
Once finished, the 'Correction Charts' are handed in to the teachers, together
with the translation, so that they can be checked more closely after the class.
Teachers can make any corrections that either escaped them in class or could not
be done for lack of time. Then the charts are given back to the students to be
filed.
This is followed by the INTEGRATION stage. Teachers can fill in their own
chart of 'Types of Mistakes' for a particular translation piece. This chart may
vary from one passage to another. It may, for instance, include the following list
of types of mistakes:

Connectors Prepositions
Grammar Punctuation
Lexical items Style, register
Misunderstanding Syntax
Nouns (agreement) Tenses
Omission

In order to fill in this 'teacher's correction chart', teachers can ask the class for
the maximum number of times each type of mistake appeared in the translated
text. By analysing this chart the necessary remedial or reinforcement work can
be integrated into class work. In this way, the feedback received from the
students can be turned into effective measures for working on those mistakes im­
mediately afterwards.
Teachers can also encourage self-assessment by their students since "an im­
portant supplement to teacher assessment, self-assessment provides one of the
most effective means of developing both critical self-awareness of what it is to
be a learner, and skills in learning how to learn" (Nunan 1984: 116).
In student-centred correction, learners must have the chance of monitoring their
own work. SELF-MONITORING is a stage during which students can check
their own progress in the course and, at the same time, become critical about
their learning. By screening their performance in various translations during the
course of the year, they can see for themselves where their difficulties lie.
 Maria Julia Sainz, Uruguay 141

At the bottom of the 'Correction Chart', students are asked to circle the figure,
ranging from +3 to -3, which they think best matches their idea about their
performance in that particular translation passage and to make any other
comments. Students have the right to express what they think about their per­
formance as well as to voice their opinion of their teachers' evaluation.

Conclusion
The charts and questionnaires were designed with translation students in mind.
They are an attempt to try to dispel the element of fear or stress implicit in any
written assignment. However, they are subject to constant revision and changes
in order to be improved.
If we do not take our students and their rights into account, we run the risk of
creating unaware and selfish professionals in the future, professionals who have
never been given the chance of developing their own opinion about their work
and who are unable to judge whether their work is accurate and appropriate
simply because nobody ever made them think about it when they were studying
at university. My contention is that we should build these "little cells" of
awareness in our translation students.
Small changes can sometimes produce great effects. If we do not do things
well in education, we see the repercussion in 10 to 20 years' time. In this light,
I would like to reword what I said at the beginning about the kind of translation
students we want and end on a question: "What type of translation professionals
do we want?"
 STARTING FROM THE (OTHER) END:
INTEGRATING TRANSLATION AND TEXT PRODUCTION

Arnt Lykke Jakobsen, Copenhagen Business School, Denmark

How I used to teach translation

I would like to begin by telling a true story from the not so distant past. It is
the story of the way my colleagues and I taught translation at the Copenhagen
Business School.
Let me mention some of the useful things we did, lest they be forgotten.
- We told students about the standard sources of information, not only printed
and electronic dictionaries, but also encyclopedias, textbooks, manuals, etc.
- We also told them about sources of potentially relevant information beyond
the standard sources.
- We told students that when working on a translation, they would probably
find more relevant information in parallel target language texts than anywhere
else.
- We told them that if they used bilingual dictionaries, they should always con­
sult monolingual, target language dictionaries for verification.
- And we struggled to convey to them our own love of words and to com-
municate our own insatiable appetite for knowledge.
What we also did, however, was give students a source text we had found, in­
struct them to go home and translate it for next week - availing themselves of as
many relevant sources of information as possible. After finishing and handing in
their translations, usually students would have to wait another week before seeing
their work again. In order not to waste valuable time, however, we asked them
to do the next translation in the course of that week, while we were correcting
the previous translation task. So, by the time we met again in class, our heads
would be full of the errors students had made one and sometimes several weeks
earlier (and forgotten all about), and their heads would be full of the problems
in a text which would not be discussed until the following week.
Not the most harmonious meeting of minds, obviously. Small wonder that our
students were not inspired and not motivated by the texts they were asked to
translate and that the same errors kept returning with nightmarish persistence
despite our intuitive or systematic efforts to get rid of them. The wonder is that
we thought for so long this was the way to teach translation.
144 Teaching Translation and Interpreting 2

Diagnosis of false assumptions

Part of the explanation for our misconceptions about teaching translation is no
doubt historical. The grammar translation tradition has been incredibly strong in
many countries - and for centuries.1 Though many teachers hesitate to accept it,
the truth of the matter is, as has often been demonstrated, that the main emphasis
in such correction work has always been on formal aspects, spelling and punctu­
ation, and on grammar (de Beaugrande 1984: 6). In addition, the translation tasks
were often so difficult that students would spend hours searching dictionaries for
possible hints in a bewildering array of potential equivalents; equivalents of
words they would probably never themselves have dreamed of using. So, what
went by the name of translation was obviously first of all an attempt to teach the
foreign language by focussing on grammar and vocabulary. If students learnt to
translate in the end - and I think the method does not entirely rule out this pos­
sibility - we can still ask if there are not more effective ways.
Perhaps the least satisfactory element in the old method was the implied as­
sumption that a translation departs from a source text only. It is a very persistent
assumption that translation is a sequential process moving, metaphorically speak­
ing, from left to right, from a source text through a translator to a target text.
That translation is essentially an interpretative activity. That a translation is only
a reproduction of another text. That a target text is not a text in its own right.2
This leads to the assumption that translation can be understood as the applic­
ation of a combination of transfer rules and translation strategies to a source text,
again without reference to the purpose, point, function or skopos of the target
text. It also implies that it is possible to approximate the ideal translation of a
given source text through a process of gradual optimisation of equivalence re­
lations between the target and source texts.
But translation cannot be understood or explained in those terms. Target text
production is never mere source text reproduction (Jakobsen 1993: 161). A target
text which is the result of the application of transfer rules and translation strat­
egies to a source text is bound to be a pseudotext. And if we assume that for
each source text there is one, and only one ideal target text - the teacher's
version - we make no provision for the creative variety and flexibility of target
text production.

Translation begins with a target

For the above-mentioned reasons, the discussion as to whether translation starts
with a source text or starts somewhere else may not be entirely trivial. I have
come to the conclusion that for the purpose of teaching translation at the
Copenhagen Business School it is preferable - for pedagogical reasons - to
 Arnt Lykke Jakobsen, Denmark 145

change the spatial metaphor and argue that translation begins with a need for a
text. It starts with the target text, and with all the usual questions we ask our­
selves when we sit down to write a text

Translation is text production

If we ask what translation is like - in behavioural terms - surely the most ob­
vious point of similarity is with writing texts. Translation is a type of text
production.
I shall not attempt to say what ordinary text production is like, and I certainly
do not wish to say that translation is ordinary text production. It is not difficult
to think of features that make translation special. For instance, if we think of
inventio3 in classical rhetoric, the first phase of language production where a
writer gets his ideas, we find important differences. And even more so in the
second phase, the dispositio, where the author's ideas are organised. The prag­
matic characteristics of translation are also different from those of ordinary text
production. Hence the familiar questions about the extent to which it is the
author/sender or the translator who communicates through the text. Here, one ex­
treme position is to argue that the translator does not communicate at all, but
merely provides a text which enables other people to communicate. The other ex­
treme regards the translator as an author, not a mere producer of texts.
But again, for pedagogical reasons, I would suggest that we begin by emphas­
ising the basic similarity between translation and text production: in each case the
primary task is to produce a text.
On closer inspection, we also find that several of the apparent differences are
not so fundamental after all. In the inventio phase, for instance, it may be true
that text production results from a bright idea suddenly invading an author's
mind, but it is also true that one may get bright ideas from reading other texts.
In accordance with the ecological order of the day, textual material is constantly
produced from recycled waste - with or without the word processor's cut-and-
paste function.4
The fact that the dispositio seems to leave the translator little room for ma­
noeuvring does not define translation either. Lots of text types have an obligatory
sequence of elements. Conversely, in functional translation, a translator may be
free to reorganise matters dramatically.
And as for pragmatics, the world is full of text production situations involving
cooperation between several parties.5 Letters may be dictated, or written from a
brief oral instruction, or from somebody else's notes. This is a frequent form of
cooperation between decision-makers and communicators, and the communicator
often has to fill in new material.
146 Teaching Translation and Interpreting 2

Texts that result from cooperation are frequently written either in response to
another text (a letter is often a reply to another letter), or on the basis of other
textual material. The editor or copy-writer of a newspaper reads reports, faxes,
and telexes from international news agencies before writing or rewriting an
article.
It is also characteristic of text production in general that textual background
material is often available in a form that makes it necessary to summarise here,
to elaborate there, and to change the style of a third passage, etc. It is this inter­
face between translation and non-translational text production, which makes it
possible to integrate writing and translating.

Integrating translation and writing

Believing that writing comes before translating, we now begin by training writ­
ing skills at the Copenhagen Business School. Writing tasks are specified with
respect to situation and purpose and based on a variety of textual material, in­
cluding parallel texts. Some of the background material provided for a writing
task may be in a foreign language. Thus, if students are learning to write texts
in Danish, some of the background material may be in English (or another
foreign language), but students are not invited to translate. The amount of back­
ground material in other languages may then be increased up to a point where it
is difficult to tell whether students produce their texts with or without translation.
The progression I am describing here, still belongs mainly in the 'visions' sub­
category in the present volume's triad of "insights, aims, visions". The course is
still in the design phase. I am convinced, however, that placing translation within
the whole spectrum of text production, and consistently exposing students to
authentic parallel texts, will help them develop greater critical awareness of ac­
ceptability norms and textual models in the target language, both when they are
translating into their native and into a foreign language. Also, by not demanding
real translation until quite late in the course, we believe we have minimised the
risk that translation tasks result in pseudo-text production.
Thus, by starting from the other end, by starting with a clear view of the target
text, we improve translation.

Process orientation
We also need to address the other main problem in the nightmare version of
teaching translation, however. This was the lack of motivation among students
due to our use of uninteresting source texts which often seemed to have been
chosen primarily to make sure that there would be a certain quantity of errors in
student translations of them. This is something we hope to remedy by process o-
 Arnt Lykke Jakob sen, Denmark 147

rienting translation teaching, a pedagogical point which follows naturally from

the emphasis on text production in the course.
Over the past few years, there has been a great deal of interest in process
oriented composition in Denmark.6 This approach, which is heavily indebted to
American ideas about composition, is now moving into the Danish school system
at all levels.7
In process oriented translation, students and teachers are active at the same
time. Thus, tuition enters into the student's world while a source text is being
interpreted, while potential equivalents are being considered, and while the target
text is in the process of being created. Tuition comes in at a point when texts are
still 'warm'.

'Warm' texts
A text is 'warm', first of all, if it is needed for a genuine communicative pur-
pose. Therefore, specimen texts must be given low priority. However, this re­
quirement is not at all easy to fulfil in class. At the Copenhagen Business School,
we would like to teach students to write or translate a variety of administrative,
legal, commercial, and technical texts, but it is difficult to create a situation in
which students feel that such texts have a genuine communicative purpose. On
the other hand, at our educational level, we cannot have students only writing
letters to their fiancé(e)s, or applications to get into exchange programmes, and
we feel that there are already enough student diaries in the world.
The challenge is clear. Writing tasks must be maximally relevant and inter­
esting to students while involving all the language elements required by the curri­
culum. Charging students with giving in-house information (in a foreign lan­
guage) to exchange students is an example. This provides a genuine communi­
cative purpose, and it involves a fair degree of professional language although the
range of text types may be limited. Another possibility is peer group review of
texts. The prospect of having a text subjected to peer group review will often in­
crease motivation considerably. Students know their fellow students, know how
to deal with their verdicts, and so tend to respect this kind of feedback (Kock and
Tandrup 1989: 16).
A text is also 'warm' if it is still in the process of being made. One main
reason why students forgot all about their translations in the old system was the
simple fact that we begin to forget a text the moment it is completed. Time is not
the crucial factor here. What matters is whether or not a text is finished. If the
text is still in the process of making, feedback received a week after production
may still be felt to be highly relevant.
Many translation and text production problems do not have simple right-or-
148 Teaching Translation and Interpreting 2

wrong solutions. Composition workshops allow teachers a better view of text pro­
duction than the finished products, and traditional student errors are not the most
interesting problems in such an environment. Here, translation and text product­
ion may be approached in a far more constructive manner. Teachers may now
bring up ideas, suggest relevant tools, solve problems - and correct errors -while
a text is still in the making, and students will immediately see and feel the differ­
ence. Teachers are no longer merely instructors and assessors, but partners in a
relevant, interesting and collaborative writing task.
This brings in a third key element in what makes a 'warm' text. A text stays
'warm' and interesting as long as a student or a group of students are responsible
for it. This is probably the crucial element, for it means that in the end we can
only expect students to learn from working with texts they themselves find
interesting.

Managing process orientation

It is obvious why some teachers have second thoughts about process oriented
class work. They would like to be able to plan and prepare classes, to know
when students are going to learn about orders and invoices, and they are not
easily convinced that students will learn proper English if all their errors are not
corrected. If we accept process orientation and if we accept that texts must be
interesting to students, are we not losing control?
In my view, process oriented class work cannot help but develop a new learn­
ing style. Process orientation means accepting that relevance criteria have to be
found in an open dialogue with students. It is never enough that a teacher finds
a text interesting. In fact, the whole issue of how much theory, how much knowl­
edge and how much skill a course should include is at stake here. Text linguist­
ics, to take an example, will not be accepted by students merely because teachers
recommend it. Unless teachers are able to make students feel - in writing, in
translating - that it makes an important difference, they are not likely to get away
with lecturing on the importance of cohesion. In accepting that relevance is nego­
tiable we are indeed giving up control to some extent, but only because we are
certain that the gains in motivation and learning are greater than the losses in
authority and control.
Some of the practical problems should not be underestimated, however. The fa­
cilities may not be available. Thus, when our own school was built in the late
1980's, nobody took into account that students might actually work there. Most
of their work was expected to be homework. Such problems can be very tangible,
but the basic problem about process-orientation is that it affects the student-
 Arnt Lykke Jakobsen, Denmark 149

teacher relationship in ways which even some students dislike. It changes our
own culture.

Conclusion: an example
Despite such problems, I take an optimistic view of all this. In a class I taught
recently, we did a role play about a chemical plant in Brazil. At some point the
student playing the role of managing director was required to make a written
statement about the company's environmental policy. The student's first impulse
was to write 'Sod the environment' - a spontaneous first attempt, which the
student rejected after a moment's reflection since it did not harmonise with what
she felt the manager ought to be writing.
The teacher's chance to work pedagogically in this new educational culture lies
in this open space immediately after the emergence of a text production problem,
when - as pointed out by Andrew Chesterman in this volume - a tentative theory
has been rejected and its successor has not yet been fully shaped. This interval
is felt to be critically important by the student, who is therefore highly motivated
to accept help. Obviously, our job is to provide relevant help to the best of our
ability and in a spirit of constructive cooperation. And without taking advantage
of the situation, we can fill this empty space with surprising amounts of theory,
grammar, stylistics, pragmatics -you name it.
It was only after several excursions into many of these areas and after looking
at many written statements both in English and in Danish that we finally agreed
that the student's fifteenth formulation of the passage in the manager's letter
would do. It now went: 'Though we have invested heavily in treatment facilities
for toxic waste, we cannot honestly claim that the environment is always out first
priority.'
In the end, it was impossible to say to what extent elements of translation
might have helped in shaping the statement, but here, finally, was an opening text
for which the student and the teacher were both willing to take responsibility, and
without thinking about it this student had also taken an important first step
towards becoming a good translator.

Notes
1. Dollerup (1993: 146): "In Denmark 'translation' in foreign language teaching at the secondary
level (lycée, college) is largely used for grammar drills in accordance with medieval practices".
2. This idea is inherent in the traditional conception of a translation as a Version' (from Latin
'vertere') of an original — whether authorised, revised or otherwise.
3. Inventio is treated by Corbett under the heading 'Discovery of Arguments' (Corbett, 1971: 45-
298). Disposition or the arrangement of arguments (or material), is dealt with on pp. 299-413.
4. The intuited similarity between paraphrase (or intralingual translation) and translation proper
(in the sense of interlingual translation) has often been noted, perhaps most influentially in
150 Teaching Translation and Interpreting 2

Jakobson (1959). However, most translation theorists have been content to point out with Jumpelt
(1961: 10) that they were only concerned with translation from one language into another.
5. Notably Holz-Mänttäri (1984).
6. For instance Andersen & Detlef & Raahauge (1991); Galberg Jacobsen & Skyum-Nielsen
(1988); Kock & Tandrup (1989); R0nn-Poulsen & Brandt-Pedersen (1986).
7. See for instance de Beaugrande (1984); and Hillocks, Jr. (1986).
TRANSLATION ASSESSMENT: A CASE FOR A SPECTRAL MODEL

Hasnah Ibrahim, Universiti Utara Malaysia, Malaysia

Introduction
Different text types and different reasons for translating call for different strat­
egies. Though long recognised,1 this fact is not universally accepted: those who
agree often reject the normative and the prescriptive, whilst those who do not, or
who accept it grudgingly, may at best pay lip service to it; most linguistic-based
writings on translation, for example, while purporting to be neither prescriptive
nor normative end up, usually, by providing norms for producing the 'best'
translation.2
The division of opinions has pedagogical implications - and hence there is a
controversy on the teaching of translation (as discussed in Holmes 1988). Total
rejection of norms and prescriptions is untenable; it would render impossible the
work of teachers of translation, whose "major task" is "to impart norms to the
students" (Holmes 1988: 109). But stylistic 'tyranny' ought also be avoided. So
we must adopt a realistic approach to the teaching of translation. The acceptance
of a variety of various acceptable translations does not imply that there are no
'wrong translations' or 'mistranslations'. Even a cursory survey of verse-trans­
lation in the Malayo-English tradition would show a difference between the
"possible", the "potential" and the "preposterous" (Hofstadter 1980: 378),3 and to
minimise the occurrence of the last type, the pedagogic branch of Translation
Studies must, when necessary, be 'prescriptive' and 'normative', and must be
able to distinguish the "possible" from the "potential" and the two from the "pre­
posterous". For teaching purposes, a means must therefore be found to differenti­
ate not just the acceptable from the unacceptable, but also one product type from
another. The assessment (that is, differentiation) of product types, implies that
there are various strategies leading to different acceptable product types: there is
a spectrum of possibilities. Students of translation in particular must be made
fully aware of the existence of many product types, and not be exposed exclu­
sively to prevailing norms. Otherwise they will not be sufficiently versatile and
creative when new norms call for new translation strategies and translation
products.
In the context of differentiation, the notion of a spectrum is not new to trans­
lation. Rose (1981) notes that translation types differ according to their position
on the autonomy spectrum. Rose's schema is polarised, the two poles being
152 Teaching Translation and Interpreting 2

'source text autonomy' and 'target audience needs'. In this schema, a position on
the spectrum indicates both the translator's relation to her/his material and her/his
relation to her/his audience. The translator can operate within the extremes of
'reverence' to 'reference', and the translation product from 'presentation' to
'adaption'. The spectrum therefore places different product types between the two
extremes in an ordered structure, perpetuating the historical dichotomy of 'literal'
versus 'free'.
The present study extends the concept of the translation spectrum, and, by
means of a field-study approach, attempts to identify and label the components
of the spectrum which are known. By introducing a neutral nomenclature, and
hence by avoiding the historical dichotomy, it hopes to introduce a measure of
objectivity into the subjective operations of assessment, that is, evaluation and
criticism, in theory as well as in classroom practice.

The translation spectrum

The fact that a text can be translated in several defensible ways,4 means that
the translation spectrum can be compared to the iridescent spectrum obtained
from the refraction of white light. Like white light, the text can be completely
represented only by itself; each version obtained in translation, like each colour
of the spectrum, is only a partial picture of the original.5
The components of the spectrum must be labelled systematically,6 in order to
make for an unambiguous, objective discussion of the various modes of trans­
lation.7 Earlier attempts to label them have not been successful, for they do not
reflect the fact that there may be several acceptable translations of a text. Larson
(1984: 11), for example, describes a continuum ranging from the "very literal",
through the "literal", the "modified literal", the "inconsistent mixture", the "near
idiomatic", the "idiomatic" (which is supposed to be the "translator's goal"), to
the "unduly free". Newmark (1988: 45) proposes a spectrum in which the modes
are labelled: "word-for-word" translation, "literal" translation, "faithful" trans­
lation, "semantic" translation, "adaptation", "free" translation, "idiomatic" trans­
lation and "communicative" translation. Such labels perpetuate the historical di­
chotomy, and because the terms are loaded, it is difficult to use them for discus­
sions of translation products, whether theoretically, or for work in the classroom.
Newmark's use of the terms "word-for-word", "literal" and "faithful" translations
is, to cite one instance only, confusing to students, for most writers make no such
distinction; "literal", or "word-for-word" translation is usually understood to be
an attempt to be 'faithful' to the original.
 Hasnah Ibrahim, Malaysia 153

A new nomenclature
The present article attempts to set up a new nomenclature which, I suggest,
avoids the problems inherent in having conflicting definitions and uses. It will
make for an increased awareness among students (and researchers) about pro­
cesses and products in translation. It is developed in accordance with Peter
Newmark's useful reminder that "any terms translation theory ... invents should
be 'transparent', that is, self explanatory" (Newmark 1981: 36). Thus each term
representing the translation process will be prefixed "trans-". All suitable exist­
ing terms will be retained, but several new terms will have to be introduced,8
since, like 'cooking', the term 'translation' has become a blanket term.9 Graham
(1985) reminds us that to sustain and so to continue discussions on translation,
several conditions must be met, namely:
At least some agreement about the use of basic terms in question must be reached. Words ...
are to be used in ways that will permit rather than prevent further revision of our ideas about
the objects and actions they designate. No immediate or even ultimate agreement about the real
nature of such referents is required; what is needed is simply a willingness to consider various
proposals as possibly true and perhaps more plausible than others advanced in the past. Those
basic terms ought to be used indexically; almost like proper names, without bearing any mean­
ing that would determine as necessary or a priori the truth of statements expressing a particular
theory or individual belief on the matter under discussion. Conventional notions of ... trans­
lation tends to be self-defeating in that they imply their own infallibility and so deny or
somehow preclude the collective search for agreement, despite differences, that characterises
the inevitably historical pursuit of an essentially empirical subject. (Graham 1985: 23)
The terms I suggest here are derived empirically and describe pathways along
which translation proceeds. The process in which the translator tries to preserve
the tone and style of the original, for example, could be labelled trans-emulation.
This is different from trans-imitation, where the translator makes a deliberate
attempt to imitate the style of another author other than the one of the original,
for example, if Thomas Hardy were translated into the style of Malaysia's
Shahnon Ahmad, which would roughly correspond to translating Tolstoy into the
style of Thomas Hardy.
Alexander Fraser Tytler (1791) uses the term 'transfusion' for the transference
of sense from the original text. The term could be resuscitated in the form of
trans-fusion, to denote a process whereby the transference of the sense of the
original transcends all other activities.10
The translation of the classics for children, or of technical texts for lay readers,
could be labelled trans-elucidation, which would be different from trans-explic-
ation, by which the original text is explained, but not necessarily simplified, in
the translation. Both are to be distinguished from trans-adaption. This last
process could describe the translation of, for example, economics texts for sec­
ondary school use, where, for the sake of relevance, illustrative cases are changed
to suit the local economic scene.11 This term, in turn, could be distinguished from
154 Teaching Translation and Interpreting 2

trans-manipulation, where the changes made reflect an ulterior motive on the part
of the translator.
Another term to consider is trans-forming. It is the least interesting type of
change in the translation of poetry, namely, a shift from verse to prose. At a
more general level the term describes the change from one genre to another, for
instance, prose to verse or a novel into a play.
Trans-metamorphosis could be used to describe the more intricate changes that
occur when one verse form is translated into another one. Trans-creation is
where the translator exercises the poetic license of an artist, without trans-form­
ing or trans-metamorphosing. A strictly literal (or 'word-for-word') translation,
must also be included in the spectrum. In view of the mostly negative conno­
tations of the phrase "literal translation", the terms trans-mapping and trans-
imaging might perhaps be appropriate. If trans-mapping were reserved for the
automatic and uncritical word-by-word translation, trans-imaging could be used
to describe a desirable and successful trans-mapping, where there is perfect
equivalence.
Trans-position is a translation in which locality is completely transposed. Early
translations of European folktales into Malay were trans-positions: Cinderella
became Chendralela, and the fairy godmother became nenek kebayan; Reynard
the Fox became the much-hated villain Sang Lamri. In trans-position thus, the
cultural nouns are completely transposed.
In the translation of plays, it may be necessary to make changes, to make the
play come alive for the new audience. The term trans-vivification would be ap­
propriate for this process. "The translation of a translation . . . of a translation"
is a trans-derivation. A text could be said to have been trans-derived n number
of times, to warn readers of changes.
What manner of operation then is translation, since it spans so many possibil­
ities? I define translation to be a heuristic, psycho-sociolinguistic process which
transfers a text from a perceived cultural state, the From-State, to a projected
cultural state, the To-State. This definition accounts for the different modes by
taking into consideration all the variables affecting the process. The psycho-so­
ciological aspects, for example, are an indication that translators, as well as the
environment of which they are a product, are part of the factors affecting the
end-product. Recognition of the heuristic nature of the process introduces the
'why'-factor; trans-imitation is distinguished from parody only by the translator's
intention. The type of text translated also dictates the translating strategy.12 The
term cultural state indicates the presence of the 'when'- and 'where'- factors. I
prefer the terms 'From-State' and 'To-State' to 'source' and 'target' because the
earlier terms are 'value-laden' (Bassnett-McGuire 1980). Besides, the use of Eng-
 Hasnah Ibrahim, Malaysia 155

lish as an international lingua franca blurs the term 'target' in translation so that
there is no target at all.

Conclusion
The above nomenclature could, I suggest provide a means for identification of
different types of translated texts and for clarifying the nature of translation for
students. Especially with students it can be argued that it is necessary to use
neutral terms, so that they are not baffled with conflicting usage in their text-
books. At a more general level, it will have been noticed that the use of neutral
terms has allowed the discussion of translation types to move away from the
usual controversies, and in so doing, opens up for a clearer picture of the trans­
lation process. The terms themselves are dispensable, and could be replaced with
more elegant substitutes, but the underlying concepts are sound.
The spectral model forces both students and teachers to be aware of the com­
plexity of the translation process. This is important, as the failure to take this into
account has doomed most linguistic theories to simplistic forms. This has serious
implications in machine translation. A machine is only as good as the software
fed into it; a simplistic notion of translation will result in inadequate software.
By considering the more complex options revealed by a spectral model, it may
be possible to develop a more sophisticated program.
Acceptance of the possibility of the various product types is important to the
teaching of translation. Like art students, students of translation ought to be
taught different styles and strategies, so as to be able to come up with the most
appropriate translation for specific texts, at a particular time and place, for a par­
ticular purpose.
The trans-processes identified here will, however, be more helpful in macro­
scopic discussions of translated texts than for micro-level discussions. The reason
is that at the macro-level, each trans-process is a complex of several basic pro­
cesses. The micro-level strategies have been widely discussed in most writings,13
but, because different authors use different terms, or the same terms with
different connotations, it is difficult to discuss this aspect of translation. This
points towards a need for a similar standardisation of terms for use at the micro-
level as well.

Notes
1. Bodmer (in Lefevere 1977: 21) notes that there will be as many translations as there are intent­
ions; Snell-Hornby (1988) notes that there is no one way of translating.
2. Bell (1991: xvi), whilst claiming to avoid giving "commandments for the creation of 'the
perfect translation'", has listed methodological options available to the translator who wishes to
be as 'faithful' as possible to the original.
156 Teaching Translation and Interpreting 2

3. Hofstadter (1980) labels pathways which are taken routinely in going from one state to another
as possible pathways and names the pathways "which would be followed only if one is led
through them by the hand" as potential pathways, which would be followed only if special ex­
ternal circumstances arise. In the translation of Homer into English, a verse translation would be
one of several possibilities, such as the translation of Chapman, Dryden, Pope, R. Fitzgerald and
Lattimore. A fictitious example of a 'potential' pathway would be a translation of the Odyssey
into the style of prose of James Joyce's Ulysses.
4. A proverb, for example, can be translated in at least three ways, according to the reason for
translating: namely by replacement with an equivalent proverb from the target culture (for a
Nidaean dynamic effect in the translation of prose fiction), by retaining the original proverb (to
introduce the foreign culture), or by just giving the meaning.
5. Savory (1957: 54) notes that most readers of translated works seem to be aware of this,
consulting more than one translation whenever possible.
6. A study of the various terms used to describe translation types shows that even terms like
'word-for-word', 'sense-for-sense' mean different things to different people (see Ibrahim 1992:
174-177).
7. Newmark (1991: 57) illustrates what happens when the various modes are not differentiated;
commenting on Chukovsky's writing (Chukovsky 1984), Newmark notes that, "On one page he
says he wants 'precision' above everything else. The next three pages he is condemning
'precision' at every turn."
8. Considering the great (on-going) contribution Translation Studies make to the professional
(technical) vocabulary of other fields, it is ironic that the field itself is devoid of an "accessible
terminology" (Bassnett-McGuire 1980), that would allow a general discussion of the translation
process.
9. See Ibrahim (1991 and 1992).
10. The term trans-fusion is different from the term 'transfusion' as used by Shelley, which im­
plied preservation of both the sense and the style of the original, which is trans-emulation in the
present nomenclature.
11. The Malayan experience would testify to the need for this mode. In the post-independence
fervour of preparing text-books in the national language, an economics book used in Australian
secondary schools was translated into Malay, without adaptation. It was then found to be un­
suitable for use in Malayan secondary schools.
12. Snell-Hornby (1988) discusses this at length.
13. See, for example, Vinay and Darbelnet (1958, as quoted in van Slype et al. 1983), and
Malone (1986).
 TRANSLATION AND THE TWO MODELS OF INTERPRETATION

Alexis Nouss, Université de Montréal Canada

There is an old joke about a rabbi and two villagers who complain about each
other. The rabbi listens to the first one and concludes: "You're right." The man
leaves happy. He listens to the second one and concludes: "You're right" And
the second man leaves happy, also. The rabbi's wife comes into the room and
tells her husband: "I heard everything. How could you tell the first and then the
second man that they're both right? They cannot both be right." The rabbi stays
silent for a moment, then says: "You're right, too."
All teachers of translation are familiar with this dilemna. We want to encourage
our students' creativity and, at the same time, teach them the rules of acceptabil­
ity, according to the norms of the target language, culture and society. True, each
translator has his own version, and one is no better than the other, but there are
limits. Or are there? Is any translation acceptable? What are the criteria of refus­
al? Should we arrive in class with our own 'perfect' version and persuade the
students of its qualities or should we be open to any suggestion?
This is a concrete pedagogical issue but we may find answers by elevating the
question up into a more general framework. True answers may be found in her-
meneutics which I define as the study of the nature and the rules of interpret­
ation, that is the study of the meaning of signs (as opposed to semiotics which
is concerned with how signs carry meaning). Hermeneutics deals with the issue
at hand: are there limits to translations (or interpretations in the hermeneutical
perspective) of a given text? Here both hermeneutics and translation theory con­
front the main question raised by post-structuralism, deconstruction and now the
ideology of political correctness: why do some interpretations (of history, of cul­
ture) prevail over others? Here we take part in a very modern debate about the
uncertainty of all meanings. It is no surprise that the great thinkers of our
century, from Freud to Heidegger and Derrida, have considered the problem of
translation.
The theories of interpretation admit two currents, indeed two interpretations of
interpretation, which Umberto Eco describes as follows: "On one side it is as­
sumed that to interpret a text means to find out the meaning intended by its orig­
inal author or - in any case - its objective nature or essence which, as such, is
independent of our interpretation. On the other side it is assumed that texts can
be interpreted in infinite ways."( 1990: 24)
158 Teaching Translation and Interpreting 2

So the question we first asked becomes: does translation have to do with the
first or the second model of interpretation? According to traditional ideology of
translation theory, translation is on the side of the recollection of meaning, of
truth to be unveiled and restored, of an original text whose essence is to be dis­
covered according to the rules of the game. But one could argue equally well the
opposite: translation entails the need for re-translations according to ever-chang­
ing linguistic, social, cultural and historical frameworks in which the meaning of
texts can be reinterpreted.
In our quest let us remember that, not only do we have a patron saint, Saint
Jerome, but we are lucky enough to also have a god: Hermes, the Greek god of
artists, robbers - and translators. As the celestial messenger, it is obvious that
Hermes is the god of translators: are they not all carrying a text from one lan­
guage to another? Are they they not also stealing a message from its original
context to deliver it to another? Hermes is not only a mythical figure. He gave
his name to a spiritual movement in the Western world at the beginning of our
era: Hermetism. Hermetism is usually studied as an esoteric trend and a rival to
Christian spirituality in religious studies. But Umberto Eco sees it as a system of
thought in the fields of semiotics and hermeneutics. Furthermore, Eco is the
author of A Theory of Semiotics and Semantics and the Philosophy of Language
as well as The Name of the Rose and Foucault's Pendulum. Both books deal with
esoteric and occultist matters.
Eco's views on the semiotic contents of hermetism have to be placed within
the context of his research, begun in the '70s, about the role of the reader in the
creation of meaning and Peirce's idea of 'unlimited semiosis', that is the potential
limitlessness of interpretation. In that work, Eco naturally encountered the ideas
of the radically critical trend known as deconstruction (Jacques Derrida, Paul de
Man and J. Hillis Miller, among others) and its claim that there is an infinity of
possible readings of any text. Eco reacts against what he judges to be too
extreme a position by establishing norms for acceptability of interpretations; for
although any interpretation is possible, they are not all acceptable.
Eco's first argument against the infinity of interpretations is well known to
translators and grounded on ethics. The interpreter and the translator have a
moral duty not to exceed certain limits of interpretation. I introduce here the
notion of morality because texts are not only dead marks on pages but play a role
in society and culture. It is impossible to justify Jack the Ripper's acts through
specific readings of the Scriptures, Eco says ironically (1992a: 24). I will add that
numerous crimes and persecutions have been committed in Western history
through faulty translations of the Scriptures. On a less tragic note, translations of
Hegel, Freud and Heidegger have produced a range of philosophical schools of
 Alexis Nouss, Canada 159

impressive diversity. Ich and es are not identical with ego and id and are not le
Moi et le Ça. It is not only a question of words: these words, indeed concepts,
are responsible for divergent German, English and French modern conceptions
of the human psyche.
But Eco also wants to make his point by showing that the discussion is not
new and that the attitude of deconstruction can be traced back to the antique
Hermetic philosophy.
Since its Greek and Latin foundations, our world has been a world of reason,
or more precisely, we see the world through the grid of rationalism. Reason im­
plies norms, standards, limits, measures. For translators, reason means awareness
of linguistic and cultural interactions and interferences: 'visiter quelqu'un' is not
acceptable in French, and to translate 'a diner' in a Kerouac road novel to 'un
bistrot' is a faulty transgression of both the American and the French social
systems.
But the human mind is also fascinated by an opposite reality: transformations,
metamorphoses, Heraclitus vs. Aristotle, the being of things vs. the constant
becoming of things, space vs. time, and, for translators, two separate texts vs. one
message, or the plurality of languages vs. the unicity of speech (to stress the
more precise difference in French between Tangue' and 'langage'). Eco finds the
expression of this concern in the system of thought known as 'Hermetism', which
corresponds to the ethnic diversity of the Greek Empire at that time (2nd century
before Christ), a mosaic of peoples and cultures close to our contemporary world
which, then as now, calls for translation.
Hermetism believes that truth is not a delimited whole but consists of frag­
ments and that there are many ways of expressing it, however contradictory these
may be. Each book "contains a message that none of them will ever be able to
reveal alone" (Eco, 1992a: 30). This position has been illustrated in translation
theory by Walter Benjamin, with his famous metaphor of the original and its
translation "as fragments of a greater language" (1977: 78). In accordance with
this definition of truth, Hermetism believes that real knowledge does not lie on
the surface but is secret, hidden: "Thus truth becomes identified with what is not
said or what is said obscurely and must be understood beyond or beneath the sur­
face of a text." (1992a: 30) How surprising to recognize in this strategy associ­
ated with Gnostic and Hermetic beliefs some elements of the translation method
proposed by Eugen Nida, whose theory comes from his practice as a Christian
translator of the Bible. In translation the boundaries between Faith and Heresy
are not so clear.
Another consequence of the Hermetic definition of truth and meaning brings
us back to our field. The hidden nature of true knowledge makes it also an alien
160 Teaching Translation and Interpreting 2

knowledge, distant in time, like a forgotten knowledge, and in space, as if pos­

sessed by foreign cultures (Druids, Celtic priests or Eastern wise men, for ex­
ample) and thus spoken in a foreign language. Whereas the unifying ideology of
the Greek Empire called for the use of one language only, Hermetism, spreading
through territories of the Greek Empire, called for translation as the only way to
reach true knowledge. This may be the place to emphasize that Hermetism offers
a specific interest to translation studies for a historical reason. Eco is not only
concerned with a specific spiritual movement from the second century before
Christ. He studies its diffusion as a recurrent pattern of thought in the history of
Western ideas. His genealogical survey tells us that Hermetism celebrated a
second victory over rationalism during the Renaissance, when, reworked by Neo-
platonism and Christian Cabalism, "the Hermetic model went on to feed a large
portion of modern culture, ranging from magic to science." (1992a: 34). Now to
us this latter period is of tantamount importance since this was the epoch when
translation came into existence as a specific activity, due to the formation of lin­
guistic and national boundaries in Europe. If translation is a circulation of mes­
sages and a transfer of significations, it has a lot in common with the Hermetic
idea of the universe as a field of unlimited semantic connections.
But the most essential feature of Hermetism for us appears in a semiotic per­
spective: its belief that interpretation is indefinite. This springs from the Hermetic
conception of the world as "one big hall of mirrors, where any one individual ob­
ject reflects and signifies all the others" (Eco 1992a: 31) A plant would refer to
the human body, which in turn would refer to a celestial body, which in turn
would refer to an angelic entity and so on. Meaning is constantly sliding in a uni­
verse created by a god in whom all contradictions are resolved or contained, as
in the limitless possibility of readings claimed by the deconstructionists.
Eco condemns what he calls such a 'suspicious interpretation' or more firmly
dubs a 'paranoic interpretation' and which (to him) is an 'overinterpretation'. "In
defence of overinterpretation" is precisely the title of a lecture given by Jonathan
Culler, a leading figure in American literary theory, in reply to Eco. He makes
the following remark from a semiotic and pragmatical perspective: "What Eco
calls overinterpretation may in fact be a practice of asking precisely those ques­
tions which are not necessary for normal communication but which enable us to
reflect on its functioning." (1992: 113-114) Are we not familiar with that strat­
egy? We, too, teach our students never to start writing their translation on first
inspiration, as an instinctive move, but to be suspicious of their idiosyncrasies
and thus go to the deepest level of comprehension of the source text, analysing
its contents and all its aspects, both formally and semantically. All the lin­
guistics-oriented theories of translation (E. Nida, J.C. Catford, P. Newmark) call
 Alexis Nouss, Canada 161

for a 'Deconstruction' of the source text, bringing to light its basic structures.
Culler gives an epistemological ground for such an interpretative strategy to
which translation studies can easily relate.
Just as linguistics does not seek to interpret the sentences of a language but to
reconstruct the system of rules that constitutes it and enables it to function, so a
good deal of what may be mistakenly seen as overinterpretation or somewhat
better, as overstanding, is an attempt to relate a text to the general mechanisms
of narrative, of figuration, of ideology, and so on. (1992: 116)
Are there no limits to possible interpretations? The answer might help the
translator save time in his interpretative work and will allow the translation
teacher to give safer guidelines to the students. Eco agrees that human thinking
functions by admitting principles of identity and similarity. But the problem is
to distinguish between the significant and the insignificant similarities. Suspicious
interpretation, not wrong in itself, becomes too extreme when it considers all tex­
tual elements at the same level of signification. To explain this, Eco cites an ex­
ample and luckily enough the game between English and French versions of his
text proves his point exactly. The French version (in Les limites del'interpret-
ation) reads: "À la limite on peut s'amuser à affirmer qu'il existe un rapport
entre l'adverbe alors et le substantif crocodile parce que - comme minimum -
tous deux sont occurents dans la phrase que vous etes en train de lire."(1992b:66)
And he states that the faulty, or paranoic, interpretation will draw a maximal
conclusion from this minimal relationship, suspecting a secret meaning. Now the
English version of this passage (in Interpretation and Overinterpretation) (un­
wittingly?) offers material for Eco's demonstration. It reads: "One may push this
to its limits and state that there is a relationship between the adverb 'while' and
the noun 'crocodile' because - at least - they both appeared in the sentence that
I have just uttered." (1992a: 48) Should I have to translate this sentence into
French, as a suspicious or paranoic interpreter, I could have noted that both terms
rhyme (while and crocodile) and then proposed: "On peut s'amuser à affirmer
qu'il existe un rapport entre l'adverbe alors et le substantif alligator..." In so
doing, I would have acted in exact opposition to Eco's statement. Introducing
such a contradiction in Eco's words would qualify as an overinterpretation. Now
don't ask me if Eco was aware of the risk in the English version. It would be no
surprise coming from the author of Foucault's Pendulum. Overinterpretation
tends to isolate certain textual elements and link them in a brightly meaningful
relationship but without any consideration of their connection with the general
meaning of the text. To choose 'alors' and 'alligator' for 'while' and 'crocodile'
proves my poetical talent but does this add anything to our comprehension of the
source sentence? On the contrary, as we saw, it contradicts the general meaning.
162 Teaching Translation and Interpreting 2

"In theory", Eco comments, "one can always invent a system that renders other­
wise unconnected clues plausible. But in the case of texts there is at least a proof
depending on the isolation of the relevant semantic isotopy." (1992a: 62) The
notion of isotopy comes from Greimas, semantic theory and designates a series
of semantic features which constitutes a whole allowing the reader to define the
topic dealt with in a text.
To produce a right interpretation, Eco states, the only way for the reader is "to
check it upon the text as a coherent whole" (1992a: 65). Such a coherence would
prevent the reader from sliding into his own idiosyncratic interpretative drives.
The strategy for the translator is thus not to deconstruct the text into minimal
semantic segments, the method often taught in translation classes, but to figure
out the global meaning structure of the text. "My idea of textual interpretation as
the discovery of a strategy intended to produce a model reader, conceived as the
ideal counterpart of a model author (which appears only as a textual strategy),
makes the notion of an empirical author's intention radically useless." (Eco,
1992a: 66) The distinction between the empirical author and the model author is
of great pedagogical interest since it helps us answer the traditional student ploy:
"If I could, I would have asked the author!" or: "How do you know? You're not
the author!" The text stands as the only valid witness between an absent empiri­
cal author and a too present reader.
Eco draws a distinction between semantic (or semiosic) interpretation and criti­
cal (or semiotic) interpretation. He explains:
Semantic interpretation is the result of the process by which an addressee, facing a Linear Text
Manifestation, fills it up with a given meaning. Every response-oriented approach deals first
of all with this type of interpretation, which is a natural semiosic phenomenon. Critical
interpretation is, on the contrary, a metalinguistic activity - a semiotic approach - which aims
at describing and explaining for which formal reasons a given text produces a given response
(and in this sense it can also assume the form of an aesthetic analysis). ... Ordinary sentences
... only expect a semantic response. On the contrary, aesthetic texts or the sentence the cat is
on the mat uttered by a linguist as an example of possible semantic ambiguity also foresee a
critical interpreter. (1990: 55)
I suppose I am allowed to articulate the notions of a semiosic translation and
a semiotic translation in the same way. "The cat is on the mat" to use Eco's ex­
ample. A simple semiosic interpretation would translate: "Le chat est sur la
carpette" where an interpretation concerned with the semiotic dimension as well
would give: "Mistigris est sur le tapis" or: "La chatte est sur la natte".
Now the question is: could both my proposals be accepted? Following Eco's
arguments, I would say that the first one, "Mistigris est sur le tapis", could be
accepted because this translation plays on a common designation of any cat as
Mistigris and the synonimy of 'carpette' and 'tapis'. My second translation, "La
chatte est sur la natte", could be taken as an overinterpretation because it relies
 Alexis Nouss, Canada 163

on an unknown, unshared piece of information concerning the sex of the animal

and on the too specific designation of a piece of fabric lying on the floor as
'natte'. So the second translation will be rejected because, following Eco's ar­
guments, nothing in the text itself (here the sentence) allows me this interpreta­
tion. Any translation theory will agree on the principle put forward by Eco: "To
decide how a text works means to decide which one of its various aspects is or
can become relevant or pertinent for a coherent interpretation of it, and which
ones remain marginal and unable to support a coherent reading." (1992a: 146)
Furthermore, to give such an overinterpretation or to propose my previous ex­
amples of overtranslation excludes me from the community of readers. Eco
reminds us: "C.S. Peirce, who insisted on the conjectural element of interpreta­
tion, on the infinity of semiosis, and on the essential fallibilism of every inter­
pretative conclusion, tried to establish a minimal paradigm of acceptability of an
interpretation on the grounds of a consensus of the community (which is not so
dissimilar from Gadamer's idea of an interpretative tradition)." (1992a: 144) The
notion of community in translation theory is very important since for the trans­
lator the intentio auctoris does not exist in itself. The work is done for a
community.
So to come back to our initial story and as a conclusion: perhaps the first
villager was right, the second villager was right and the rabbi's wife was also
right. But in his interpretation, the rabbi himself was wrong. For a rabbi, his wife
and two villagers form a community and, given that social structure, the rabbi's
interpretation would not gain the consensus of the community. He would be a
bad translator.

Notes
1. He develops this reflection mainly in The Limits of Interpretation (1990) and Interpretation and
Overinterpretation (1992a), the latter being a collection of lectures; both books are discussed in
this article.
INTERPRETING AND CLASS
DRONNINGGAARD
 INTERPRETING STUDIES AND THE HISTORY OF THE
PROFESSION

Margareta Bowen, Georgetown University, USA

In recent years, the profession has shown a growing interest in the history of
interpreting. The professional organizations - the International Association of
Conference Interpreters and the American Translators Association - have included
articles on the history of interpreting in their publications, as have a number of
university schools for interpreters. In 1988, the AIIC General Assembly organized
an exhibition of memorabilia, and in 1992 it showed a videotape on the history
of the profession. The History of Translation project launched by the Inter­
national Federation of Translators and slated for completion in Spring 1994,
under the leadership of Jean Delisle, will include a chapter on the role of the
interpreter through the ages.
Given this growing momentum, we may well ask ourselves the following ques­
tions: What can the history of interpreting contribute to teaching and learning
how to interpret? Can such information be useful, directly or indirectly, to the
teacher of interpreting and to the future interpreter? To what extent can the
lessons of the past allow us to draw conclusions about professional ethics today?
What are the sources available to us?

Sources
I shall first address the categories of sources, written and oral, their reliability,
and the relative ease of access to them, since this will influence conclusions on
the other questions. We should distinguish between information from interpreters
themselves, comments from the users of interpreting services, and those from
third parties who, for any number of reasons, may mention interpreting services.
The last category is, understandably, open-ended: administrators, journalists,
politicians, are only some who come to mind. Historians, it should be noted, have
seldom gone into the details of specific interpreting situations and the possible
effect of the interpreters' work on the outcome of a conference.
Interpreters and delegates have given interviews and written autobiographies,
reminiscences, and diaries, which in turn were edited. Interpreters, if they are
government officials or international civil servants, need the approval of their
superiors to publish. Once retired, or as a free lance, the interpreter is free from
this formal constraint. In the interest of confidentiality, however, extreme caution
168 Teaching Translation and Interpreting 2

is nevertheless advisable. Christopher Thiéry, for many years the chief interpreter
of the French Ministry of Foreign Affairs, has gone on record stating that inter­
preters must never write their memoirs.
Interpreters' memoirs have been written, however, and we should not neglect
them. Most of those that exist in book form, as the title or the preface indicate,
were written as a contribution to the historical record and to diplomatic history,
not as case histories of interpreting as a profession. To a greater or lesser degree,
A.H. Rirse's Memoirs of an Interpreter, Charles E. Bohlen's Witness to History
1929-1969, Eugen Dollman's Dolmetscher der Diktatoren, Robert B. Ekvall's
Faithful Echo, Paul Schmidt's Statist auf diplomatischer Bühne, and Vernon
Walters' Silent Missions all concentrate more on events and personalities than on
interpreting. Many of these authors did not consider interpreting their profession.
Other objections have been raised to autobiographies in general. Salvador de
Madariaga, whose memoirs are among the sources about interpreting at the
League of Nations, wrote in the preface:
I remember having read somewhere that Freud thought all autobiographies to be but lies. If
such is the fact, I mean that Freud did say so, it may count as yet another case of that over­
statement of which the Viennese magician seems to have been fond; for autobiographies need
not be lies. What they nearly always are is inaccurate. But an inaccuracy only becomes a lie
when it is deliberate; while most of the inaccuracies in a life written by the man who has lived
it come from other causes than the intent to deceive the reader." (Madariaga 1973: ix)
The grounds for inaccuracies listed by Madariaga are the nature of human
memory and a certain lack of objectivity when writing about oneself, which may
go as far as a desire for self-enhancement. This last motive one could easily at­
tribute to Dollman and, to a lesser degree, to Bohlen and Schmidt. Dollman also
seems to seek self-justification in his book. On both counts the interpreters, how­
ever, are no different from some of their principals. Sergei Witte, chief negotiator
for Tsar Nicholas II with the Japanese at Portsmouth is outright self-congratulat­
ory in his memoirs, at least in their abridged English version, and claims that his
press relations signally contributed to the success of the Treaty of Portsmouth.
Eugen Trani (1969: 178) notes that "One must be careful in using Witte's recol­
lections." The diary of Witte's secretary, I. Korostovetz, gives a very detailed
description of the negotiations, including the language arrangements:
The use of many languages slowed the proceedings considerably, especially because of the
Japanese insistence upon absolute precision. Witte spoke mostly in French, but his use of the
language was far from skilled, and he often lapsed into Russian. Adachi [Adachi Mineichiro,
who was first secretary of legation in Paris and knew French well] translated the French into
Japanese. When Witte spoke in Russian, Nabokov translated into English. Komura [the leader
of the Japanese delegation] spoke in Japanese, which Adachi translated into French. (Esthus
1988: 97, quoting from Korostovetz. Pre-War Diplomacy: 65)
Vernon Walters' autobiographical book gives a great many comments and ex­
amples on interpreting as such, including admissions of the odd mishap here and
 Margareta Bowen, USA 169

there, and is therefore of particular interest under the heading of "teaching."

These few examples show that verification, preferably by primary sources on
interpreters, like personnel files, diplomatic dispatches, correspondence, etc., is
necessary. Normally, such material only becomes accessible after a certain lapse
of time, depending on the various countries' and organizations' rules and regu­
lations. Searches for direct sources are time-consuming and often involve travel.
Archives have had to become more protective of their materials and open
shelves, the researcher's ideal, are becoming less prevalent in many places, in­
cluding the United States.
These practical difficulties of access and verification would speak against the
incorporation of the history of interpreting in a course of studies for future con­
ference interpreters. But we should remember that some sources are always avail­
able. Depending on the location of the university in question, its programs for
one or two semesters abroad, and the networking that may be possible, students
could be directed to projects that are feasible and would acquaint them with ap­
proaches to research, which can only pay off in the future. At an international
panel discussion organized at the university of Vienna in May 1993, it was men­
tioned that after a four-year course of studies, students often have not learned
how to do systematic research. The investigation of past occurrences could pro­
vide hands-on research experience that is bound to be useful for future work
because the candidate will have learned how to use different library resources,
go through a mass of material, organize notes, etc. Also, considering that uni­
versity faculty must engage in both teaching and research, schools for interpreters
should not try to seek exceptions to the rule. This would only lower their stand­
ing in the system. For recruitment of faculty, a pool of practicing interpreters
who have had research experience during their studies is necessary.

Contributions to teaching
From oral history, teachers of interpreting have been using some elements of
professional lore in their courses. The caveat about students' engaging in histori­
cal research applies equally to teachers including the anecdotal in course work.
Care must be taken that these enjoyable excursions do not take up too much
time. Still, anecdotes have their place. They liven up class meetings and they
help students remember what was said. From my own student days, I remember
Colonel Dostert telling us about General Eisenhower getting too close to the front
line for his entourage's comfort, and Louis Paulovsky acting out the travails of
the interpreter who came unprepared to a session in consecutive. For years I
would think of the late Paulovsky every time when, before going to a meeting
in consecutive, I would make sure I had enough pencils or ball-point pens in my
170 Teaching Translation and Interpreting 2

handbag, and I stopped to buy the right kind of note-block. The fact that the
interpreter's work may include an element of danger is being forcefully brought
home to us by the elaborate security measures at most conference sites.
Some anecdotes - true or false - have been spread by journalists or interpreters
themselves and are taken seriously by outsiders. The exchange (delegate) "This
is not what I said," and (interpreter) "But this is what you should have said,"
which has been attributed in turn to André and Georges Kaminker, has been
quoted to me by any number of people who never knew either one. It will prob­
ably continue to be quoted. Both denied the story as long as they lived. Inter­
preter students should be prepared for this kind of question or comment and be
able to respond, putting things in perspective. When including this kind of
material in class, teachers must make a clear distinction between fiction, the
anecdotal, and historic fact.

Motivation
"Studying the history of translation means, in a way, to go over the history of
the world, the history of cultures, but through the study of translation." (My
translation. Van Hoof 1986: 5, as quoted by Delisle 1991: 63). Many university
programs for the training of interpreters include history courses, either as pre­
requisites, as we do in Georgetown University, or as background courses during
the earlier part of a longer course of studies (the German and Austrian univer­
sities, the Moscow school, etc.). In either case, the history courses are designed
for a general student population, often with the history major in mind. This fre­
quently leaves interpreting candidates with a feeling of irrelevancy, and they tend
to devote only scarce attention to the course. A reversal of Van Hoof's sequence,
namely studying world history through the history of the profession, would cer­
tainly be more motivating for interpreter students and provide many insights into
the consequences of interpreting performance, while providing a better knowledge
of history.
The interpreter's role in peace-making is often given as a reason for wanting
to become an interpreter, either as a volunteer or as a candidate to a professional
course. Vernon Walters describes the first moves towards peace between the
United States and Vietnam. The negotiations to end the Spanish-American War
are an instructive example of the interpreter's delicate position when serving both
delegations, reminiscent of the position of the court interpreter who is constantly
being watched to prevent any favoring of one side over the other. A detailed
study of these examples, and their list could be lengthened considerably, would
make young people more aware of the working conditions of interpreters.
 Margareta Bowen, USA 171

Ethics and working conditions

When discussing professional ethics with an interpreting class, examples are
needed. The motto "do no harm" in the abstract may be evanescent, but examples
like the one given by Ekvall (1960: 160) drive the point home. He describes the
dramatic incident during the 1954 Asia Conference, when during the closing
session a double misunderstanding was caused by a mishearing between Paul
Henri Spaak, the Belgian Prime Minister, and Chou-en-lai, Premier of the
People's Republic of China and led to bad feelings that could not be entirely
overcome. One interpreter had thought Spaak was saying "dans l'autre texte," [in
the other text] when actually he was saying "dans notre texte." [in our text]
During the exchange that followed, more confusion ensued and both statesmen
concluded that the other was trying to invalidate what had been so painstakingly
agreed upon. When the misunderstanding was finally straightened out, some del­
egates were left with the nagging doubt that something untoward might have
been attempted and all had been deprived of the sense of a job well done that a
smooth session would have provided. In a more recent article, Shlesinger (1989:
31) reports on a similar problem from a rogatory hearing in Israel. The interpreter
had misheard just one word in the question "Were there any officers higher than
[name of the accused]?" and rendered it as if the Defense Counsel had said
"Were there any officers hired by [name of the accused]?" Since two interpreters
were on duty, working in consecutive, one was monitoring the other and the error
was corrected immediately. The court either did not notice or decided to overlook
the exchange between the interpreters. Such examples show not only the dangers
of misunderstandings, but also the importance of working conditions (the number
of interpreters) and the need for teamwork.
Another example (Walters 1973:494-95) tells about a dinner speech by General
Eisenhower in France:
... he was pausing from time to time for translation. In this speech, he came to a phrase that
said, 'There are those who say that General de Gaulle is a very stubborn man.' At this point
he paused for translation. This was awkward for me because the French word for "stubborn",
entet6 or tetu, is not a kindly word and is more pejorative that the word "stubborn" in English.
Since I did not know what the next sentence would say, I was caught on the horns of a con­
siderable dilemma. So, in translating it, I said in French "There are those who say that General
de Gaulle is a very tenacious man." General de Gaulle roared with laughter, as did most of the
people in the room who understood what had happened. President Eisenhower then went on,
'But when that stubbornness is in the service of his country and the cause of human freedom,
then it is a quality to be admired rather than reproved.' A French newspaper, describing what
had occurred, said, 'The colonel diplomat drew a delicate veil across the sensitive word.' Here,
once again, I had demonstrated to myself the danger of attempting to tamper with what is being
said by one of the principals for whom one is translating.
Walters (1973: 276) also discusses briefings, for instance when in 1953 he was
172 Teaching Translation and Interpreting 2

sent to the summit meeting in Bermuda, between Eisenhower and Churchill. "I
flew back to the United States and was briefed at the White House and the State
Department on the various projects which the United States expected to present
at this meeting and broadly the US view of the purposes of this meeting." Paul
Schmidt's memoirs also show the value of interpreter briefing, an approach
which unfortunately seems to be disappearing today.

The perception of interpreters

Let us repeat the first question we raised: What can the history of interpreting
contribute to teaching and learning how to interpret? The same question can, of
course, be raised about the history of medicine, a subject taught at most Medical
Schools, and the history of law, included in many Law Schools. I asked a lawyer
and one-time translator whose answer was immediate: any profession needs a
sense of perspective. If doubts persist on the place of interpreting studies in the
university, let us remind ourselves that analytical history only came about as a
result of 18th century enlightenment (Butterfield 1955: 4).
A sense of perspective, I submit, is essential for the image of the interpreter.
What journalists select for publication on interpreting are often the marginal
aspects. It is true that the published biographies deal almost exclusively with the
highest level of the job. The authors of these autobiographies had direct access
to chiefs of state, generals, and cabinet members, but what applies to the famous
and powerful often has general validity. The place in the hierarchy of an organiz­
ation is important for a profession's image. In his article in this volume (p. ???
above) on the interpreting services at the European Commission, Christian
Heynold makes the point that "Our service reports directly to President Delors."
Edouard Roditi (1982:1) showed us another aspect of the profession, when he
stated that interpreters often were from border regions, from minorities or mixed
parentage. In a democratic society such considerations do not matter, qualifica­
tions do. Again, Walters (1973: 173), is a valuable witness on the qualifications
of interpreters. In July 1948, while travelling throughout the countries of the
Marshall Plan, he had to interpret Harriman's discussions with political and labor
leaders and with financial experts.
Sometimes it was embarrassing because I would have to translate some rather complicated
financial or currency convertibility matter which I did not fully understand and it is quite
impossible to translate adequately into another language something you do not understand your­
self. Many people believe that this type of translation is a purely mechanical process. It is quite
impossible to convey in another language someone's meaning unless you have fully grasped
and understood it in your own.
 Margareta Bowen, USA 173

Conclusion
If we remind ourselves of the historian's warning: "The dead hand of vanished
generations of historians, scribes, and chroniclers has determined beyond the pos­
sibility of appeal the pattern of the past." (Carr 1962: 13), and if we consider the
importance of a profession's image, we can only conclude that we should not
neglect the study of the history of the profession. We should not leave the col­
lection of material to those who would select them from the outsider's point of
view. In the interest of the efficient use of time in our programs, I do not advoc­
ate the institution of special courses, but would rather like to see some interdis­
ciplinary work on the part of full-time faculty, some assignments for students
built into existing courses, some theses written in our institutions.
We are about to witness and - it is to be hoped - participate in far-reaching
changes in the language professions. The present-day translators and interpreters
and their students need the sense of perspective that a knowledge of history can
give. We should not have to re-invent the wheel every time we have to explain
basic concepts of our profession.
Holmegård
 TEACHING AND LEARNING STYLES

David Bowen, Georgetown University, USA

When young people contemplate studying at a school for interpreters, they tend
to ask "Which is the best school?" I will try to show that the question cannot be
answered in the abstract. Not only do the strengths and weaknesses of schools
vary depending on the culture and the educational system in which the school is
located, but the general approach and the personality of both student and teacher
make for an optimal or suboptimal fit.

Working within the U.S. university system

The size of the institution, its rules and regulations, whether it is a private or
a public university - and the degree of control the teachers have over admissions
- are the first consideration. By an American universities' standards, my insti­
tution, Georgetown University, is small; it has a student population of about
twelve thousand, including Law and Medicine. It is a private university and
highly selective. Admission is not automatic upon presentation of a high school
diploma or a Scholastic Aptitude Test score (whose validity is once more being
questioned). Good test results are required in most cases, but they alone will not
get you into any of the top twenty universities, of which Georgetown is one. The
Division of Interpretation and Translation has an entrance examination (described
in Bowen and Bowen 1989). Since the applicant pool is international, that is
candidates offer a wide spectrum of previous experience and studies, we make
every effort to be non discriminatory as to their cultural background, and to
orient ourselves by the profession's requirements only.

Admission
Having control over who gets admitted to the course puts the responsibility on
the Division's faculty who administer the entrance examination. They must
decide whether or not a student is ready to begin interpreting studies - and at
Georgetown, for a full time student in our Division, this means interpreting
classes from the very start, not language enhancement or beginning a new
language. Only those who lack prerequisite courses from the areas of philosophy,
economics, history, government and English (or their native language) have a
"preparatory" year in other departments or other universities. Some counselling
takes place during the entrance examination, especially with borderline cases.
176 Teaching Translation and Interpreting 2

Someone recently asked me what the ideal profile of a student in interpreting

and translation is. I couldn't really answer, because there isn't one. Various
attempts to design a questionnaire for career counseling have not brought much
light to the matter either, as shown by the most recent project of the Princeton
Educational Testing Service, which is to include a set of questionnaires on trans­
lation and interpreting in its career kits. The basic tenet has always been the old
cliche that translators are introverts and withdrawn, whereas interpreters are
extroverts and expansive. This is simply not the case; I fail to see much dif­
ference between sitting in a booth all day, two floors up from the delegates'
floor, or sitting in an office in front of a computer or a translator's workstation.
Certainly, business sense and what personnel managers call "interpersonal skills"
will also play a part in a beginning interpreter's chances of success, but this kind
of prediction, regrettably, goes beyond the scope of an entrance examination. One
can only mention the business factor and include it in the course on professional
ethics offered by our Division under the heading Interpretation and Translation
as a Profession.
Major factors for admission are age and language combinations. The law of the
United States, as increasingly elsewhere, forbids discrimination for the handicap
of age. For some European institutions this has meant doing away with the upper
age limit. We never had one and did not fare badly with older students, and if
there are doubts about an older candidate, advising has shown itself to be more
effective than a strict "No, we don't want you." The "No, we don't want you"
approach, regrettably, is sometimes necessary with the younger set. Usually it is
accompanied with "At any rate, not now," which young people tend to see as the
same thing.
In the present volume, Christian Heynold deals with the truly daunting number
of languages that have become official conference languages as a result of the
successive enlargements of the European Union. There is, clearly, also the
problem for the interpreters of maintaining the required range of three, four or
more passive languages. A serious investigation would be desirable on questions
such as: How well can people know these languages, and at what age? What is
the lead time for acquiring an extra language, of the same linguistic family as
those already known as opposed to one of a totally different family? Some
staunch souls have, to date, started modem Greek under the Commission's policy
of educational leave; this may turn out a picnic compared to Finnish.
Our location in a city with the largest number of foreign embassies sometimes
entails a different problem: Which, if any, is the candidate's 'A' language? Is it
a real 'A' or an honoris causa 'A'? Did the candidate's family move from one
foreign service post to another, from one school system to another? The school
 David Bowen, USA 177

systems are different and consequently also affect the time needed for interpreter
training. Here we also have the first one of several personality aspects: some
people work best under pressure, others need time to work at their own pace.
This does not necessarily imply that the second group is going to be too slow for
simultaneous interpreting, but it may be a warning sign.

Student expectations
Student expectations may be completely wrong. They may think it is easy. "All
you do is sit there and talk". Some people are misadvised by well-meaning lan­
guage teachers. Fortunately, guidance counselors are less inclined today to tell
female students they should opt for languages, even if truly talented for the
sciences. Our environment, to a certain extent, also propagates the wrong ideas
by separating foreign language departments, where there are any, from the Eng­
lish department (which every U.S. college and university has). As a result, Eng­
lish does not look like a language you study or perfect. You major in foreign lan­
guages. The mention "insufficient knowledge of the native language" which was
listed many years ago by M. Bachrach as one of the three reasons why few can­
didates pass the European Parliament's recruitment test for translators, usually
comes as a surprise to students. Yet, in U.S. colleges, English proficiency ranges
from "honor student" in English to "remedial English."

Motivation, 'Operation Bootstrap' and 'Unrequited Love'

Regrettably, it is often the highly motivated students, those who were serious
language majors and devoted great efforts to acquiring what language teachers
call near-native fluency, who have to be told that they can not work into that lan­
guage as conference interpreters. Not as simultaneous interpreters, anyway. One
may have to explain patiently that all the sweet elderly gentlemen who had
assured the candidate that her French was perfect, if attending a conference as
delegates, would be the first to complain to the organizer that what was coming
out of the French booth was pure nonsense. "But I love French!" is the usual re­
action. The non-prescriptive approach to language teaching may be responsible
for this kind of unrequited love, as well as some modern language teachers'
fascination with near-native pronunciation. American candidates occasionally
send us cassette recordings with a sample of their "perfect" accent, which can be
quite deceptive if not accompanied by an in-depth knowledge of the unspoken
but implied.
In most cases, motivation alone will not turn such situations around. What
makes matters worse, is the lack of appropriate courses to improve a student's
foreign language beyond a certain point and the uncertainty of the outcome. The
178 Teaching Translation and Interpreting 2

easy way out taken by most teachers is to advise a prolonged stay in a country
where the language is spoken. Sometimes this works. But we all know people
who have lived in our country for ten or twenty years and still have not acquired
the elusive "near-native" fluency which still would, at best, get them a B rating.
We do know some success stories, fortunately, but we do not know enough about
them to explain them. Often, what worked for one person, will not work for the
next.

Learning styles
Beyond language acquisition, we have seen that, generally, there does not seem
to be a Royal Road applicable to all. Therefore we looked at our successful can­
didates, those who worked well in class and did well once released upon society,
and tried to compare their approaches to what experimental psychology has to
say about learning styles. We soon came to the conclusion that learning styles,
which are described differently by different authors, must meet matching teaching
styles. The authors we found most helpful are Frederic Vester, Pierre Casse and
Eric Berne. Both Vester and Casse distinguish four styles. Berne actually des­
cribes the patterns of interaction between people and the secret agenda they may
have.
The sociologist Pierre Casse (whom we met at a seminar of the Society for
Intercultural Education, Training and Research) gave us permission to use his
self-assessment test on large groups of candidates to our entrance examination
and made a video-recording of a lecture on his method for us. His four groups
are: 'idea oriented', 'process oriented', 'people oriented' and 'action oriented.'
These categories, which seldom are found as pure types, come close to Vester's
groups of schoolchildren (Vester 1978: 40 and 205, illustrations) who may learn
best by experience, or by a friendly talk with a classmate, or by detailed demon­
stration, or from listening to an abstract presentation. Obviously, a predominantly
process oriented person will become frustrated when the teacher concentrates on
criticizing results. Vester's questionnaires can be adapted for asking students to
make suggestions in an interpreting class.

Teaching styles
In the present volume there are some outstanding examples of teaching styles
to suit the process oriented student (Daniel Gile). Students appreciate a teacher
who is genuinely interested in them, who changes the approach in accordance
with the material studied and their needs. This dynamic approach to teaching
means that one occasionally has to make concessions to a student's learning
style, since we are bound to admit people who are different from one teacher or
 David Bowen, USA 179

the other.
The role of the teacher is not to terrorize the undeserving, but to build con­
fidence, which must be justified. A false sense of security is dangerous. When
teaching translation and interpreting, teachers are all too often tempted to discuss
only the mistakes. Of course, they must be taken up. Danica Seleskovitch recom­
mends that in the beginning only the most serious ones be taken up, and as the
students progress, more and more detail is to be insisted upon (Seleskovitch and
Lederer 1989). We feel that at the same time students should be told when they
have found a very good solution by themselves or shown specialized knowledge
that can be shared with the rest of the group. Evaluation grids can help by show­
ing different kinds of mistakes and awarding stars for unexpectedly felicitous
solutions (Margareta Bowen 1989: 58, 59). Some mistakes must be corrected im­
mediately, while the student is still working on a speech and is bound to repeat
it. A post mortem of the performance is not sufficient to avoid the ingraining of
certain mistakes. This need not be a long interruption; a disbelieving "WHAT?"
from the teacher or a classmate will be enough to let the culprit know that a
'false friend', for instance, or a stylistically discordant colloquialism will not go
unrebuked. One may add, for reinforcement, a pun or a brief anecdote, to help
the students remember the correct formulation (see also Bowen p. 169, above).
If, at a later time, the same mistake is repeated by the same person, one will
know that the comment did not work as a memory prompt and should not be re­
used, and instead one may resort to a play-back of the recorded performance, pre­
ferably on the student's own time.
In addition to building confidence, the teacher must monitor listenability. Any
particularly annoying speech mannerisms or patterns, for instance the "shrill sub­
urban housewife" syndrome, or the "voice like a buzz-saw" must be corrected be­
fore training begins. A rising note of hysteria while under pressure and a "Chris­
tian Maiden Being Fed to the Lions in the Coliseum of Imperial Rome at its
Most Decadent" delivery can be remedied by confidence building. Even the most
dedicated teacher cannot teach constructive adrenalin flow.

Students in action
The recognition of the four basic types means that one can readily react to
students. There may be the teacher-dependent student, who needs to be encour­
aged to be more assertive and make decisions independently. More common is
the show-off, with a tremendous repertoire of trivia, which may or may not be
of interest. There is also the discussant, who if left alone would take half a
semester arguing over two paragraphs. There is the 'team-teacher' who helps you
teach your course; she may get on your and other students' nerves and, worst of
180 Teaching Translation and Interpreting 2

misinstruct gullible classmates. There are also those who must have a theory,
system or rule. They are the ones who use card files, complain that the course
is not well "organized", take to packets of texts, etc. An early warning of this
student reaction is important, if the teacher is not personally given to card files
or their modern equivalents. Schools that use the student-teacher evaluation
system provide such feedback, but one may wish to get a reaction before the
mainframe computer has spit out the results a few weeks after the end of the se­
mester and ask all students to write an item for the suggestion box. This may
well avoid the unpleasant surprise colleagues from one school had at a meeting,
when a recent graduate had been invited as a panelist and this pride and joy of
her school, like Oliver Twist in the orphan's home, politely said "Un peu plus [de
pédagogie]!" ("A bit more pedagogy")1
One remedy to this longing for systematicity is special assignments to those
students. Depending on the seriousness of the case, these could be more pre­
scriptive material like resolutions, or appropriate readings on theory, the conclus­
ions to be presented to the other students as an exercise for Public Speaking
class, preferably recorded on videotape. The ensuing discussion can cover both
content, presentation, and interpreting performance by classmates to give all their
due.
The other extreme is the intuitive student who, if good, is excellent; if bad, is
helpless and floundering around, never knowing what he did and why. In be­
tween, there is a continuum, on which most people fit. For those who want a
system, it is usually in the attempt to beat it and to find shortcuts, for instance
by looking for rules instead of reading background material, or trying to limit
text preparation by studying word frequency lists. Unfortunately for this type of
student, most speakers put their time to better use, and do not favor simple
declarative sentences, similar to those of an earnest fifth grader. Those with ex­
tremely rule-bound 'A' languages, who are always asking "but, what is the rule?"
need to learn how to work in a language operating on unwritten rules.
Once you have these people together, they begin to interact. If you are out of
luck, they will fight. The show-off student, for example, is often resented by the
others, and care must be taken not to let the most withdrawn one sit in a corner
thinking abstract thoughts and going off on tangents. To have a workable class,
one must watch over time management and still let students make the kind of
creative contribution to the class that suits their personality.

Conclusion
Indeed, selecting students for interpreting studies is rather like choosing a
doctor: you want one mature enough to have made any major mistakes elsewhere,
 David Bowen, USA 181

but not old enough to retire or die on you. Once you have selected them, you
must live with them.

Note
1. AIIC Schools symposium, Strasbourg, 1986.
Nylars Kirke
 EXPERIMENTS IN THE APPLICATION OF DISCOURSE STUDIES
TO INTERPRETER TRAINING

Robin Setton, GITIS, Fu Jen University, Taipei, Republic of China

Introduction
A review of the steadily growing literature shows the continued existence of
several competing approaches to interpreter training, a specialisation within an
already highly specialised field. A keen rivalry has developed between the ex­
ponents of a traditional, practice-based, intuitive approach, and researchers origin­
ating in 'harder' disciplines, who have proposed numerous teaching methods
based on findings from cognitive psychology, communication science, and statist­
ical studies.
The controversy over the most appropriate paradigm for research into confer­
ence interpreting is thus sadly reminiscent of an old-fashioned confrontation
between the 'arts' and the 'sciences'. Interpreting being a social function, no
test-tube type investigation can be expected to give a satisfactory account of it;
any forced attempt to produce results in interpreting research, with its small
samples and multiple factors, complying with 'scientific' imperatives such as re­
producibility and predictability must run the risk of seriously distorting or
over-exploiting the data. On the other hand, we lay ourselves open to the charge
of technophobia, mystification or obscurantism if we wilfully ignore the latest
methods and findings in those areas where parameters are quantifiable (scientific
observation), as well as the latest extended conceptual apparatus in the human­
ities - notably, the burgeoning linguistics of parole. Recent writing suggests that
developments in the study of natural language at new levels (pragmatics, dis­
course analysis, speech act theory) may offer a credible framework in which to
formalise the 'naive-empiricist' training strategies pursued intuitively by trainers
in the past.
In this article I shall describe some strategies used in the training of
Chinese-English conference interpreters at the Graduate Institute of Translation
and Interpretation (GITIS), Taiwan, focusing on devices to enhance the familiar­
ity of trainees with discourse-level characteristics of public speech in the two
'working cultures'. This cross-cultural training experience has particularly high­
lighted the need for enhancement of pragmatic and rhetorical competence in both
comprehension and production. In the discussion, a simple multi-level discourse
184 Teaching Translation and Interpreting 2

model is proposed, both as a guide to identifying problems in performance, and

as a component in a comprehensive description of interpreting, complementing
other facets such as the very thorough catalogue of situational parameters (mode
of presentation, speed, visual contact, preparation) explored by Franz Pochhacker,
and the wide-ranging analysis of interpreter strategies begun by Sylvia Kalina
(both in this volume). The low assessment of P5chhacker's recorded extract
(below, p. 239), for instance, can be attributed to rhetorical failure: heartfelt
thanks and welcome delivered in a hesitant, doubtful tone.
The exercises and training strategy described here were designed to address
particular difficulties, some of which are local or culture-related, others task-
related; these are described in more detail in a previous article (Setton 1993a).
The influence of different historical conceptions of rhetoric and the role of the
speech act in East and West, and other cultural contrasts, are also described
elsewhere (Setton 1991 and 1993b).

Theoretical position
However, in view of some current controversies, I begin with a short statement
of my pre-theoretical position on interpreting:
1. That insofar as the product (sometimes called the 'target text') of successful
translation or interpreting is experienced as natural spontaneous language, the
same criteria should apply in evaluating its effectiveness.
2. That spontaneous (not translated or interpreted) text is generated from a deep
structure (to borrow modern terminology), and not from a prior and different sur­
face structure; we assume that thought and the desire to communicate are pre-lex-
ical, and hence that the success of any text as communication is proportional to
(a) the extent that it is generated from deep structure, that is to the extent that it
is assimilated at a pre-lexicalised level; and (b) its interactionality.
The question of equivalence presupposes agreement about some reliable definit­
ion of units of discourse and 'text'; and a reliable measurement of co-text and
context domains. Since the latter demonstrably expand in the course of the text,
and are constantly modified by the act of utterance itself, this direction of inquiry
currently seems too open-ended (rather as generative semantics appeared by the
1970's).
However, translating or interpreting into a non-native language will require
more frequent fallback on 'pragmatic' equivalence-based solutions.
3. That since there are at least two bipolar parameters (source text/target text,
speaker/hearer), translation or interpreting are most usefully investigated by con­
sidering at least two components: comprehension and production. The assumption
is that however fast the process, it is merely a blurring, or at most only partial
 Robin Setton, Republic of China 185

'pragmatic' short-circuiting, of two separate and overlapping activities: compre­

hension and formulation.
4. That it is more productive to focus on these activities in themselves indepen­
dently before speculating on the whole process, thus dividing research and train­
ing into three distinct fields:
- Discourse comprehension at various levels: segmental/semantic/pragmatic, in­
cluding any and all factors involved in drawing inferences and making interpret­
ations.
- Discourse production: degrees of expressive proficiency: language perform­
ance and enhancement, memory and activation, lexical retrieval; higher expres­
sive functions such as discourse ordering and presentation and rhetorical effects
such as emphasis, foregrounding, abstraction, compression, ellipsis and cohesion.
- Discourse handling: technical skills and aids to the interpreting task (such as
consecutive notes); and physical conditions favourable or unfavourable to one or
both of the activities.

Chinese-English interpreter training

Local constraints
At this stage it is necessary to explain some of the special constraints on the
Chinese-English and Chinese-Japanese training of interpreters in Taiwan - and
probably on other non-European programmes - as compared to established pro­
grams in Western Europe.

Environment and culture-related problems

Training interpreters to work between two languages with very different social
and cultural backdrops, such as Chinese and English, poses problems at levels not
encountered in European programmes. The most obvious are:
(1) trainee recruitment, related to the status of the profession: it is difficult to
find sufficient numbers of candidates highly qualified in both cultures and to at­
tract them to what is still, outside Europe, a small, marginal and poorly paid pro­
fession (Setton 1993a).
(2) the culture gap (or the 'ethnosemantic' dimension): the traditional 'second
language deficit' here extends into wide areas of culture and thought and the dif­
ferent conventions of communication, debate and the presentation of ideas.

Task-related challenges
The market requires interpreters to be fully 'bi-active', i.e. able to work accu­
rately and produce acceptable grammar, vocabulary, register, etc. in two lan­
guages, both in consecutive and simultaneous. Also, since most multilingual con-
186 Teaching Translation and Interpreting 2

ferences require Chinese, English and Japanese, relay interpreting (in which inter­
preters have to rely on a colleague's version from source languages they do not
themselves understand) is the rule, not the exception, placing high demands on
production of the acquired language, particularly in terms of clarity, compression
and cohesion.
These constraints have prompted research at our institute into new strategies to
improve students' familiarity with the pragmatic and cultural dimensions of dis­
course. Fortunately, small class numbers (3 to 5 students per class) have made
it possible to implement, refine or apply new experimental methods fairly rapidly.

Curriculum design
While some widely-accepted training principles, such as the progressivity of the
interpreter training curriculum from consecutive to simultaneous interpreting,
have been conserved (and extended, with 'B' to 'A' before 'A' to 'B', and in­
tensive simultaneous-with-text in the final semester), the special constraints des­
cribed above have prompted the following adjustments at various stages:
1. Entrance examination: one-language, context-related tests.
2. Introductory courses (2-3 months) in each of the working languages, con­
sisting of various exercises to enhance discourse familiarization,
3. 'Reportage' sight translation.
4. Extensive training in simultaneous-with-text.
5. Relay training.
This article discusses my experiments using three exercises:
In comprehension: anticipation (introductory and again before beginning simul­
taneous).
In production: speech construction (introductory and with consecutive note-
taking).
In 'handling': 'smart shadowing' or same-language 'chunk' paraphrasing (trans­
ition from consecutive to simultaneous, as explained below).
I now take a closer look at the context in which the exercises were introduced.

Entrance selection: combined cognitive and language testing

Since no-one may be admitted to postgraduate studies in Taiwan solely on the
basis of an oral examination, the written part of the entrance exam must serve as
a final test for translation students while also testing prospective interpreters, for
whom it counts for 50% of the grade for admission. The written papers in the
'A' and 'B' languages must thus be designed to test active and passive language
proficiency, but also whole-text comprehension, interpretive skills and general
world knowledge. At the same time, in view of the scarcity of good candidates,
 Robin Setton, Republic of China 187

it must allow potential interpreters such as overseas Chinese, non-Chinese etc,

whose writing skills often cannot compete with locally-educated applicants, to
have a chance at the orals.
This has led us to include tests such as paraphrase, summary, and the insertion
of synonyms in context, as well as an evolved version of the Cloze test, which
can be considered in its basic approach as a forerunner to some of the exercises
in subsequent interpreter training.

'Discourse-level' Cloze
The standard Cloze test used in standard language proficiency testing consists
of text with words omitted, usually randomly or at regular intervals, with a
choice of only one 'correct' answer. The test is thus easy to grade (by an optical
reading device), since it is context-independent; but as Sergio Viaggio has
pointed out (1992b), this is precisely what disqualifies it as a test for trainee
translators or interpreters, since it implies a one-to-one correspondence between
sign and signification which not only flies in the face of linguistic science, but
also enshrines the word as the only possible sign-unit, thus confining the test to
one of 'isolated' morpho-syntactic proficiency.
The 'discourse-level' Cloze test given in our entrance examination is similar
in principle to Viaggio's 'cognitive clozing', except for the progressive element
(it is given as a one-time written test) and the length and choice of the omissions:
candidates may be asked to supply a whole paragraph.
Completed versions are assessed to different standards ('A' or 'B' language)
for grammatical and lexical, but also logical (semantic) acceptability, and for
appropriateness on the basis of a minimum assumption of knowledge about the
world. The use of this to test for admission shifts the emphasis to text compre­
hension, which is tested at several levels: the ability to construe meaning, to ap­
prehend logical development and to recognise and adapt to the illocutionary force
in the passage. Cognitive preparation (general knowledge of world affairs, etc.)
can also be tested by an appropriate choice of text. Grammatical and lexical pro­
ficiency - 'language' proficiency in the restricted sense - are of course naturally
tested both on the comprehension and production (performance) side.

The introductory courses

Goals
The three-month introductory module in our curriculum is designed to fulfil
several functions:
a. to enhance language comprehension and proficiency,
b. to train memory, particularly each student's familiarity with his/her own
188 Teaching Translation and Interpreting 2

memory.
c. to create a sensitivity to different types of discourse and register, and to the
relative significance of discourse parts in different contexts, while introducing
the subject matters of international meetings.
d. to draw attention to the pitfalls of interference.
Contrary to expectation, the morphological and etymological distance between
Chinese and English does not remove the problem of interference. Chinese has
been heavily influenced by English in the last hundred years, both in terms of
partial imitation of syntactic structures (Kubler 1985) and 'direct' translations of
concepts which appear straightforward, but in fact have undergone the same
subtle semantic and functional skewing as borrowings between all language pairs:
the Chinese word tupo(xing), for instance, conjures up the English word 'break-
through', but it more often means 'groundbreaking', and can be used as adjective
or verb, rarely as noun, while the English word is hard to use other than as a
noun.

Introductory exercise
Exercises in this phase include:
7. Oral (consecutive) paraphrasing and memorisation.
These exercises begin along the lines of those described by Weber (1991:
404-9) as preparation for and introduction to consecutive, such as public
speaking (reading aloud of speeches followed by same-language paraphrase)
and memory exercises from simple narrative or descriptive extemporaneous
speeches, but are then continued in the same language for a second and third
month, with more difficult input from conference speeches, TV interviews,
presentations on economic and political topics.
2. Language enhancement and 'agility' exercises such as at-sight paraphrase
and gist extraction (Mackintosh 1991: 391). Register switching has proved dif­
ficult in the early stages, and is not recommended in the 'B' language before
the second year, when students begin simultaneous into 'B'. Other exercises
quoted by Mackintosh are recommended to students for enhancement of active
proficiency in their own time: shadowing, reading out loud, active reading.
Mock conferences are staged from the second semester, notably for training in
relay interpreting.
3. Speech analysis and speech construction.
The majority of candidate trainee interpreters in Taiwan know little or nothing
of public life, still less the world of international conferences; initiation to the
vocabulary, procedures and structure of international organisations, treaties and
conventions, budgets, voting etc., is begun in the introductory semester, then
 Robin Setton, Republic of China 189

consolidated in the third semester.

Against this cognitive background, attention is drawn to differences in the
patterns of discourse organisation and presentation in different language/cultures;
comparative studies of speeches, and anticipation exercises (discussed below),
raise awareness and familiarity on the receptive side. On the production side,
confidence is increased by the speech construction exercise. Students also shadow
real videotaped speeches on their own.
In a regular analysis of speechmaking, large numbers of speeches taken from
actual conferences and public occasions (as far as possible in an international
context) are analysed, and attention drawn to recurring patterns (a heuristic ap­
proximation is given in Figure 1 overleaf: p. 190). Articles and extracts from
popular guides to speechmaking (such as Dale Carnegie's) are occasionally
handed out as background reading.
Speech construction has been experimented with, both as active language pro­
ficiency and public speaking exercise and as a support to teaching consecutive
interpreting (below),
In this exercise, the instructor constructed idealized consecutive notes from a
speech. Conventional note-taking principles of verticalisation, indentation super-
position and division were known to the student, as well as the use of arrows,
horizontal lines and a number of commonly used symbols. After having other ab­
breviations and symbols explained, the student delivered a speech from the notes
which are shown in Figure 2 (overleaf: p. 191).
The idealized notes are of course here merely a construct to illustrate this func­
tion/toolkit paradigm of consecutive with notes. The student's freedom of inter­
pretation and representation (step 1) is not interfered with, since the exercise
begins directly at the rendition stage.

Consecutive interpreting
The numerous analyses of the consecutive interpreting task in the literature are
here interpreted (again borrowing from recent linguistics terminology in an at­
tempt to relate the study of consecutive interpreting to mainstream discourse
studies) in a simple two-step 'generative-functional' model:
1. Audition: Listening, comprehension, interpretation, representation in notes
and memory.
Notwithstanding variations in the use of source language vs. target language ab­
breviations, symbols (individual or common) etc., by different interpreters, con­
secutive notes with their symbols, relations, layout and hierarchy can be des­
cribed as an interlexical phase; they can be conceived of as representing one
layer of a grid of functions, references, ties and rhetorical effects which is com-
190 Teaching Translation and Interpreting 2

Figure 1: Speech analysis

" 1Guten Morgen, meine Damen und Herren. I am 2honoured to be with you today, and I am
3
grateful to Minister Stoiber for the invitation. It is also a 4pleasure for me to share the program
with my 5colleague and good friend, BKA President Hans Zachert.
I am 6privileged to address this 6illustrious group about 7the United States experience in fighting
the plague of organised crime. I accepted this invitation 8in the hope that our experiences might
help you as you engage in your 8current crime-control debates. 9xBut I would like you to know
that I have 9Nnot come to speak in support of any political platform in the debate over organized
crime legislation. I have 9Nnot come to tell you how to solve your crime problems - 9Ybut rather I
have come to share with you some facets of the FBI's experience in our s35-year war against
crime.
Today, 10abecause organized crime and illegal drugs have become a global problem, they are of
increasing concern to 10aall nations. These problems are also of great personal concern to 10bme.
11
I have lost two close friends and professional colleagues to violence perpetrated by organized
crime [12names, dates].
A
like an infectious disease,....to fight this disease, 13all governments need to...
14/1
Today, I would like to discuss with you how the FBI's techniques have evolved over the
years. 1 4 / 2thenI'dliketosharewith you some of the techniques.... 1 4 / 3 Finally I'd like to share
with you...work together....First,
FUNCTION REFERENCE
d.e. institution/person events place time topic
me you other
greeting 1
honour h pleasure p 2h 4p-N5
compliment c
thanks 3N 6 7
introduction
setting scene 8timely - motive:
x 9 N-Y
description 11 12/V
reason/argument l0b 10a
moral
recommendation 13-13-13 14/1-2-3

Abbreviations:
RESOURCE: quotation, statistic S, news item NI, proverb, name N, analogy A
DEVICE: negation N/assertion Y, suspense (*), contrast (x), concession (/)
D.E. (displacement effect): humour, metalanguage
STRINGS: topic definition, analogic and discursive strings

*from: "Combatting Organized Crime", delivered by William Sessions, Director of the FBI, at a
Symposium in Germany on Dec. 4th, 1992 and reprinted in Vital Speeches of the Day, Vol. LIX,
no. 9, Feb. 15th, 1993, City News publishing Co., Mount Pleasant.
 Robin Setton, Republic of China 191

Figure 2: Speech construction

Context, situation (an international conference in Pyongyang, Korea), proper

nouns and titles were given to the student in advance.

(1) Constructed notes (originally hand-drawn):

Guide to abbreviations:
Hon: honour; pleas.: pleasure
DPRK: Democratic People's Republic of North Korea
: beginning of a cliche meaning 'crossing seas and mountains'
: "court and populace" (also "government and opposition")
indicates cordiality, underlining indicates emphasis, tick indicates approval.
(2) Reconstruction (by a second-semester student with Chinese A, English B, after
approximately 6-7 weeks of consecutive training. Pauses of one second, or of two or more
seconds, are marked [ ] and [pause] respectively.

Distinguished Guests, Ladies and Gentlemen -

It is my great pleasure and honour to be here to extend my welcome to all

delegates [pause]. We have come [pause] across the world to participate in this
conference [pause] and on behalf of all the delegates here I'd like to express my
wholehearted thanks to the peoples and the government of the [ ] Democratic
People of [pause] Republic of Korea for their warm welcome [ ] and for the
facilities provided to us. Especially I'd like thank the President of [ ] the
Chairman of [ ] the Republic [ ] the democratic Peoples of the Republic of Korea
[ ] for his interest of being here [ ] and, er [pause] the active participation of both
the people and the government here showed that [ ] the [ ] Korean people's
hospitality which we are very grateful for...
192 Teaching Translation and Interpreting 2

plemented by a second component in the interpreter's memory.

This terminology conveniently brackets the issue of whether discourse can exist
in a 'deverbalised' state, leaving it to philosophy or a future breakthrough in cog­
nitive science.
2. Rendition: Lexicalisation of the function-reference grid as a new 'surface
structure' from the available language 'toolkit'.
Needless to say, the mapping processes are multiple and diverse: any form of
utterance material (continuous or discontinuous speech acts, combinations of ver­
bal/nonverbal + cognitive complements) can map potentially into one grid-item
at audition/noting, then, at rendition, into many potential realisations (non-verbal
included) with the pragmatic adjustments, including the addition or subtraction
of utterance material, required to form a cohesive 'neotext'. Since these prag­
matic adjustments will involve the new shared presuppositional context and the
various 'Gricean' cooperative principles (relevance, informativeness, clarity, etc.)
they will necessarily be situation- and culture-dependent.

The toolkit (or lexicon)

On the issue of 'lexicalisation', Gile's discussion of retrieval and activation (the
'gravity model' (1988)) is instructive. Gerard Ilg's "penny in the pocket" tips on
useful phrases, though suspected by some as 'stock-equivalent' based, come into
their own for interpreters who have to work into a 'B' language.
Frequent instructor's demonstrations, and attendance at conferences, have
proved to be extremely effective at this stage.

Transition from consecutive to simultaneous

"Smart" shadowing, or real-time paraphrase
A review of the literature shows that there seems still to be some vagueness
about the first steps in simultaneous interpreting. Among exercises proposed by
cognitive psychologists, some, of the multitasking type (for example counting
backwards while listening), have been accepted even by diehards among the 'em­
piricists'. Shadowing, on the other hand, is more than controversial. Some Ja­
panese training schools fill the gap with 'booth sentence-by-sentence consecutive'
in which trainees translate the previous sentence into the target language during
pauses.
Our experience has been that in spite of everything, students at this stage still
need every possible incitement to claim sovereignty over their own formulation
and struggle free from the morphological straitjacket of the incoming language.
Admittedly, one of the motivations for trying out these exercises was variety in
classwork, since the traditional 'multitasking' exercises soon become mono-
 Robin Setton, Republic of China 193

tonous.
"Smart" shadowing has been very well received by our students and appears so
far to be the most effective way of guiding students into coherent simultaneous
interpreting.
At first, 'processing units' are suggested by the instructor by pauses at possible
sense-unit boundaries. Students can be encouraged to make complete syntactic
units (even sentences) at each pause. Over a period of six to eight class hours,
the various dimensions of the problem are introduced in steps:
- reader or speaker pauses after incomplete sense units: students' attention is
drawn to various strategies: holding pattern; filler material (depending on length
of pause or delay before next 'clue' is heard); 'open' grammar.
- pauses are shortened: trainees learn to keep listening while talking, finish
their sentences etc.
- the stop-go flow glides into normal speech.
Time off is taken for suggestions as to opening structures, instructor demonstra­
tions, and comments on lagging and leading.
Any of these steps can be done either from 'B' into 'A' or in the same lan­
guage, providing practice in paraphrasing and verbal agility.
One spin-off of the technique has been an insight into 'segmentation'. Several
authors have sought to measure lag, describing input discourse in terms of
numbers of successive words or linear segments (Carey 1971; Moser 1978;
Lambert 1988; Davidson 1992) but it appears that when real simultaneous begins
to flow, it soon passes from the overlapping processing of a succession of seg­
ments (if it ever was that) to a series of sharpening approximations, in which
semantic fragments (not words or clauses) are seized on to produce provisional
non-committal output. Such fragments are suggested in bold in the following
excerpt from same-language paraphrasing, where the pace of the original is slow
and the interpreter cannot wait for too long:
Source text: We have also suffered the consequences of the uncontrolled international
Interpreter: ...another problem we had is ... all
Source text: exports of hazardous nuclear waste
Interpreter: over the world... ...dangerous nuclear waste is being sent to Thailand and...

Anticipation: a probe into the 'lead' dimension

In the anticipation exercises, speeches are read out, or tapes of speakers pro­
jected, after the trainees have been filled in on as much as they would be ex­
pected to know about the event, situation, players, place and so on. The tape or
speaker stops occasionally and students try to continue the sentence. The exercise
is reported here simply on the basis of student response, suggesting its value as
a consciousness-raising exercise.
194 Teaching Translation and Interpreting 2

The role of anticipation in simultaneous interpreting is widely recognized in the

literature, although - perhaps due to the greater difficulty of identifying and
measuring it - it has not attracted the same interest as 'décalage' (lag).
Lederer (1978) posits two distinct types - anticipations based on sense ex­
pectation and those based on language prediction, as when the components of
stock collocations like 'play...a role' or 'shoulder...responsibility' are separated
by long stretches of intervening text. Wilss (1978), drawing largely on work by
Mattern in German-English simultaneous interpreting, calls these moments anti-
cipation cues, or A cues, classifying them as follows:
- 'co-textual' (intralingual);
- extralinguistic (situational); and
- context-independent cues, such as those based on a knowledge of standardised
communication processes ("on behalf of my delegation I would like to...thank"),
as well as cliches or "petrified, idiomatic phrases (collocations)".
Not surprisingly, interest in anticipation has been highest among those working
with syntactically asymmetric language pairs requiring rearrangement at sentence
level: typically, in simultaneous interpreting from German. Chinese and Japanese,
being 'structurally problematic' source languages of this type, have seen similar
studies (Davidson); strategies proposed by Zhuang (1991), for example, mainly
address Wilss' third type of 'A-cue'.
Wilss claims that "the development of a subtle syntactic and semantic antici­
patory ability is a useful goal in the linguistic and psychological teaching of
[simultaneous interpreting]" (1978: 350). However, differences between Japanese
or Chinese and English or French are not confined to the sentence level, suggest­
ing the potential value of cultivating a 'wider angle' of anticipatory ability.
Again, discourse analysis might contribute to building a knowledge base for such
work.

Discussion
Personal experience of Chinese-English interpreter training, and a perusal of the
literature on translation and interpreting between Japanese and Indo-European
languages, indicate that problems 'above' the sentence level arise more frequently
than in interpreting between European languages. At the same time, recent literat­
ure from Europe shows that 'macro' perceptions of interpreting are gaining
ground generally, and a clear need is emerging for a more comprehensive ac­
count of the various components of the activity. In this volume (below, pp. 233-
242), Franz Pochhacker presents a good working overview of some objective si­
tuational factors affecting interpreter performance, and Sylvia Kalina has
embarked on an extensive corpus-based study of interpreter strategies (below, pp.
 Robin Setton, Republic of China 195

225-232). As a third component, as it were 'upstream' of these dimensions, the

theoretical and training project needs a framework capable of giving an 'ex­
tended-linguistic' account of tasks and difficulties in the comprehension, handling
and production of discourse.
The 'handling' stage is now being explored, after a generation of plausible but
unsubstantiated intuitive descriptions, by numerous process-oriented studies and
contributions from the new cognitive sciences, while useful practical experimen­
tation continues in the schools.
The last two decades have seen the emergence of an 'extended linguistics'
offering new tools to describe macro-phenomena in language communication,
raising hopes for a happy convergence of the arts and sciences in the overlapping
fields of pragmatics and cognition. But in applying discourse analysis to con­
ference interpreting we have to be very selective. It is doubtful at present whether
the mainstream exponents of these disciplines are even aware of this application:
the bulk of work still addresses written corpuses, while, even in the area of the
spoken word, Grice's maxims were devised with reference to everyday conversa­
tion. There are also numerous studies of interviews with mental health patients,
etc. This latter work will be of value to the community interpreter; but confer­
ence and media discourse, where the flow of communication is different from
conversation, and speech is part spontaneous and part prepared, will require a
different perspective, in which some traditional discourse analysis issues, such as
turn-taking, or narrative structure, will be of marginal relevance:

CORPUS (TEXT TYPE):

WRITTEN literary technical graffiti

(narrative) informative recipes

d iscursive
ORAL conversation pathology
turn-taking (patient
interviews)

Figure 3: Areas within current discourse analysis studies pertinent to

conference interpreting (shaded)

Conclusion: theory and pedagogy

The big question remains of what can be taught. Traditionally, students were
196 Teaching Translation and Interpreting 2

told to 'go away and learn the language', and later, to 'go beyond the words to
the ideas'; but in some cultures - where speech is traditionally viewed as ritual,
its content authoritative, and its forms canonical - this latter is not as obvious a
contrast as it is to Europeans, and may have as little effect as the instruction to
'learn it all by heart' in a Western liberal arts college. Consequently - leaving
aside the question of whether such 'exhortatory pedagogy' is ideal even in
Europe - we are forced, in Asia, to find higher, credible explanation and illustra­
tion. Years abroad are indispensable, and at the skills level, instructor demonstra­
tion is very effective, but they are not always enough; we should not abandon the
search for a structured pedagogy of conference interpreting articulated to the
mainstream of communication studies.
At this stage, certain avenues in discourse analysis look more promising than
others. Semantic and cognitive approaches still appear too abstract - and contro­
versial - for use in pedagogy: Kintsch and van Dijk's proposition-based represent­
ation of discourse content (1978), for instance, is called into question by Chafe's
suggestion that "knowledge is not stored propositionally ... the basic form of store
may consist of individuated events and objects, each with an associated analogic
content...until a need to verbalise them makes propositional decisions necessary"
(1977: 54, as quoted in Brown and Yule 1983).
Some other approaches offer a less ambitious, but currently more solid basis
for comparing input and output discourse; one example is the distinction between
'given' and 'new' items in a flow of discourse which provides a clear insight into
the distribution of pragmatic features such as intonation and pitch, repetition, el­
lipsis and anaphoric pronominalization, and last but not least, the notorious use
of the definite or indefinite forms a/the, all of which constitute an alien dimen­
sion to students with acquired English learned at undergraduate level.
The need for a formulation of discourse features at the macro-level becomes
clear when we have to address certain higher, text-level expressive functions
which are not generated automatically in trainees from outside the Aristotelian
rhetorical tradition. Examples of pragmatic weakness, in particular, from Chinese-
English student interpreter output show patterns which may not seem new, but
are more frequent and striking than in European training.
A tentative 'extended-linguistic' model showing seven degrees of liberty - or
error in Chinese-English interpreting is given in Figure 4 on the opposite page.
Features of the two languages as source (input) and tool (output) respectively can
be seen either as resources or as hazards.
In offering this table I am fully aware of the jaundiced eye with which prac­
ticing interpreters and others may by now look on 'explanatory' systems and
models. But this search for benchmarks to guide theory and training is prompted
 INPUT (Comprehension) OUTPUT (Production)
RESOURCES HAZARDS RESOURCES HAZARDS
Level Chinese English Chinese English
1 phonetic/ accents accents
phonological language knowledge homophony

2 morphological language knowledge compact nflected

3 syntactic A cues embedding 'sharpening approximation' topic-pred SVO

holding patterns, 'open
4 semantic A cues topic-pred grammar', filler material

culture- culture-
5 pragmatic context, extralinguistic specific diffuse abridgement, anaphora, low high
knowledge (general and substitution
topical - situation, players)
reference, presupp., implicature

6 rhetorical cues/transitions 'narrative' 'discursive' cohesion/ties marked

(familiarity with descriptivepersuasivei generalisation, abstraction discursive
Robin Setton, Republic of China

speech conventions) exemplary

professional selective listening, noting real-time (linear) 'toolkit'

(handling) lead/lag voice/delivery

7 'displacement effects' ! ! voice/deliv imagination

(humour, irony, metalanguage)

Figure 4: Linguistic/cultural resources and hazards in Chinese-English interpreting

197
198 Teaching Translation and Interpreting 2

by several considerations: firstly, to take advantage of the opening up of linguis­

tics to areas more relevant to our craft and to attempt to articulate our concerns
within an established field, in the hope of alleviating the difficulties translatology
has had in establishing an academic footing; and secondly, to propose a con­
ceptual vocabulary to formalise and vindicate the intuitions of a generation of
practitioners.
In conclusion we might ponder the rueful words of the late celebrated translato-
logist and grammarian of spoken Chinese, Chao Yuan-ren, in his 1969 essay on
the dimensions of fidelity in translation. After stressing that such dimensions -
semantic vs. functional fidelity, frequency of occurrence, style - are necessarily
non-measurable, dependent variables, Chao alludes to the then keen search for
algorithms to express such variables with a view to the quantitative study of qual­
ity for application in machine translation (Carroll 1966):
We are far from a workable definition of any of the dimensions, not to speak of formulating
a mathematical function with a view to maximizing its value... the present state of affairs is still
what in some of the formal disciplines is called the pre-systematic stage, which is another way
of saying that the ideas are still half-baked, ... but so far as that is concerned, in what field is
one not troubled with multidimensionality? (Chao 1976: 168)
Progress has nevertheless been achieved since the 1960's, if not towards the
mechanisation of discourse, at least towards a fuller understanding of its com­
ponents.
ON TEACHING NOTE-TAKING IN CONSECUTIVE INTERPRETING

Bistra Alexieva,
'St. Kliment OhridskV University of Sofia, Bulgaria

Note-taking: the pros and cons

The growing need for Consecutive Interpreters as mediators in bilateral ex­
change of exact and specific information in the fields of economy, science and
technology once again brings to the fore the much disputed issue about the role
of Note-Taking in Consecutive Interpreting. On the one hand, experience pro­
vides evidence in support of Mahmoodzadeh's statement that "Even with the best
of memory it is next to impossible for the interpreter to remember all that is said
in lectures, negotiations or press conferences, particularly if names, dates and fig­
ures are involved" (Mahmoodzadeh 1992: 235). On the other, we have data from
Gile's 1984 experiment in which the students asked to interpret without notes
"had a better restitution for names" than those who did it with notes (Gile 1990a:
81). One explanation may be that the students - who are in the process of ac­
quiring Note-Taking - put a greater effort into it, which diminishes their total
processing capacity (Gile 1990a: 80), and more specifically, impairs the work
their memory can do. In other words, their Note-Taking has not as yet become
a sufficiently good tool as memory reinforcer, for it uses up more of their
processing capacity and contributes less to the completion of the task.

A longitudinal study
This explanation seems to be corroborated by data from a longitudinal survey
which I have conducted with four groups of students at three different stages of
their training, namely:
1. The pre-Note-Taking stage in Consecutive Interpreting training (the first two
weeks) characterised by lower performance parameters, in particular, by greater
tension and faltering voices, which the students ascribe to their fear that they will
not be able to remember the Source Language text in its entirety.
2. Pre-Instruction Stage. During the third week they are given permission to
take notes of what they think could facilitate recall This helps them not only
subjectively but also objectively, for there is greater ease in their voices, and, in
terms of content retrieval, their performance shows a slight improvement. Their
self-confidence, however, wears out very quickly, for the 'left-to-right horizontal'
note-taking they use soon proves inadequate for the purpose.
3. The Note-Taking Instruction Stage. The most marked improvement in
student performance occurs at the very beginning of the Note-Taking instruction
stage, with the introduction of Vertical Writing (the node- or step-patterns),
200 Teaching Translation and Interpreting 2

which seems the easiest to acquire and which immediately makes them aware of
the importance of the lay-out of the notes for the reproduction of the semantic
relationships in the source text. However, further instruction in Note-Taking,
which introduces the currently used systems, brings about a trough in students'
performance, which remains consistently low for a comparatively long period. It
can be inferred that the period of Note-Taking acquisition is characterised by a
weaker memory operational capacity, for most of the students' energy is spent
on: (a) the trial-and-error strategy employed for deciding whether they should use
the signs offered by their instructors or whether they should invent their own; (b)
the effort to recall the signs and (c) decision-making about what information to
note and What should be remembered.
It is essential that trainees do not get stuck in this 'trough'. "Despite individual
variations due to personal aptitudes", as Gran suggests (1988: 1), there is "a com­
mon basic approach" in the modern note-taking systems developed by eminent
scholars and practitioners, thanks to studies on memory processes, language com­
prehension and production and Text Linguistics (for instance Ilg 1980 and 1982;
Allioni 1989).
Therefore, further optimization of Note-Taking and of Note-Taking Instruction
in particular can be attained only if our strategies take care of both the general
(systemic) and the specific (idiosyncratic) ways our memory operates in the com­
prehension and production of texts. This is why I shall address the following
issues: (a) the priorities of note-taking and (b) the code in Note-Taking.

The source language text content

In order to discuss the part of the source text content that should be written
down in order that the whole be recalled and reproduced, we must take into ac­
count: (a) the structure of the content, and (b) the procedures for retrieving it
(Beaugrande and Dressier 1981: 87).
The text content structure is a coherent network of sequences of scenes, each
consisting of configurations of predications (propositions = PN), in which there
is a hierarchy consisting of:
- Participants (Entities), which play different semantic and pragmatic roles.
They can be further subdivided into 'hyper', 'major' and 'secondary' depending
on their importance and frequency;
- Relationships between Participants or, as Halliday has it: 'Processes', (1985:
101-157), which can be static or dynamic, and occur under specific Circum­
stances, in terms of Space, Time, Manner, Quantity, Intensity, etc.
Thus the basic building block of the content of a text is the predication (= PN):
"Participant(s) involved in a Relationship/Process (static or dynamic) under speci-
 Bistra Alexieva, Bulgaria 201

fic Circumstances".
In the present context it is important that the Participants and the Predications
play different roles in the continuity of meanings in terms of information load
and control over the text. The different positions a PN can occupy in the hier­
archically organised textual network of meaningful relationships, allow for a dis­
tinction as follows:
1. Text-controlling (Hyper) PNs, (such as "NGOs have a major role to play...")
in this excerpt.
EXAMPLE 1
Non-government organisations (NGOs) have a major role to play in influencing Bulgaria's
evolving environmental policies. ... We all recognize the crucial role of NGOs in carrying out
education programmes and in influencing environmental policies, therefore better communica­
tion and coordination between the Ministry of Environment and the NGO community, as well
as between NGOs themselves is essential, particularly in view of the limited resources available
for such efforts.
2. Major PNs, containing at least one major Participant and representing rami­
fications of the hyper PNs in the recurrent structures ensuring the continuity of
meanings, for instance "NGOs influence environmental policies" at the beginning
of the second sentence; and
3. Minor PNs, which contain secondary Participants and Relationships (Pro­
cesses) and supply additional information about major Participants, as in "...the
limited resources available..." at the end of the second sentence in Example 1.

Procedures involved in the retrieval of the source text content

The recurrence of Predications and the role of familiarity in memory operations
(Jacoby 1991: 513-38) make it possible to predict that:
1. Predications which recur no matter whether at the micro- or macro-text level
leave deeper traces in our memory. Hence, their retention and recall will be much
easier, for "familiarity as a basis for recognition memory judgments [is] - ... in­
variant across full versus divided attention" (Jacoby 1991: 513). We can also ex­
pect greater ease in retention and recall of such familiar PNs and the configu­
rations of which they are part, because they represent big chunks of knowledge
rather than individual items, thus ensuring greater economy of search (Beau-
grande and Dressier 1981: 90), particularly when there is a good match between
the textual world and the global cognitive patterns (frames, schemas, plans,
scripts) with which the Consecutive Interpreter operates. We can therefore
conclude that familiar PNs can be reproduced without note-taking.
2. The PNs introducing new Participants, Relationships and/or Circumstances,
however, are more difficult to retain. In order to recall and reproduce them the
Consecutive Interpreter will need memory reinforcers; therefore, they should be
noted.
202 Teaching Translation and Interpreting 2

3. Notes, rendering the rhematic core of the source text content, can also act
as memory reinforcers of the remaining portions, for they can activate a larger
area of the semantic network and thus facilitate recall of what has not been
written down.

Training Strategies
The pedagogical implications of these assumptions are:
1. The trainees' acquisition of the skill to analyze the source text's continuity
of meanings as a hierarchical network of hyper, major and minor predications
(Participants, Relationships and Circumstances) will help the students to identify
recurrent PNs more easily and to distinguish between central and peripheral in­
formation. In my experience, exercises centering on this radically improve the
students' analytic capacity to discover what Beaugrande and Dressier (1981: 95)
term control centres in the text and to sort out its content into what can be easily
retained and what must be noted.
2. It follows that trainees in consecutive interpreting can gradually acquire the
skill to be selective, that is NOT TO TAKE NOTES OF EVERYTHING. This
will improve their performance, since it will reduce the time when attention is
divided between, on the one hand, listening and writing and on the other, writing
and looking for visual signs (face expression, gestures, etc.) relevant to the
speaker's intentions. The point is that writing is a strong distractor in face-to-face
communication.
The importance of economy in note-taking is illustrated by a test conducted
with four groups of Bulgarian students who had the following passage:
EXAMPLE 2
We must be global, we cannot wrap the environment up into neat national parcels, we cannot
say this is mine, that is yours. What happens in Darlington or Detroit today may affect Accra
or Djibouti tomorrow, and indeed vice versa.
Three of the groups were instructed to take notes only of what they thought
was important, with a step-pattern layout, in which they were expected to leave
empty slots, while one group (the control group) was left to its own devices. The
control group scored only an average of 59 out of 100 points, whereas the three
groups employing the "greater economy" strategy made a higher average score
(73 points), with the following notes taken from the best performances:
- wrap
env
parc
Darl & Detr
1 = 3
Acr & Djib
 Bistra Alexieva, Bulgaria 203

It will be appreciated that these extremely economic notes carry the most im­
portant information and activate the whole network of relationships.
The advantages of economic note-taking confront us with the need to develop
exercises which can better the students' selectional and inferencing capacity. The
latter can be improved by what I have labelled the 'Semantic Network Activation
Exercise'. Its goal is to develop the students' skills in building chains of infer­
ences from single key word(s) or phrase(s), supported by the minimum additional
context. Such an exercise can be conducted along the following lines:
(a) The students are given a word or a phrase exponent of a major participant
and/or the hyper or major PNs, as well as information about the communicative
situation. For example: using the formula CO2 and the additional information that
it is from a discussion of environmental issues, the students are asked to build
as many as possible semantic networks of which it might form a part of.
(b) The second version of the Activation Exercise consists of giving the stu­
dents notes and information about the hyper PNs of the text (but not the text it-
self). The following notes were for instance proffered with the information that
the text is about the ways in which pollution prevention can be made to pay and
the corresponding measures taken in the steel industry in Germany.
EXAMPLE 3

recy(cle)
90% water
solid wastes
In spite of the variety of versions these notes generated, there is a great deal
of similarity concerning the major PNs, for 90% of the students have made the
necessary inferences and produced sentences like "The German Steel Industry has
developed no-waste technologies. It recycles 90% of its industrial water and
converts 90% of the solid wastes into useful materials."

Choice of code in note-taking

The discussion about which part of the content should be written down con­
nects with the choice of code in Note-Taking. I shall address only the following
two issues, namely, the choice: (a) between language and non-language codes,
and (b) between the Source Language and the Target Language.

Language or non-language?
The guiding principles in the effort to find an answer to this question can, in
my view, be formulated as follows:
1. The code chosen for Note-Taking should have the maximum economy in
terms of: (a) the time used for writing it; (b) the effort of the hand, notably the
204 Teaching Translation and Interpreting 2

use of signs, which can be written down with one stroke of the pen, instead of
elaborate drawing and backwards movement; and (c) the mental effort to produce
the sign, for it takes much interpreter energy to make a complex combination of
squares, circles, arrows, crosses and language signs (usually abbreviations).
2. The second parameter is the amount of information the note sign can carry.
The greater its load, the greater its activation power. Answers from students cor­
roborate my view that a very general, and hence semantically poor, symbol such
as "", or ':' for "say" is hardly worthwhile, for it can be easily inferred from the
remaining notes. Example 4 serves to illustrate this point.
EXAMPLE 4
The representative of the Vienna agency discussed the repair of the nuclear power station in
Kozlodoui in the nearest future.

rep.nuc.st[
The note on the third step (the place for Argument Two of the PN), namely
"rep. nuc.st.", makes for easy retrieval of the first two empty steps, since the
Speaker is well-known to the interpreter and the Predicate of verbal expression
- Talk about' - can be easily inferred.
The requirement for a greater semantic load, however, should be combined with
the requirement for economy, particularly concerning the issues discussed above
under 1. (b) and 1. (c), for an elaborate time-, hand- and mental-effort-consuming
drawing can hardly be the best choice under the circumstances. It seems that the
natural language signs are best for carrying the greatest amount of information
in a comparatively small number of letters which can be written down easily and
quickly with the least effort and loss of time. The optimum code for Note-Taking
may therefore combine features from the following codes, each used for specific
portions of the source text semantic network, namely:
1. The Vertical lay-out (the step- and/or node-pattern) can be used for the re­
lationships within predications and configurations of PNs, and thus ensure ad­
equate recall of the semantic network of the source text;
2. Natural language words (written in full when short, or abbreviated when
long) are most suitable for semantically dense portions, as they can carry a large
amount of semantic information; and
3. Non-language signs and symbols which p n act as general operators signal­
ling: (a) the speaker's attitude, in terms of positive/negative evaluation and
different types of modality; (b) Relations between Participants in terms of ident­
ity, difference, similarity, comparison (along with the quality and quantity para­
meters); (c) Relations of causality, material implication (conditional) and
entailment, and (d) temporal and spatial coordinates.
 Bistra Alexieva, Bulgaria 205

Source language or target language notes?

The last issue I would like to address relates to the possibility of choosing be­
tween the Source Language or Target Language as the code of note-taking. In my
view, this choice is not a mere surface structure issue. It goes much deeper, for
it affects the nature of the process of Consecutive Interpreting, specifically its
first phase. If the Target Language is chosen for Note Taking:
1. The first phase will not only consist of Listening, Note-taking and Memory
operations (Gile 1990a: 80), but would also include Transcoding, that is an Inter­
lingual Transformation, for in order that the Consecutive Interpreter can take
down notes in the Target Language he must first mentally transform the textual
segment into it. It is not merely an increase in the number of functions that the
Consecutive Interpreter has to perform. It is more complex, for it involves the
workings of the interpreter's memory and the signs of the languages used in the
process: if notes are taken in the Target Language, then Transcoding should
precede both Note-Taking and Remembering.
2. The second reason for discouraging Note-Taking in the Target Language
connects with the reliability of the transcoding operations in the absence of the
context to follow, for the decisions taken may not be justified by what comes
afterwards. This may lead to erroneous notes. Certainly, the Immediacy of Inter­
pretation Hypothesis (Just and Carpenter 1980, 1987, as quoted by King and Just
1991) seems to hold good here too, for an interpreter will try to grasp the mean­
ing of each word as soon as he encounters it (King and Just 1991: 582). But his
strategy will attain optimum results if the "immediate" operation he performs on
the incoming portion is something like a first "scanning" only, accompanied by
flashes of transcoding, while the final decision about its meaning is postponed
until the Production Phase, after the reception of the whole segment.
One can therefore conclude that postponing Transcoding for the Production
Phase will: (a) make for less frustration in the memory operations relevant to the
retention of the source text and the selection of what should be written down, and
(b) permit the Consecutive Interpreter to employ the delayed commitment strat­
egy (Connine, Blasco and Hall 1991: 246), which is in fact the most important
advantage Consecutive Interpreting has over Simultaneous Interpreting.
In addition a strategy of Note-Taking in the Source Language creates a better
balance between the Analysis and the Production phases, which is important for
trainees, for in this way they can cope better with the first phase, a prerequisite
for the realisation of the communicative act in Consecutive Interpreting.

Conclusions
We may therefore conclude that the optimization of Note-Taking, and more
206 Teaching Translation and Interpreting 2

specifically, the optimization of the process of its acquisition on the part of

trainees, can be attained when the Consecutive Interpreters improve their skill
and efficiency in:
1. Analysing the source text content in terms of the specific contributions its
components have to the hierarchical network of continuity of meanings, which
allows for an optimum decision-making for determining what can be remembered
and what has to be written down for memory reinforcement, and
2. The selection of signs for their Note-Taking code, satisfying the major re­
quirements for economy of time, hand-and-mental-ejfort, high semantic load, ac-
tivation power and avoidance of the Transcoding overload of the first phase, thus
creating a better balance between the Consecutive Interpreter's functions during
the two phases and a more effective use of his processing capacity.

Notes
1. The survey was conducted during the academic years 1990/91 and 1991/92 with students from
the Translation and Interpreting Specialisation, organised by the English Department of Sofia
University.
2. Apart from the different intratextual values of a PN discussed above, it can also enter into a
variety of intertextual and situational relationships, that is, it can partially or completely coincide
with a PN or PNs (be they hyper, major or minor ones) from the preceding body of texts (the
preceding macro-text, Alexieva 1985: 196) or feature Participants and Relationships that may be
part of the communicative situation itself. The minor PN "The available resources are limited",
which occurs in Example 1, is a major PN in the statement made by the previous Speaker.
3. The students' performance was assessed by the parameters: preservation of referential content;
cohesion and coherence of the target text; fluency of delivery; and accent, prosody, rhythm.
4. The sentence in brackets was produced by one student only who said she knew that solid
wastes could not be properly recycled the way that water can, but should rather be converted into
a re-usable material. This stresses the importance of the Interpreter's knowledge of the subject
matter, but this is not a question I am going to discuss here.
5. The use of language can also facilitate the acquisition of the skill of Note-Taking because the
number of new signs the consecutive trainee has to learn will be smaller.
6. The frustration caused by the instruction to use the target language as a Note-Taking-code has
been documented by a series of tests with four groups of consecutive trainees, for the performance
of the fourth (control) group taking notes in the Target Language was markedly poorer than that
of the others. The greater capacity of our memory to retain the source text in the absence of an
effort to translate is also evidenced by the larger text segments reproduced as a Listening Task
only, rather than as an Interpreting Task.
 WHOSE LINE IS IT ANYWAY?
OR TEACHING IMPROVISATION IN INTERPRETING

Viera Makarovd,
Comenius University, Slovakia, and University of Warwick, United Kingdom

Background
This article deals with the experience of interpreter training at the Comenius
University in Bratislava, Slovakia. At our Department of English and American
Studies, students who study English in combination with other languages can
choose whether they want to become teachers or translators/interpreters. If they
opt for the translator/interpreter specialisation, they have classes of consecutive
and simultaneous interpreting for three years. They usually have different
teachers each year and I have been teaching simultaneous interpreting. The vo­
cabulary of most of my students has been excellent, because my colleagues have
stressed the importance of that throughout their interpreting sessions.
Unlike most of my colleagues I do not teach new vocabulary, but I do teach
improvisation. There are several reasons for this. First of all the students who
come to my classes usually have a good vocabulary already. Then, the acquisition
of new vocabulary can be done outside the class, by the students themselves.
And, finally, they have other teachers who insist on mastery of words and expres­
sions rather than on interpreting skills. In addition, it is my conviction that an
interpreter will never, even at the age of Methuselah, come to possess all the rel­
evant and potentially useful vocabulary relating to architecture, archaeology, as­
tronomy, biology, chemistry, demography and the rest of the alphabet. I fully
agree that interpreters have to specialize and should acquire as much knowledge
about their area of specialization as possible, but in interpreting situations we can
never avoid unpredictable vocabulary or unpredictable associations of the well-
known vocabulary. There is also the specific situation of the Slovak market,
where interpreters cannot specialize too narrowly, because there are not enough
events taking place and most interpreters have to work in several areas. I have
based my teaching method on a knowledge of this situation - having been inter­
preting in Slovakia for ten years, both consecutively and simultaneously. I see the
ability to improvise as a form of extralinguistic preparation, which complements
the basic, that is the linguistic preparation of an interpreter.
208 Teaching Translation and Interpreting 2

Teaching improvisation: components in the course

During the interpreting classes I give my students a prima vista text, usually
a sample of spoken style, not a recording of a written article or written material,
and I listen to their interpreting. During these sessions I accept renditions based
on lack of knowledge of a word or an expression rather than precise equivalents.
If, for example, cystic fibrosis is mentioned in the text, I accept the solution of
students who say this disease, the disease, the discussed disease. In this way they
provide the listener with at least partial information, rather than stopping in mid-
sentence to try to find the most exact word or expression, and thus losing whole
sentences and passages that follow. This is basically a question of setting one's
priorities, among which I see the rendering of at least partial information in a
particular textual unit much more important than belated full information which
would entail the loss of a whole passage in between and all that connected the
two passages in question.
This method of teaching requires a system of checks and balances, so that the
students do not abuse it. However, in the one year I have been teaching improvis­
ation, I have not noticed a single case of misuse. On the contrary, those students
who have been encouraged to substitute a generic term for the exact technical
(but unfamiliar) term, would either do this and continue interpreting without in­
terruption, or would retrieve the exact word or expression more quickly, because
their vocabulary would have contained it anyway. Certainly, in professional inter­
preting one has to insist on serious preparation before any commission, a careful
preliminary study of vocabulary. However, in any real-life interpreting situation
there are always at least two interpreters sitting in a booth, so one can always
look the word up in the dictionary or the conference materials or even ask the
delegates at the conference.
Improvisation cannot have a central place in real-life situations, it would be
against the ethics of the profession to interpret without a sufficient knowledge of
the area in question, but improvisation has to be known to the interpreter as one
of the possible ways of crisis management. In this connection I never fail to point
out to the students that it makes a world of a difference if the interpreter does not
know the Slovak for warted spindle tree when it is mentioned by chance and
only once or twice at that, whereas not to know the Slovak for warted spindle
tree at an annual conference of the Amateur Warted Spindle Tree Growers' Asso-
ciation amounts to one of the seven deadly sins. In other words, the decision
about whether one can or cannot improvise would depend on the centrality of the
unknown word or expression in the text to be interpreted. The more central the
particular technical term is in the given text, the more damage is done by
resorting to improvisation.
 Viera Makarovd, Slovakia and United Kingdom 209

After the initial interpreting in classes we always have a detailed discussion of

the vocabulary. The students are also encouraged to take tapes home and listen
to the recordings at their leisure so that they can pay due attention to every
detail.
Students are deliberately not given a full list of new words and expressions
before the lesson, since I try to ensure that they do not concentrate excessively
on new words, becoming accordingly less attentive to other parts of the text.
In most real-life situations in Slovakia simultaneous interpreters do not receive
any written conference materials beforehand. Sometimes they obtain certain ma­
terials immediately before the start of a conference. It must, therefore, be one of
the skills of an interpreter to skim a text for technical terms. That is why we
cover rapid underlining or highlighting in the class, in order to train students in
this skill. I bring any conference materials I have, distribute them in the class and
allow the students five minutes to underline whatever they find relevant. This ex­
ercise is geared towards real-life situations, where the technical terms have to be
picked up as one goes along, because terminology lists and other relevant ma­
terials are not available.
Another exercise consists in guessing the missing words and expressions from
newspaper and magazine articles that have been torn to pieces. Students realize
the relatively high degree of predictability in their mother tongue: in most cases
they can supply a close synonym for the missing word or expression. This train­
ing prepares them for situations when they do not hear something and have to
substitute a plausible synonym. In simultaneous interpreting they will not be able
to verify what was said and they will not be able to return to a passage they did
not hear in the first place.
Coming back to checks and balances: one has to bear in mind that an inter­
preter enters the communication process at a stage, when the participants have
already established certain patterns and have discussed the topic, so that if the
interpreter says the price will change, because he or she has not heard properly
whether it is going to go up or down, it is a much smaller mistake than it might
seem. The participants will probably have already discussed it or at least men­
tioned it, so they would not be too surprised by its occurrence and, should the
worst come to the worst, they can always ask for verification. My point is, that
again at least partial information has been given.
My speciality in training covers nonsensical texts. These lack logical syntax,
use incorrect gender of nouns and adjectives in Slovak, incorrect singulars and
plurals, have incorrect relations between verbs and objects, make several starts
and then stop without saying anything, use slang idioms, alternate rhythm from
very fast to irritatingly hesitant and so on. The students are forced to work with
210 Teaching Translation and Interpreting 2

such texts and dig out of them the 'pure nuggets of gold'. I have to use this type
of text because the majority of authentic Slovak texts available for interpreting
are stylistically poor, with incorrect cases of noun objects, bizarre collocations,
sentences that start anywhere, or make a couple of repeated attempts at saying
something, and fail completely. So exposure to the funniest distortions are no­
thing compared with real-life situations.
In order to prepare the students for all the unexpected situations in that wild
world called interpreting, I expose them to a succession of texts with constant
switching of languages. In the first few cases I record a couple of bars of music
in between passages in different languages, to establish the Pavlovian reflex.
Later I withdraw the music and still expect them to salivate at the right moments.
I make recordings where I lisp, stammer, speak with a heavy accent, cough,
omit words, create noises that prevent the students from hearing and then, listen­
ing to their performance, jump with joy if they still manage to produce something
by, for instance, adding neutral pieces of information where they heard or under­
stood nothing.
No matter how much interpreters specialize, there will always be gaps in their
knowledge and I am sure it is always better to provide the listeners with at least
partial information, if the interpreter cannot provide complete information. In
many cases it is a question of something or nothing at all.
My segment of our interpreter training programme is based on the 'forewarned-
is-forearmed principle'. Let me nevertheless stress that I am the first believer in
hunting for the meaning of words and expressions, in reading everything one can
lay one's hands on; I am a staunch believer in constant language enhancement,
I recognize the need for specialization, etc. etc., but in any interpreting I see
space for flexibility and I would argue that this gap can be filled by clever,
skilled improvising, and also that improvisation can be taught.
 TRAINING FOR REFUGEE MENTAL HEALTH INTERPRETERS

Nancy Schweda Nicholson, University of Delaware, USA

Introduction
With increasing frequency, interpreter training courses meeting specialized
needs are being developed in the United States. One such class for Southeast
Asian refugee mental health interpreters was held at the University of Minnesota
under the auspices of the Twin Cities Interpreter Project (= TCIP) during the
summer of 1991. The languages involved were English, Hmong, Vietnamese,
Cambodian (Khmer) and Lao.
As a principal instructor, I assisted in planning and implementing a specialized
curriculum geared to the working requirements of this group.
This article (1) presents the background for the course and hence a framework
for a discussion of the pedagogical techniques; and (2) examines successful
teaching strategies when the instructor and the students share only one working
language, in this case, English. The discussion concludes with my personal obser­
vations regarding the feasibility of 'generic' interpreter training.

The Twin Cities Interpreter Project: background

Formed in 1989, "the Twin Cities Interpreter Project (TCIP) is an informal,
voluntary, interdisciplinary coalition of professionals in Minneapolis and St. Paul
concerned with improving the quality of language interpreting services for
refugees and others, especially in health care facilities and other social services"
(Downing and Tillery 1992: 5). It continued and expanded on previous work and
also drew on local experience regarding the provision of interpreting services in
health care situations (or, perhaps better stated, the lack of such services!).
Initially, I became a consultant to the project in the area of curriculum develop­
ment and then was invited to teach during the intensive week of the Interpreting
II course (discussed below).
The Twin Cities area is home to many Southeast Asian refugees, most of
whom need access to social services. In the context of language and cultural
barriers, making the initial contacts and securing the required assistance are often
difficult and formidable tasks. There is a great need inasmuch as approximately
one out of every twenty people in the Twin Cities (Minneapolis and St. Paul,
total combined population of about 600,000) is a Southeast Asian refugee (Min­
nesota Department of Human Services 1988). The substantial refugee and immi-
212 Teaching Translation and Interpreting 2

grant population clearly warrants attention regarding cultural and linguistic issues.

The TCIP Community Interpreter Training Program

The Program offered two separate courses. Beginning in April of 1991, Inter­
preting I was open to interpreters employed at least half-time in medical and
mental health settings. The course consisted of 40 hours of very basic interpret­
ing training, and was characterized by large group lectures and small group
practice.
An advanced course, Interpreting II, was offered in July and August of 1991
to participants who had completed Interpreting I. Both Interpreting I and Inter­
preting II were free of charge to residents of Minnesota, thanks to outside fund­
ing.1 Interpreting II provided ninety additional hours of training for twenty-five
individuals. The trainees' backgrounds varied significantly: some had college
degrees and others did not; some had been interpreting for ten years while others
had only six months' experience. Unlike Interpreting I, Interpreting II began with
an intensive week of forty hours of instruction, during which I was a principal
instructor. The class then continued meeting for ten hours per week (Friday after­
noons and all day Saturday) for five more weeks. The goal of Interpreting II was
to raise the participants to a professional level in medical and mental health
interpreting by the end of the course. Included in the curriculum were (1) guid­
ance on the development of a system for organizing and storing vocabulary; (2)
instructor input on how to successfully internalize and implement strategies for
self-assessment as well as for giving and receiving peer feedback (See Schweda
Nicholson 1991 and 1993b for a detailed discussion); (3) extensive practice
aimed at improving English fluency; and (4) scenarios which highlighted the in­
terpreter's role and ethical responsibilities (See Schweda Nicholson 1994 for an
examination of ethical issues in community interpreting).
The training team was composed of health care professionals, certified inter­
preters, English-language specialists, and bilingual tutors. Unlike conference in­
terpreter trainees, many of the participants were not fully competent in English.2
For this reason, trainers included English-language specialists as well as bilingual
tutors. These experts worked with each language group to improve speaking
skills, explain cultural differences, and assist with vocabulary building, both in
a general sense and, more specifically, in the areas of medical and mental health
terminology.
Most of the participants had preconceived notions about what the training
would entail. They thought it would be a vocabulary-building course, not one in
the techniques of interpreting. The students were obsessed with vocabulary and
incorrectly believed that, once they increased their knowledge of technical terms,
 Nancy Schweda Nicholson, USA 213

their job as interpreters would be trouble-free. In their minds, terminology ex­

pansion was the key to becoming a good interpreter.

Teaching strategies
My work with the Interpreting II students consisted of (1) sessions in English
with the entire class, including all languages; (2) small group work when specific
language combinations practiced consecutive interpreting during either self-gener­
ated or teacher-assigned scenarios; and (3) informal consultations with students
between or after classes to offer one-on-one advice and address ethical questions
which trainees might have been reluctant to raise in class.
Consisting of both lectures and interactive exercises, the large group sessions
covered much material, such as: (1) the differences between oral and written lan­
guage; (2) the fundamentals of consecutive interpreting; (3) the role of linguistic
and extralinguistic factors in the interpreting process (Schweda Nicholson 1987);
(4) the rationale for note-taking (Schweda Nicholson 1990a; 1993a); (5) the role
of contextualized and non-contextualized information (Schweda Nicholson
1990a); and (6) note-taking techniques (Schweda Nicholson 1993a). The term
'lecture' is actually misleading, as students were constantly asking questions and
relating their own experiences to the issues raised. Interruptions and discussion
were heartily encouraged.
I provided more handouts for the students than I usually would because I had
been informed that it was difficult for the trainees to take notes and listen in
English at the same time.
Moreover, the intensive week also included a section on pre-interpreting exer­
cises which I have developed, and these met with an enthusiastic reception
(Schweda Nicholson 1990b).
The bilingual tutors were an integral part of the training and contributed in a
number of ways. They would, for example, describe their own experiences, pro­
vide me with cultural elucidation, and explain difficult concepts and words to
their group in its native language.
After the first day of preliminary lectures and discussion, much of the rest of
the week was spent on consecutive interpreting practice in small groups. As the
primary instructor, I roved from language group to language group, observing the
scenarios and providing feedback in areas such as note-taking, public speaking
skills, choice of English words, phrasing, and ethics. Although not a speaker of
Hmong, Cambodian, Lao or Vietnamese, I could, in fact, comment on extralin-
guistic behaviour which was often directly related to a linguistic challenge. This
situation can be illustrated by a typical example. During a small group session
with the Vietnamese students, I noticed that the interpreter (obviously stymied)
214 Teaching Translation and Interpreting 2

hesitated, crossed and uncrossed her legs, and looked away from the principals
several times. I took note of the English subject matter and, after the scenario,
questioned her about the problems. In each instance, the trainee explained that
she was searching for a specific medical term. Inasmuch as the students knew
that I did not speak their languages, they were often surprised that I could focus
precisely on a point of difficulty. It goes without saying that, as I learned more
about their culture, I was also able to identify more subtle manifestations of lin­
guistic problems through their extralinguistic behaviour.
Taking notes along with the students was an effective strategy. At the end of
an exercise, I showed my own notes to the trainees so that they could observe the
techniques of an experienced note-taker. This was never done with the intention
of imposing specific methods; the notes were simply offered for the students' ex­
amination. Inasmuch as note-taking is a highly personalized activity, the trainees
made their own decisions regarding the adoption of particular symbols, abbrevia­
tions, and spacing.
It was also of great value that I, as the principal instructor, was a native
speaker of the language which was non-native for all of the participants. While
it is true that an instructor in this position cannot offer feedback and guidance to
students regarding native language word choice and phrasing, she can provide
much-needed input on use of the non-native language and native-speaker accepta­
bility as well as cultural norms and appropriate extralinguistic behaviour. In my
experience, this is the area where students generally require most assistance and
direction.

Conclusion
When I first learned of 'generic' interpreter training courses in which the in­
structor and students share only one working language, I believed that such
classes could not be successful. My own experience as a trainee and a trainer
convinced me that it would be impossible to provide interpreting instruction
without knowing both languages well.
I firmly believe that the situation in the TCIP training program was as close
to ideal as possible for this type of class. If the instructors are native speakers of
the students' acquired or weakest language(s), the trainees receive valuable input
on linguistic and extralinguistic skill-building. Moreover, the presence of English-
language specialists as well as bilingual tutors added much to the overall effect­
iveness of the course. Students were always provided with feedback and guidance
in a variety of ways and in a variety of languages. The instructor-student feed­
back was strengthened by the student-student feedback, where the various group
members offered comments to one another, both within their own language com-
 Nancy Schweda Nicholson, USA 215

bination and outside of it. The learning environment provided an opportunity for
much interaction.
Looking back, it would have been ideal if the trainees' English competence had
been much stronger at the outset. However, we live in the real world, not an
ideal one. As mentioned, the individuals selected for the training course were
already part of the system - they were interpreters in hospitals and other health
care facilities. It seemed logical to provide the skills and knowledge which would
enable them to perform at a higher level. During the course, especially during the
lecture/discussion sessions with the entire class, I consciously tried to adjust my
vocabulary and also paraphrased and repeated myself when I noticed some
puzzlement in the group. Additionally, more than the usual number of examples
were provided to clearly illustrate all points.
From an instructor's perspective, my experience was extremely gratifying inas­
much as each and every one of the trainees was thoroughly interested and highly
motivated. The students were hungry for training and were there to learn. They
were excited about making videotapes and receiving critiques so as to continually
better their performance.
My additional experiences offering courses in consecutive interpreting to Lan­
guage Specialists for the Federal Bureau of Investigation (with up to eight lan­
guages in one class) and to members of the London-based Institute of Translation
& Interpreting add weight to my strong endorsement of such programs.
To conclude, a skeptic has now been converted. There is no doubt in my mind
that this type of course can meet with success, given the right mix of students
and instructors. Training is in great demand, especially at the community level,
which includes the courts, health care, and a diverse group of social services. As
a result, one notes more and more specialized curricula springing up worldwide.
From a practical perspective, the approach described allows interpreter trainers
(who are a rare species and whose working languages are generally traditional
conference languages) to reach out to far greater numbers of students. Classes
like mine offer at least a partial solution to the paucity of interpreting courses
available. If the bottom line is 'generic' training or no training at all, the former
alternative is the most workable and practical option.

Notes
1. Funding was obtained from the Office of Refugee Resettlement through the Refugee and
Immigrant Assistance Division of the Minnesota Department of Human Services.
2. Language ability are also discussed by Christian Heynold and Robin Setton in the present
volume. Heynold refers to the necessity for interpreter trainees to speak both fluently and natural­
ly in their mother tongue (above, p. 13-14). Setton stresses that, in Taiwan, he often has difficulty
finding prospective students whose English skills are strong enough to undertake interpreter
training in Chinese-English (above, p. 185).
 INTERVENTION AS A PEDAGOGICAL PROBLEM
IN COMMUNITY INTERPRETING

Leonor Zimman, King's College, University of London, United Kingdom

This article does not define good or bad interpreting but focuses on ethics in
interpreting, notably on how to teach adequate and balanced intervention in com­
munity interpreting situations.
The methods and teaching materials discussed in this article were developed
in connection with my courses for community interpreters.

Background: the controversy about intervention

Training courses for community interpreters are relatively new in Britain.
As in all other countries, teachers and students are immediately faced with the
crucial controversy concerning the intervention of the community interpreter (who
will henceforth be referred to by the neutral 'she').
Should she only transfer words and concepts from L1 into L2 and from L2 into
L1? Or is it part of her role to intervene, to be an advocate and to act in some
informative or advisory capacity? Is there any difference between a conference
interpreter and a community interpreter?
Anderson (1976: 209-216) was clearly aware of the conflicting role of the
interpreter:
Understanding the role of the interpreter may also aid understanding of interaction between
people of different statuses and backgrounds within a single-language community. For example,
paraphrasing, which may be viewed as a special case of translation, commonly obtains in la­
bour negotiation, doctor-patient interaction, parent-child interaction and resolution of disputes
between dominant and minority groups...
The interpreter commonly serves two clients at the same time. He is the 'man in the middle'
with some obligations to both clients - and these obligations may not be entirely compatible...
These [situations] illustrate the fact that the interpreter's role is always partially undefined - that
is, the role prescriptions are objectively inadequate... (Anderson 1976: 209-216).
Conversationalists have interpreted dimensions of language use in conversation
in different but complementary ways: they have been seen
(a) as a negotiation of mutual understanding when speaker and hearer have dif­
ferent perspectives (Stubbs 1983),
(b) as governed by rules which include negotiation of the speakers' rights and
obligations during the speech event (Myers Scotton 1983),
218 Teaching Translation and Interpreting 2

(c) as governed by principles and maxims respected by speakers (Grice 1975),

(d) as knowledge of norms for language usage (Hymes 1977).
However, there is an overall agreement that social context affects the use of lan­
guage. Therefore, the rules governing conversation are not only linguistic rules
or linguistic 'maxims' but also partly socio-culturally determined.
When two monolingual people hold a conversation there is constant monitoring
of the cultural and linguistic norms governing the speech situation; both of them
know the 'maxims' of conversation, know their rights and obligations. They also
'know how' to flout the norms and share an understanding of this flouting.
When speakers do not share a common language and culture the interpreter
acts as a mediator who can arguably be expected to 'interpret' not only words but
also the conversational norms which include use of language according to role
relationships. In other words, who is who in the speech situation, the respective
status of the two parties, their role in the speech event and in the situational
context. Hymes writes:
Communities indeed often mingle what a linguist would distinguish as grammatically and as
socially or culturally acceptable. Among the Cochiti of New Mexico J.R. Fox was unable to
elicit the first person singular possessive form of "wings", on the grounds that the speaker, not
being a bird, could not say "my wings".
The nonidentity of the two kinds of rules (or norms) is more likely to be noticed when a shared
variety is a second language for one or both parties. Sentences that translate each other gram­
matically may be mistakenly taken as having the same functions in speech, just as words that
translate each other may be taken as having the same semantic function (Hymes 1977: 54).
Given its intercultural features there are then specific points that need to be
borne in mind in a community interpreting speech event. There are cultural
values, attitudes and beliefs that are not shared by the sender and the addressee
between a non-English speaking client and an English speaking service provider
in an English context.
In community interpreting situations, such a linguistic and cultural 'mismatch'
of conversational maxims, rights and obligations in role relationships, exchanges
and values may need to be matched by the community interpreter - most notably
so when the mismatch concerns knowledge of the system of the service needed
by the non-English speaking client.
A social worker I interviewed on the subject, Elena Pollard, said that the com­
munity interpreter may "explain to the client how the system works because it's
very frightening for some clients to actually be interviewed by housing officers
and social security officers without really knowing what it is about" (Zimman
1989c).
The same social worker pointed out that, conversely, "unqualified interpreters
tend to get overinvolved, overidentified with the client, particularly if the client
comes from the same culture" (Zimman 1989c). And a community interpreter,
 Leonor Zimman, United Kingdom 219

Maite Bell, explained during an interview that she thought that the "limit [of the
community interpreter] has been reached when it comes to making decisions for
other people. They need the information but they don't need you to make up
their minds for them" (Zimman, 1989b).
The Guide to Good Practice of the British Association of Community Inter­
preters (1989) lists four reasons why a community interpreter can intervene:
- to ask for clarification if she has not fully understood the concept she is being
asked to interpret;
- to point out if a client has not understood the message although the rendition
was correct;
- to alert a client to a possible missed inference. (An inference is information
which has not been stated but the knowledge of which may have been as­
sumed);
- to ask a client to modify their delivery to accommodate the interpreting
process.
There is no controversy about the first and fourth cases; there is some contro­
versy about the second; and the third one is heatedly debated: in order to decide
between wrong or right inferences the community interpreter needs to be bicul-
tural, have information and knowledge about both social systems, be familiar
with the subject matter and be able to judge for herself 'on the spot'. Shackman
has clearly stated the problem:
A community interpreter has a very different role and responsibilities from a commercial or
conference interpreter. She is responsible for enabling professional and client, with very dif­
ferent backgrounds and perceptions and in an unequal relationship of power and knowledge,
to communicate to their mutual satisfaction (Shackman 1987: 18).

The controversy and trainees

It is important to define the role of community interpreters to interpreters, since
they will at one extreme meet with the statement that they should be 'language
machines' who translate from L1 to L2. At the other extreme it is said that a com­
munity interpreter can take over the interview and should have the power to do
so. Between these two extremes there is a continuum of views on the role of the
community interpreter. Perhaps the controversy reflects the fact that circum­
stances and role relationships that are crucial for defining the role of the com­
munity interpreter are not always taken into account. Instead of being a partici­
pant in a speech event and the person responsible for making sure that the lin­
guistic or extra-linguistic manifestation of role relationships and culturallinguistic
norms are transferred in the interpreting, she is merely considered a professional
working in a vacuum.
Essentially, then, there is less controversy than one may think, but this is far
220 Teaching Translation and Interpreting 2

from obvious to the uninitiated. To people uninvolved with community interpret­

ing professionally, it seems tangible enough, but I believe that among clients (ser­
vice providers and non-English speakers alike) it has led to their mistrusting
interpreters.

The interviews
Accordingly I considered it appropriate that the prospective community inter­
preters (students/trainees) were confronted with the controversy in order to identi­
fy and clarify their professional role. So I carried out four interviews.
The four people interviewed were selected because they represented the whole
spectrum of viewpoints mentioned above and because they were familiar with
community interpreting either as users, organizers or interpreters.
Interviews were carried out with two aims: in addition to illustrating the con­
troversy, I also wished to use the interviews as authentic material on the topic for
the students' work.
I carried out the interviews by means of a questionnaire which served as a
guideline and allowed for open answers. Each of these focused on the contro­
versy but differed slightly according to the position of the person interviewed.
In order to illustrate the procedure, I reproduce here one of the questionnaires:

QUESTIONS FOR INTERVIEWS

(24th of July 1989)

INTERVIEW 1
Elena Pollard, you have worked as a Social Worker in London for a long time. I would like
to ask you a few questions related to communication with clients who do not speak English.

1) When you work with an interpreter, do you expect anything more from her than translating
words?
2) What else might she do beyond literal interpreting?
3) Would you be prepared to give some information to the interpreter before the interview?
4) Would you be prepared to agree for the interpreter to give some relevant information to the
client?
5) In your opinion, could the interpreter act on behalf of the client as an advocate?
6) Can you summarise what you see as the role of the community interpreter?

The interviews as teaching material

The interviews have been incorporated in course work, and I shall present the
teaching materials I wrote in the order they were used in the training courses I
ran in 1989-1990 at Morley College, London (Zimman 1989a).
In the first exercise the trainees are given a worksheet. Then they listen to the
recorded interviews in the language laboratory individually and work, in their re­
spective languages, through the worksheet. There are four different worksheets,
 Leonor Zimman, United Kingdom 221

each tallied to one specific interview. This, for example, is that for the above
series of questions:

INTERVIEW 1 - Elena Pollard

Elena Pollard is a bilingual social worker. She has worked extensively in her mother tongue
(namely Spanish) as well as through interpreters in order to communicate with clients who do
not speak Spanish or English.

1 - Decide whether the following statements are true or false in the opinion of the interviewee.

2 - Explain and elaborate your answer:

(a) I do not need anybody to tell me what my job is no matter whether my client speaks Eng­
lish or not
(b) A professional community interpreter needs to have a good knowledge of the agencies she
is working for
(c) The professional community interpreter needs to organise what the client says and bring it
to the level of the agency worker
(d) Any information given to the interpreter beforehand jeopardises the job of the agency
worker
(e) It is a good idea for the interpreter to give some information to the client about how the
system works before and after the interview
(f) The community interpreter must occasionally be allowed to take over the interview
(g) The interpreter should never act as an advocate
(h) Agency workers do not use interpreters because they do not know that they are available

3 - List the qualities and/or comment on the role of a good community interpreter according
to the interviewee.

In the next exercise the students form four groups, one for each interview, they
exchange the results of their exercise and advance their own view on the subject.
The main objective is to make them aware of the controversy about the role of
the community interpreter and at the same time to help them to identify their
future roles as community interpreters.
The third step is to confront trainees with real life cases so that they can use
the preceding discussion and decide the most adequate course of action in each
situation.
For this purpose the trainees are presented with five cases which they discuss
in pairs. After this discussion they have to give reasons for agreeing or dis­
agreeing with the course of action taken by the interpreter in each case. The fol­
lowing are examples of cases the trainees are presented with at this stage.

CASE A
This is an extract from an interview between the headteacher of a primary school and a Chilean
whose eight year old son was repeatedly bullied in the school's playground by a classmate. The
father expresses his anger at what happened and the headteacher explains that the child re­
sponsible for the bullying has been suspended for two weeks:
222 Teaching Translation and Interpreting 2

SPANISH SPEAKING FATHER: (angry) But, what have you done about it? Tell me what
you've done about this barbaric behaviour.
BILINGUAL INTERPRETER: (neutral tone) What have you done about this intolerable
behaviour?
HEADTEACHER: We had a chat with the child and his parents, and later on he was suspended
for two weeks.
BILINGUAL INTERPRETER: We had a chat with the child and his parents, and now he
cannot attend school for two weeks.
SPANISH SPEAKING FATHER: (tells the interpreter in confidence and in a sarcastic tone)
So she is telling me that the child who insulted and beat my kid has been sent on holidays for
two weeks; in bloody England a child who is a bully gets a prize; what a cheek!
(still angry) Can you tell me what sort of regulations you apply for discipline?
BILINGUAL INTERPRETER: (neutral tone) Can you tell me the regulations you apply for
discipline?

CASE B
A community interpreter is asked to go to a police station to interpret for an elderly Chinese
man (who speaks no English) who has been mugged. This is part of the interview:

POLICE OFFICER: Can you describe the clothes the mugger was wearing?
BILINGUAL INTERPRETER: Can you describe the clothes the mugger was wearing?
CHINESE ELDERLY MAN: Well, he was wearing a blue.... No, I think it was black, no, no....
It was definitely grey, a grey coat, it was too big for him.
BILINGUAL INTERPRETER: Well, he was wearing a grey coat which was too big for him.
POLICE OFFICER: Can you tell us the colour of his eyes, his hair and other features you
remember?
BILINGUAL INTERPRETER: Can you tell us the colour of his eyes, his hair and other
features you remember?
CHINESE ELDERLY MAN: I don't think I saw his eyes very well. Everything happened so
quickly, but as he was blond his eyes must have been blue, yes, that's right. I think he had blue
eyes.
BILINGUAL INTERPRETER: He was blond and I think he had blue eyes.

CASE C
You are an interpreter and a Turkish female client (who speaks no English) phones you at 9
p.m. (you interpreted for this person once in the past, that is why she has your phone number).
She tells you: 'My son has just fallen down the stairs. My husband isn't home yet. He won't
stop crying and his arm is swollen. Please help me. What shall I do? Where should I go?'
You, the community interpreter, tell the client: 'I'm sorry, I can't really do anything. I'm only
an interpreter. If you still need an interpreter tomorrow, contact the Borough's Interpreting
Services in the morning, they will make the necessary arrangements'.
(adapted from Shackman 1987)2

It is immediately understood that, ideally, the trainees should be able to come

up with good and reasoned explanations of the course of action they would take
in each case, bearing in mind role relationships, intercultural differences, lin­
guistic and cultural norms, knowledge or lack of knowledge of the system, vari­
ations in interpreting needs for different settings. The list is far from exhaustive,
 Leonor Zimman, United Kingdom 223

but it suffices for the bottom line: it is all right to summarise in some cases but
in other circumstances a summary may send an innocent person to jail.
As a fourth step the trainees are asked now to role-play an interpreting scen­
ario. In groups of four, three of them can role-play the scenario and one can ob­
serve their performance for later comment. One group is asked to perform the
interpreting assignment strictly in accordance to the briefs. In the other groups,
the trainees who are 'the interpreters' are allowed to modify the brief according
to their own informed judgement of the situation. The scenario all groups must
enact is as follows:

ROLE-PLAY
PATIENT'S BRIEF: You are a young Moroccan woman. You have had a lot of bowel trouble
and constipation. You saw the doctor a week ago and he gave you a laxative. The laxative
worked and you want some more because you are constipated again. You respect your doctor
and agree with all his advice. You say 'yes' and 'thank you' to everything he says and leave
the surgery without being sure of what cereal you should eat (you have never eaten cereal for
breakfast before), what 'bran' is and where to buy it.
INTERPRETER'S BRIEF: Neither the doctor nor the patient have briefed you. You have not
met this patient before. The doctor has asked you to interpret for her. You do not know what
is wrong with the patient. You interpret everything that is being said. At some point you have
the impression that the patient is saying 'yes' without actually understanding what the doctor's
advice is. At the end of the appointment you say goodbye to both parties and leave. You wait
for the patient outside and ask her if there is anything she did not understand. You then explain
to the patient a few things about high fibre diets without informing the doctor you have done
so.
DOCTOR'S BRIEF: Patient with bowel trouble. You saw her a week ago and prescribed some
laxative. You want her to re-establish natural patterns, so you do not want to give regular laxat­
ives. You ask her about regularity, pain, flatulence, and the patient's normal diet. You give
your normal advice, which is to eat bran with the morning cereal. (Adapted from Shackman
1987)2

The trainee groups enact the scenarios for the whole group. Accordingly every-
body can see the problems and judge on suitability of the trainee's intervention.
As a fifth and final exercise after all the groups performances it would be inter-
esting to have comments from the observers.
Given all the material in classwork it should be possible for the trainees to
make some conclusions about the role of the community interpreter and the type
of intervention she ought to make. The main points to surface in these con­
clusions would be:
a) Types of intervention
b) Circumstances that determine the intervention or non-intervention
c) Consequences of underintervention
d) Consequences of overintervention
224 Teaching Translation and Interpreting 2

Conclusion
In practical work the result of both non-intervention and intervention in com­
munity interpreting may be disastrous unless all the circumstances are borne in
mind by the mediator. Although the perfect and well defined roles do not exist
for the community interpreters, it is of immense importance that this fact is
brought home to would-be interpreters at a very early stage of their career. In so
doing, the teacher can, however also make each trainee fully aware of the fact
that each community interpreter has to develop a clear and justifiable role for
herself with the capacity to adjust to particular circumstances, concepts and situ­
ations which will depend on such factors as the service (police, social services,
health, etc.), the clients (both service provider and non-English speaking client),
the cultural differences and the immediate circumstances.

Notes
1. There are minor changes in the questionnaires and assignments cited in the main text to make
it easier to read for outsiders/persons with no first-hand knowledge of the people or topics
involved.
2. Reproduced with the kind permission of the copyright-holder, National Extension College, 18
Brooklands Avenue, Cambridge CB2 21 IN, United Kingdom; telephone + 44-223-316644.
 ANALYZING INTERPRETERS' PERFORMANCE:
METHODS AND PROBLEMS

Sylvia Kalina, Heidelberg University, Germany

History of the corpus

This article is about empirical research in interpreting, without which any
discussion of theoretical models will remain in the abstract. Such empirical
research is also, I think, a prerequisite for developing adequate methods of
teaching interpreting, and it should therefore be of help both to the re­
searcher and the teacher.
In a previous article (Kalina 1992), I set out to suggest a model of dis­
course processing which I then applied to interpreting, emphasizing that
there are a number of specific strategies characteristic of this type of pro­
cessing and especially of simultaneous interpreting.
As empirical studies of simultaneous interpreting are much rarer than the­
oretical contributions, I felt it would be advisable to look for authentic em­
pirical evidence for my hypotheses. Until then, I had based my assumptions
on experiments carried out in a more or less artificial setting, often with ma­
terial collected from student interpreting performances, and I felt I needed
some real life interpreting by professionals to see whether the strategies I
had identified could be found there as well
Unfortunately, it is not easy to get hold of professional interpreting re­
cordings. It may be that the conference organizers and contributors wish to
keep their discussions confidential or want to publish their papers them­
selves, or that the interpreters claim that the conference in question is not at
all typical, or that they cannot work easily and spontaneously when they
know they are being recorded by a colleague, etc.
However, in 1992,1 had the unexpected chance of working at a three-lan­
guage (German, English, French), three-booth simultaneous conference
where all the booths were being taped.
The organizers and my interpreter colleagues working at the conference
agreed to let me use the recorded material to analyze interpreting strategies,
and I am most indebted to them for their consent. It is all the more valuable
as they agreed unanimously that the conference was very difficult, conditions
were far from perfect, and there was insufficient time for preparation, so that
interpreters were unable to perform to their normal standards.
226 Teaching Translation and Interpreting 2

As the recordings of the original conference were made by the company

that supplied the technical equipment, we obtained only single track re­
cordings of the original speakers and the three booths respectively, and
whenever tapes were full and needed to be changed, a few sentences were
lost. Sound quality was another problem, with people not waiting until they
had a microphone, etc. - all the well-known irritations of simultaneous inter­
preting work.
Despite these problems, when I returned from the conference with more
than 20 tapes of 90 minutes each, I felt as if I had stumbled on a goldmine
and could set about immediately to exploit my riches.
Recordings and transcripts of good interpreting performance are under­
standably more readily accessible and have formed the basis for most
research hitherto. Some authors tend to neglect difficult circumstances, and
some even claim that they will not work under such conditions. However,
such situations are, in fact, part of interpreters' daily lives and must
therefore be coped with. So how do interpreters cope with them? And can
this be taught?
As stated above, we had already collected some material based on experi­
ments in artificial settings, with students as interpreters, in order to study
certain well-defined research questions, such as consecutive interpreting and
note-taking, error analysis, results of think-aloud protocols, user attitudes and
expectations (Kalina 1991). We had even called in audiences with as little
knowledge of English as possible to make our setting similar to real-life
conditions. Nevertheless, we always felt after such events that one important
feature had been missing, namely the user's complete dependence on the
interpreter's rendition; and that another feature namely the presence of other
student interpreters and, even more crucial, of teachers, might affect the
results.
Moreover, we wondered whether perhaps professionals make use of strat­
egies in ways other than those chosen by trainees. So one hypothesis to be
tested was that professionals' strategic processing differs from that of
trainees. This leads to the question: what are those differences? In order to
obtain more information about these questions, we organized a 'mock
conference', where we tried to have the same conditions as at the 'real'
conference: It therefore took place during the same hours on the same days
of the week and the same documents were made available at the same time
as at the original conference, etc. There were inevitably serious deficiencies
and differences which could not be overcome: we had only audio recordings
(which, in fact, resembled the real-life conference, as the view of the
 227
Sylvia Kalina, Germany

speakers had been very poor), and furthermore there were no legal experts
or high-ranking officials among the audience.

Methodological problems
In studying real-life conditions and professional interpeting, one problem
is that one will rarely find several interpreted versions of the same text, a
fact which makes direct comparison impossible. Even a large volume of
authentic material will thus yield only limited information. Furthermore,
there is a high degree of individuality in interpreting, and this makes it even
more difficult to draw valid conclusions. Accordingly, large numbers of
studies need to be carrried out to validate any given hypothesis.
There are certainly strategies that vary depending on the languages, pro­
cessing direction ('A' language into 'B' or vice versa) and cultural differ­
ences, but for the purpose of this study we tried to identify those strategies
that interpreters seem to use when working with English and German (and
French) and likely to be found in more than one language combination. One
major difficulty was the fact that at the original conference all booths were
staffed by native speakers, whereas in the mock conference we had only
German mother-tongue students, who were expected to work from English
(or French) into German and vice versa - and were, in fact, eager to do so.

Recordings and transcription

Transcription turned out to be much more of a problem than I had ex­
pected, especially in terms of the procedures to be used for it.
Real problems began to crop up. Even a superficial transcription of the
German and English versions alone would be a full-time job even for some­
one familiar with the subject, let alone a student. It became clear that this
time-consuming effort had to be made by more than one person.
Yet, when different people set out to transcribe even the same recording
and along identical principles, there will still be considerable deviations.
Indeed, the same person will do the job quite differently when repeating it
at other times (Shlesinger 1992b).
After careful consideration of various transcription methods and criteria
described by different authors (Brown and Yule 1983; Chafe 1987; Halliday
1967), we decided that, rather than trying to develop one single method to
be applied to all transcribing, we would opt for different methods that were
to be a function of the research goal defined. A catalogue of rather general
guidelines was drawn up; these were to be followed by everybody, but
228 Teaching Translation and Interpreting 2

within this broad framework specific methods were adapted to individual

research questions. For instance, if the objective was to analyze intonational
phenomena, types of presentation were used other than those used where
compression or other macro-strategies were at issue. After all, it should be
borne in mind that the material to be studied is not the transcript but the
recording.

Definition of potential research questions

I wanted to identify strategies, but this may in itself represent quite a
problem. This, for instance, goes for anticipation as a strategy in the inter­
preting process. Firstly, the use of this strategy is only to be traced in cases
where the interpreter utters an anticipated hypothesis before the speaker has
produced it, or in cases where the anticipated hypothesis proves to be wrong,
which may trigger off other strategies such as relativation, outright cor­
rection, or may lead to an error. Secondly, anticipation can only be analyzed
when dual track recordings and sophisticated technical equipment for tran­
scription are available, so that one can gain a precise picture of the degree
of simultaneity between utterances of the source text producer and the
interpreter.
Interdependence of individual strategies is even more difficult to trace. In
the booth, one can hardly distinguish between comprehension strategies and
production strategies, as there is a considerable degree of interaction between
them, and the success or failure of one strategic operation will affect other
strategic choices to be made.
Other strategies worth analyzing, as findings may be relevant for teaching
simultaneous interpreting, include preparation strategies (expert knowledge
vs. interpreters' knowledge, manuscript marking etc.), text compression in
cases where source text is produced at an extremely fast rate, approximation
strategies, and situationally and culturally induced elaborations or deletions;
each of these strategies, however, has its own analytical problems which
must first be solved. And it is only after careful studies of an interpreter's
strategic choices to solve a given problem that it will be possible to prove
the high degree of interdependence between individual strategies.
There is another methodological problem in connection with a large corpus
such as the one described above. Should the recording of one person be ana­
lyzed longitudinally as it evolves, perhaps from a turbulent start to a phase
of smoother processing, when a certain degree of ease and reassurance is
reached, and eventually to processing under fatigue, in order to find out
something about differences in processing strategies under these conditions?
 229
Sylvia Kalina, Germany

Or should the rendering of the same part of text by different interpreters be

compared - which is only possible, or course, when using experimental
material, in this particular case our 'mock conference'?

Preliminary results
Some of the very preliminary results have led us to establish the following
hypotheses:

Professionals vs. students

There turned out to be a major difference between a superficial impression
of 'listenability' and detailed investigation. The professional English booth
was underrated when one merely listened to its performance, owing to se­
vere deficiencies in delivery. When it had been analyzed for errors, com­
pleteness, relevancy, etc., and compared with the performance of student
booths, we found that the professional performance was semantically much
more reliable. One conclusion that may be drawn from this finding is that
the reception of semantically reliable information produced by the interpreter
is adversely affected by the inadequacy of performance-geared strategies.
This underlines the importance of professional self-control and efforts to
brush up (e.g. by recording one's own performance from time to time, etc.).
This can also be stressed to student interpreters with a view to their future
careers when they will not be supervised by others but must rely on them­
selves. Moreover, it suggests that students should already be made aware of
the continued need for refinement of their performance throughout their pro­
fessional careers.

Strategic approaches
As mentioned above, a hypothesis resulting from our analyses was that
professionals tend to use strategies in ways differing from those used by stu­
dents. We found that the strategies of professionals seem to be at a higher
level. As regards monitoring, for example (a term which refers not only to
output control and repair operations but also to planning and aspects of sem­
antic equivalence, namely all components of an interpreting process), profes­
sionals face fewer interference problems, have a lower correction rate for
minor errors and a higher correction rate for significant errors. They are also
more user-oriented, as can be observed in increased cohesion or connectivity.
The professionals' attitude towards their own deficiencies seems to be dif­
ferent, too, with problems not successfully solved having fewer far-reaching
230 Teaching Translation and Interpreting 2

effects on other parts of the text than is the case in student performances.
Professionals seem to manage their capacity more efficiently (to borrow a
term from Gile 1991a) and sometimes even build in a higher degree of con­
nectivity, whereas students seem to try to break down their task into smaller
units.
Sentence splitting is a case in point. Contrary to our expectations, we
found that the professionals made less use of this strategy than the students,
and that professionals even tended to interconnect more utterances than
source text producers did. This finding certainly needs further investigation,
but, if it can be verified, the conclusion might be drawn that professionals'
total capacity is not as quickly exhausted by interpreting long interconnected
utterances as that of students.
Moreover, if the situation allows them to do so, professionals seem to be
able to facilitate the user's comprehension task by making a text more con­
nective and stating things more explicitly than the source text producer did.
This could lead to the hypothesis that sentence splitting is a strategy which
is useful for beginners but which interpreters tend to use less and less as
they become more experienced.

Intonational strategies
Shlesinger described the strange character of simultaneous interpreting in­
tonation (1992b). In our corpus, we found a significantly higher number of
nucleus (stressed) syllables even where intonation was not awkward. This
strategy seems to serve to give different segments of an utterance more
weight, facilitating the interpreters' own structuring of their output and facil­
itating listeners' comprehension. But interpreters may also stress more syl­
lables with the aim of clarifying and memorizing the semantic content and
information structure of a message for themselves, or they may simply have
to make up for deficient knowledge by increasing communicative content of
those utterances that they have clearly understood.
Monitoring strategies operate during the whole process, probably at all
levels. However, when capacity is exhausted, the monitoring strategy is the
one which tends to break down first.
As has been shown in a number of other studies, there are various types
of monitoring: (a) anticipatory monitoring which cannot be traced in the ut­
terances of the interpreter, (b) simultaneous monitoring, where corrections
may be made as or just after a word has been pronounced, and (c) retro­
spective monitoring, which may or may not entail repair operations of a
higher order (Kohn 1990; Hilverkus 1991). These include interference, where
 231
Sylvia Kalina, Germany

it is as yet unclear whether one can speak of an outright 'interference avoid­

ance strategy' (Riedmuller 1989; Hack 1992), as well as self correction.
Again, professionals tend to use such repair strategies only when grave er­
rors or planning faults are identified, and they tolerate quite a high rate of
minor semantic or grammatical imprecisions. Students, on the other hand,
seem to be over-concerned about even the slightest inadequacy, resulting in
an extremely high rate of repair operations, often not even successfully com­
pleted nor even an improvement on the first version, and leading to an ex­
cessive amount of attention being given to this process, so that inevitably
other components are neglected.

Open questions and implications for teaching

As the above discussion is far from conclusive, I cannot offer any 'con­
clusions' nor an accurate description of processes or valid statistical evalu­
ation, nor, indeed, a model for teaching. I have rather tried to describe some
problems encountered in empirical studies of interpreting processes and strat­
egies at work. I am aware that the experience gained in collecting and ana­
lyzing data is very personal, but I think that many of the problems I have
mentioned are of a more general nature and will be faced by any researcher
conducting empirical research in simultaneous interpreting. Among those
points which are of general interest and need further investigation, I would
emphasize the following:
(a) the representativity of corpora consisting of individual conferences,
interpreters, or experimental sessions;
(b) the accessibility of material for comparative studies;
(c) problems of transcription from recorded tapes, with the constraints en­
tailed by a typed and readable transcript;
(d) the persons expected to carry out such research work: students, profes­
sional interpreters, linguists, psychologists may all have useful contributions
to make.
Despite these open questions there are, I think, some useful lessons out­
lined above for the teaching of simultaneous interpreting:
1. Adequate teaching of simultaneous interpreting must be based on a the­
ory of the processes involved, and for developing such a theory, empirical
research is necessary.
2. Professional standards can be defined on the basis of such research, and
teaching will have to be geared to making students fully aware of what is
required from them and enabling them to achieve such standards.
232 Teaching Translation and Interpreting 2

3. The strategies that beginners learn will have to be continuously refined

and adjusted if their performance is to become truly professional.
4. Students must be motivated to continue checking the quality of their
performance even when they have finished their training and have achieved
a professional level.
5. Teaching must therefore not concentrate on ideal conditions and arti­
ficial settings alone, but must prepare students to cope with all the problems
and difficulties of professional interpreting.
 QUALITY ASSURANCE IN SIMULTANEOUS INTERPRETING

Franz Pochhacker, University of Vienna, Austria

Aims - Insights - Visions

The quality of the services rendered by professional interpreters has been
among the prime concerns of the international conference interpreting commun­
ity. Defining and measuring such quality, however, particularly in simultaneous
interpreting, has apparently been a lesser preoccupation. The Practical Guide for
Professional Interpreters published by the International Association of Confer­
ence Interpreters AIIC (1982) refers to "quality" as "that elusive something which
everyone recognises but no one can successfully define" (AIIC 1982: 1). In terms
of the subtitle of this volume, it is the "aim" of this article to pin down "that
elusive something", to try to get a grip on the notion of quality in simultaneous
interpreting.
I will review some of the "insights" gained by fellow researchers and hope to
add to these by reporting on my own research. While the approach to be outlined
here has been developed for the analysis of professional interpreting in an auth­
entic conference setting, it is equally relevant for the teaching of simultaneous
interpreting. Since the level of performance to be attained by graduates must be
in line with professional standards, it is essential to establish the quality criteria
by which candidates will be assessed in their final examinations.
In this attempt to establish criteria of quality in simultaneous interpreting I will
emphasize a much neglected component of quality assurance - the inspection of
the actual product. I am aware that inspection and analysis of interpreters' output
is only part of an overall scheme for quality assurance in simultaneous inter­
preting. But it is a necessary, indeed an indispensable step towards a "vision" of
quality assurance both in the relationship between interpreters and clients in pro­
fessional conference assignments and in the joint efforts of trainers and students
to improve performance in the interpreting classroom.

Views concerning quality

Over the past decade the issue of quality has clearly gained prominence in the
literature on simultaneous interpreting. In 1983, Cartellieri's reflections "On
Quality in Interpreting" essentially came down to the statement: "Much still
remains to be done to overcome the present unsatisfactory state of affairs in the
sphere of reliable quality parameters." (Cartellieri 1983: 213) Gile (1983) also de-
234 Teaching Translation and Interpreting 2

plored the lack of an objective and precise definition of quality and outlined a
methodological approach in which the interpreter's output ("presentation") was
to be judged by delegates with the help of questionnaires, while "informational
fidelity" was to be investigated by a comparative analysis of source and target
text recordings. This method was not applied to on-site quality assessment among
delegates until seven years later (Gile 1990c), but several empirical studies were
carried out to establish the relative weight of factors considered relevant to
quality judgments in simultaneous interpreting (Bühler 1986; Kurz 1989; Meak
1990). It is significant that the relative importance accorded to quality criteria
like native accent, pleasant voice, fluency of delivery, logical cohesion of utter­
ance, sense consistency with original message, completeness of interpretation,
correct grammatical usage and use of correct terminology (Btihler 1986: 234) was
found to vary not only between conference participants and representatives of the
interpreting profession (Kurz 1989) but also among different groups of users
(Kurz 1992). This variability of user expectations, which is also reflected in the
results of Gile's (1990c) direct quality assessment study, has recently been con­
firmed and elaborated by Kopczyński (1994).
There is thus a growing body of empirical evidence for (a) the principal factors
and criteria underlying judgments of quality in simultaneous interpreting and (b)
the relative variability of expectations among different groups of users. Yet these
data on various 'views on quality', on the inevitably subjective perception of
quality in simultaneous interpreting, do not address the issue of how to describe
and analyze the 'objective' reality on which such judgements are based. (Object­
ive in the sense that the interpreter's recorded product is open to replicable ana­
lysis and assessment.) Even though Gile's (1990c) study is concerned with deleg­
ates' subjective perception of quality rather than the comparison between that
perception and some clearly defined objective reality, it does hint at the potential
discrepancy between delegates' judgements and the actual features of the inter­
preter's recorded output. Elsewhere Gile (1991b) explicitly states that "in some
cases, the correlation between 'satisfactory quality' as perceived by a given com­
munication actor and the quality of fidelity, linguistic acceptability, clarity and/or
terminological accuracy of the Translator's output is weak, to say the least." (Gile
1991b: 193) The question then arises as to how we should best go about defining
and analyzing the text produced by the interpreter as an 'objective', that is, phy­
sical reality. What are the textualized parameters and variables underlying judge­
ments of quality in simultaneous interpreting, and how can they be measured and
quantified in a corpus of texts?
The process-orientation of most previous research on simultaneous interpreting
may explain why there are few answers to these questions in the literature con-
 Franz Pochhacker, Austria 235

cerning interpreting. The statement by Stenzl (1983: 47) that there are "practically
no systematic observations and descriptions of interpretation in practice" has lost
little of its validity over the past ten years. On the other hand, the criteria used
in the ranking studies by Buhler (1986) and Kurz (1989; 1992), as well as in the
above quotation from Gile (1991b), indicate that there is a general consensus
within the interpreting community on the quality standards for professional inter­
pretation. We seem to know what the product should be like, but we are less sure
about a method for establishing what a particular product is like in a given situ­
ation. Quite obviously, researchers, teachers and trainees need a method for
looking at the product.

Product-oriented research
A term like 'product inspection' may seem like a blatant misnomer in the field
of simultaneous interpreting, since, as a rule, the target text in interpreting is
nowhere to be seen. It is precisely this evanescence of the text {'verba volant')
that makes product-oriented research in simultaneous interpreting an improbable
undertaking. It should have become clear from the above, though, that such an
undertaking must be of prime concern to any researcher interested in knowing not
only what sort of quality users expect but also what sort of quality they actually
get. Such research would in turn provide the analytical tools and criteria for inter-
subjective quality assessment in the training of simultaneous interpreters.

Methodological problems
The methodological problems confronting the product-oriented researcher are
nothing short of daunting throughout all stages of research design and procedure -
from the recording and the methods of transcription and documentation to the
issues of analysis and evaluation. Four major questions might be asked to identify
some of the main problem areas: (1) How does one gain access to a text (cor­
pus)? (2) What should be done with the recordings? (3) What should one look
for in a textual corpus? and (4) How can parameters for quantitative description
be turned into values for qualitative analysis?
Question number (3) is crucial as it hinges on the underlying theoretical con­
ception and the hypothesis to be tested against the data. In any case, the search
for parameters in a corpus of spoken texts is likely to be hampered by "a lack of
analytical methods and techniques" (Stenzl 1983: 42). Thus, it should not be too
embarrassing here to suggest a list of textual features which may be relevant to
the description and analysis of an authentic textual corpus. The fact that the ana­
lytical scheme described below was developed on the basis of an authentic con­
ference corpus indicates that the difficulties referred to in question number (1),
236 Teaching Translation and Interpreting 2

such as professional secrecy and interpreters' unwillingness to submit their work

to analytic scrutiny, are not altogether insurmountable. The answers to question
number (2) are largely interlinked with question number (3): the way a textual
corpus is transcribed and documented will constrain the depth and scope of ana­
lysis, while the parameters to be considered in turn constrain the 'trans-
description'.
For the time being, I shall leave aside question number (4), which concerns the
link between quantitative parameters and qualitative evaluation, and will turn
from problems to solutions by outlining a model for descriptive text analysis in
simultaneous interpreting.

A multi-parameter model for description and analysis

Although my focus here is clearly, if not exclusively, on the text, my overall
approach must not be misconstrued as a narrowly linguistic one. I have previ­
ously stressed that texts in simultaneous interpreting must be placed within the
wider multi-level context of assignment type and situation (Pochhacker 1992a).
I have argued that it is only within a holistic conception of the conference as a
"hypertext" and the "situation" as a socio-psychological constellation of interact­
ing parties that one can safely narrow one's focus to the parameters of the pro­
duct or text as such.
In simultaneous interpreting the "text as such" is again a multi-parametric semi-
otic whole, which, in its full complexity, often defies description. In Pochhacker
(1994) I have suggested a text model with constituents in both the auditive and
the visual channel, on a "verbal-paraverbal-kinesic continuum" (Poyatos 1987).
From this constituent model of the "audio-visual text" in simultaneous interpret­
ing one can derive a number of textual features or parameters, such as slips and
structure shifts in verbal production, voiced hesitation markers, peculiarities of
voice quality and articulation, the use of (pictorial or verbal) visual information
(such as slides), as well as prosodic and/or paraverbal features. To these textual
constituents must be added temporal phenomena like speed, pausing, and rhythm­
ical pattern, which are often dominant in shaping the overall impression of a
spoken text. Finally, the modalities of text presentation (such as extemporaneous
speech, use of a manuscript, notes or slides) merit particular attention in the
documentation of (source and target) text corpora in simultaneous interpreting.
Thorough description and analysis of recordings from conferences with simul­
taneous interpreting thus require a combination of detailed transcription (includ­
ing false starts, slips in production, extended pauses, hesitation markers, and para-
linguistic elements like clearing one's throat, laughing, etc., as well as indications
of the delivery modality and temporal characteristics) and some sort of intersub-
 Franz Pochhacker, Austria 237

jective assessment of prosodic/paralinguistic features. A number of these delivery

parameters can be projected onto a parametric grid as suggested in Figure 1:

Figure 1: Text delivery profile (parametric grid)

It lies outside the scope of the present article to justify and explain, in suffi­
cient and convincing detail, why specific parameters were chosen for incorpor­
ation in the text delivery profile. The entire analytical scheme was developed on
the basis of a thorough theoretical and conceptual discussion laid down in
P5chhacker (1992b).
It is possible to assign numerical values (mostly between 1 and 5, from bottom
to top on the scale) to all parameters in Figure 1 and use the text delivery
profiles for a quantitative analysis of an entire conference corpus. The parameters
cover the following information:
ADR: The addressees of the text can be 1 or 2 persons, a Group of listeners,
the Plenary (everybody present) and/or, when the media are being
addressed, the General Public.
PRE: The degree of pre-planning of a text can range from extemporaneous to
read, with presented and preconceived texts as intermediate stages. One
can read or present a publishable manuscript (MS print) or a speech
written specifically for oral delivery (MS speech). A preconceived text
is one produced on the basis of - but without reading - a written text or
by drawing on written notes or merely a mental plan.
MED: Media use includes manual aids such as boards or flipcharts, pictorial
or verbal (overhead or photographic) slides and even video films (for
238 Teaching Translation and Interpreting 2

instance when surgical techniques are shown and commented on by a

speaker at a medical conference).
TEM: The presentation rate (tempo) of a speech assessed in syllables per
minute on the basis of three one-minute samples (first, middle and last
minute of a speech, adjusting for pauses of 1.5 seconds or more) can be
converted into syllables per second and marked on the central scale. (In
my corpus the average was 219.5 syllables/minute = 3.6 syllables/
second.)
MEL- The melodic, dynamic and rhythmic accentuation of a speech (as as-
DY- sessed by four bilingual professional voice specialists who listened to the
RHY: audio-recordings) can be marked on a five-point scale, where the middle
point (3) refers to the normal or baseline value and ratings below indica­
te insufficient and above exaggerated accentuation.
VOX: Voice quality can be normal (3), slightly abnormal (2), or markedly
abnormal (1).
ART: Articulation is similarly assessed as normal, slightly abnormal, or
markedly abnormal.
This description of the analytical tools, however sketchy, should suffice to in­
dicate the variety and complexity of the parameters that may be considered in a
product-oriented analysis of the texts in simultaneous interpreting. To illustrate
what 'a look at the product' actually looks like, Figure 2 (overleaf, page 239)
shows a sample of text 'trans-description' and analysis.
I am not really concerned here with the specifics of particular transcription con­
ventions or sets of parameters. Rather, I want to emphasize the fundamental prin­
ciple that product-oriented research in simultaneous interpreting means, to me at
least, a study of the target text as a complex and multi-faceted whole within a
communicative situation. This prospective, intratextual view of the target text is
a logical outgrowth of the functionalist principles of the General Theory of
Translation and Interpreting on which my specific conceptualizations and models
are based (P5chhacker 1992a). Emphasis is shifted to the interpreter's product as
a whole rather than the temporal and semantic correlations and correspondences
between source and target texts. The latter are central, for instance, to Lederer's
(1981) transcription method for the study of mental processes and have played
a dominant role in a number of diploma theses on error analysis in simultaneous
interpreting. The target-text oriented approach is a direct reflection of a func­
tionalist theory of text transfer which rejects the chronically ill-defined notion of
equivalence (Snell-Hornby 1988) on the level of words, phrases, sentences or
larger linguistic units in favour of the functional adequacy and coherence of the
(target) text-in-situation.
 Franz Pochhacker, Austria 239

Total duration: 7" 12

Tempo: 172 syll./min.

Pauses (>11/2 s): 1.2 s/min.
Hesitation (e): 5/min.

Slips - uncorrected: 4
Slips - corrected: 2
False starts: 9
Lexical blends: -
Syntactic blends: -
[ vMEMBERS OF THE PRESIDIUM, LADIES AND GENTLEMEN. AFTER THIS VERY
INTERESTING INTRODUCTION OF MISTER KIRCHHOFF I'D LIKE TO WELCOME YOU
ON BEHALF OF THE FEDERAL ECONOMIC CHAMBER. AUSTRIA'S ECONOMY IS
PROUD OF THE FACT THAT THE I C S B HAS CHOSEN THE AUSTRIAN CAPITAL AS A
VENUE FOR THIS YEAR'S CONGRESS. FOR THE FIRST TIME, 7 THE CONGRESS IS 5
TAKING PLACE IN VIENNA . . . OR RATHER IN EUROPE, AND THIS IS VERY
IMPORTANT FOR US FOR MORE THAN ONE REASON. OUR ECONOMIC STRUCTURE
CONSISTS OF A LARGE PART OF SMALLER AND MEDIUM-SIZED ENTERPRISES.
EIGHTY-FIVE PER CENT OF ALL AUSTRIAN COMPANIES HAVE LESS THAN TEN
1" EMPLOYEES, .. .. AND THIS CONGRESSV IS OF PARTICULAR IMPORTANCE FOR US, 10
BECAUSE OUR ECONOMIC POLICY IS ABOUT TO UNDERGO A CONSDRABLE CHANGE.
AUSTRIA .. IS FACING MEMBERSHIP IN THE EUROPEAN COMMUNITY AND
WISHES TO BE A FULL MEMBER AS SOON AS POSSIBLE. AND OF THIS COOPERATION
WE EX WE EXCEPT ALSO WE EXPECT v ALSO A STRENGHTENING OF THE SMALL AND
MEDIUM-SIZED BUSINESSES. W E HOPE THEY WILL FIN • OR THEY HAVE TO FIND 15
NEW NICHES, NEW MARKET NICHES, WHERE EXPERTS ARE IN DEMAND AND EXPERT
KNOWLEDGE. BUT ALSO, 3 OUR COUNTRY HAS A CERTAIN LINK3 3 TO THE NEW *
2" MARKET ECONOMIE • S IN THE COUNTRIESv EAST OF US. W E HAVE * CONSIDERABLE
EXPERIENCE* * IN* BUSINESS WITH THESE COUNTRIES, AND WE HOPE TO PASS ON
OUR EXPERIENCES TO OTHER COUNTRIES, TOO. N O W , FOR MANY SMALL AND 20
MEDIUM-SIZED COMPANIES THE WORLD OVER I SEE MANY POSSIBILITIES TO GET
INTO BUSINESS IN THESE* NEW 3 vECONOMIC SYSTEMS, PERHAPS ONLY 3 FOR A
SHORT TER TIME OR PERHAPS ALSO ON A LONG-TERM BASIS ONE THING IS
SURE, AND WE HAVE EXPERIENCED THIS IN THE PAST FEW MONTHS: IN THIS
3" PERIOD OF CHANGE3 3 THE DISADVANTAGES3 OF* THE OF THE v PAST* *M CREATE 25
PROBLEMS AND THE NEW SYSTEM IS HAS NOT REALLY * GAINED GROUND, AND
THIS LEADS TO DISCOURAGING ASPECTS, AND THAT'S THE WORST THAT CAN
HAPPEN TO AN ECONOMY THAT NEEDS OPTIMISM, SENSE OF IMAGINATION AND
CONSIDERING NEW IDEAS. IT IS CERTAINLY NO ACCIDENT THAT THE SMALL AND
v
MEDIUM-SIZED COMPANIES *M PROVE THEIR WORTH IN THIS PERIOD OF CHANGE, 30
THEY CAN RESPOND QUICKLY, THEY ARE FLEXIBLE AND CAN ADAPT QUICKLY TO
NEW SITUATIONS. O F COURSE, PARTNERSHIP WITH THE LARGE CORPORATIONS IS
VERY IMPORTANT FOR THESE* SMALL-SCALE INDUSTRIES. .. ZM OUR
4" ORGANIZATION, THE FEDERAL CHAMBER OF COMBERCE, REPRESENTS ALL 7 * THE

Fig. 2: Sample "trans-description" (from day 1 of ICSB Congress

(June 24-26, 1991, Vienna), German - > English.

Legend: Text profile: ■ = same as original; □ = value in target-text profile.

Transcript: | = mike on; v = time marker every 30 seconds), full minutes in left
margin; * = voiced hesitation ("uh"); • = hesitation; .. . = pause (one dot = 0.5
seconds).
240 Teaching Translation and Interpreting 2

From descriptive analysis towards evaluation

Now that we have the textual data before us, can we say anything about quality
standards or norms and thus relate the 'quality described' to the 'quality
perceived'?
On the most superficial level, the professional interpreter's output reproduced
in Figure 2 shows that the widely accepted standard (or "don't") that voiced
hesitation markers ("uh's"; transcribed as 3) are inadmissible (Bühler 1987: 16)
is respected only to a degree. It would be interesting to know whether this would
be reflected in delegates' assessment of the criterion of "fluency of delivery" used
in the ranking studies by Bühler (1986) and Kurz (1992). One should note, how­
ever, that fluency is actually a multi-parametric criterion and would have to be
studied by a combination of pause measurements, as included in the transcription,
and the assessment given for the scale of rhythmic accentuation (= RHY) in the
text profile.
With the exception of "pleasant voice", which can be linked to the assessment
on the voice quality scale (VOX), establishing a correlation between objective
and subjective text quality is even more complicated for the other quality criteria
associated with 'views on quality'. There is some hope that new methods in con-
trastive phonology may permit a sound analysis of "native accent" (Markus
1993), but no ready-made tools for larger corpora of texts are as yet available.
Similarly, establishing verifiable sets of parameters for "correct grammatical
usage" and "correct terminology" is much more complex than it might appear at
first sight. In the absence of a comprehensive normative grammar of spoken lan­
guage, linguistic output quality can be assessed only by using informants. Gile
(1985) has convincingly demonstrated the variable reliability of such judgments
in a group of ten native speakers of French. As regards "correct terminology", ex­
perts in terminology standardization would have to provide us with a canon of
"correct terminology" in the subject area of a particular conference, against which
interpreters' use of terminology could be checked. There have been attempts, co-
ordinated by Infoterm, Vienna, at providing advance and in-conference termino­
logy documentation for conference interpreters, and it remains to be seen whether
these initiatives will help us give an objective assessment of terminology use in
a product of simultaneous interpreting.
A truly fundamental problem arises for the criteria "sense consistency" and
"completeness". If we reject linguistic/semantic equivalence relations between
elements of the source and target texts as our analytical yardstick, the notions of
sense (cognitive content) and completeness (which is linked to variables like ex-
plicitness vs. implicitness and redundancy) become extremely cognitive. Since the
information to be transmitted by means of a text is constructed in the minds of
 Franz Pöchhacker, Austria 241

the listeners as a function of their general, specialized, personal and contextual

knowledge and competence as well as their expectations and communicative in­
tentions, it is beyond my analytical grasp to establish whether and to what extent
specific renditions "make sense" (Pöchhacker 1993).
To end this section on a more optimistic note, it seems plausible that the cri­
terion of "logical cohesion" is somewhat more amenable to analysis. It must be
borne in mind, however, that, in essence, the 'logic' of the rendition studied
stems from the producer of the source text. Verifying logical consistency will
therefore necessitate intertextual rather than intratextual product analysis.
As a result of the conceptual and methodological issues raised above, the
analytical scheme actually applied to my conference corpus is admittedly
'superficial', limited, that is, to features of the 'text surface', and covers only the
most obvious cases of logical contradiction and lack of coherence. The method
outlined is therefore only a modest first step towards product inspection in simul­
taneous interpreting. Nevertheless, it is in accordance with Cartellieri's suggestion
for resolving the question of quality: "The answer may be found in trying to
describe a number of features whose quantity may very well develop into quality
criteria." (Cartellieri 1983: 213)

Conclusion
The methodological approach described in this paper is undeniably fraught with
limitations. My scheme for text trans-description and analysis takes into account
some but by no means all conceivable parameters in the recorded text. In parti­
cular, it neglects the analysis of linguistic and paralinguistic variables which call
for the application of (text)linguistic methods and procedures. Clearly, my sug­
gestions for describing and analyzing the product in simultaneous interpreting can
only be a first step towards a method for detailed and rigorous product inspect­
ion. And even if such a method were achieved, our analytical study of the pro­
ducts would have to be complemented and correlated with delegates' (subjective)
assessment of the texts under study. The methodological difficulties of such on-
site evaluation are immense, since listening for errors of grammar and termino­
logy and listening for the substance of a given topic are altogether different
things. Direct delegate response is indispensable, however, if we want to establish
the thresholds at which a particular constellation and quantity of quality-related
features in the material text reaches 'critical mass' and leads to 'bad marks' for
the interpreting services.
Once empirical work on both delegates' in-conference quality assessment and
the analysis of the material text yields a broad enough body of findings, we will
have to take yet another series of steps and relate the 'objective' and subjective
242 Teaching Translation and Interpreting 2

quality of a text to the conditions under which it was produced. It is a well-

known fact that speakers' non-native articulation, speed of delivery, or use of
written documents not available to the interpreters, among regrettably many other
things, often leads to inevitable deficiencies in professional interpreters' output.
The text delivery profile presented in this article includes some of the relevant
parameters. Ultimately, then, documenting not only the text(s) but also the cir­
cumstances of meeting organization and text delivery will enable us to evaluate
not only 'quality' but 'quality under the circumstances'. This would be my vision
for comprehensive quality assurance in simultaneous interpreting - in support of
the interpreting community's endeavour to assure clients of the professional qual­
ity of interpreters' services and as a basic prerequisite for putting the training and
assessment of future interpreters on a sound conceptual and methodological foot­
ing. Admittedly, it will take some time and plenty of effort on the part of re­
searchers in the field of simultaneous interpreting to turn that vision into reality.

Højerup Kirke
SCREEN TRANSLATION
MARIENLYST
7ed
HELSINGOER
 RELEVANCE AS A FACTOR IN SUBTITLING REDUCTIONS

Irena Kovačič, University of Ljubljana, Ljubljana, Slovenia

Introduction
Special courses for subtitlers are still rare, a fact partly due to the widespread
belief that the special skills needed for a good subtitler can only be developed
through experience. The reasoning behind this attitude is that adaptations neces­
sary in subtitles are made intuitively and there is no global principle underlying
them.
In the present article I would like to argue against this view, using one par­
ticular type of subtitling adaptation, reductions, to demonstrate how apparently
random, intuitive and very different adaptations all have the same general 'raison
d'etre'. It is such general principles and the particular subtitling strategies which
result that should be part of subtitlers' training.
Reductions are a typical feature of subtitling. They are dictated by the extralin-
guistic requirements of the media: reduction depends not only on the speed of the
dialogue, but above all on the systemic similarities and differences between
source and target language. Subtitling into a language with similar syntactic pat­
terns and a similar average word length as those of the source language may call
for little reduction, but when the target language does not allow for condensing
patterns abundantly used in the source language, the importance of performing
reductions in the optimal way comes to the fore. Consequently, the principles
underlying reduction deserve a prominent place in the training of subtitlers work­
ing with such pairs of languages.
Subtitling English texts in Slovene is a typical example. The Slovene language
does not accept premodification of nouns by other nouns, it is very restrictive in
the use of non-finite, especially participial clauses, etc. According to some
counts, this makes Slovene translations of English texts 10 to 30 percent longer
than the originals. Consequently, subtitlers frequently have either to leave out the
structure in question or, if it is indispensable for the understanding of the story,
to sacrifice other parts of the text.
This leads to the fundamental question: what criteria (if any) does the subtitler
use in deciding what and how to reduce? A discussion of this point is pertinent
not only to practising subtitlers, but also to teachers and students of translation,
who should be conscious of problems in their chosen trade and ways of coping
with them.
246 Teaching Translation and Interpreting 2

In a previous attempt to analyse the processes underlying subtitling reductions,

I began by using a model of language functions and traditional structural cat­
egories of linguistic analyses (Kovačič 1992). As the work progressed, I found
myself increasingly expanding the model by introducing more general cognitive
notions. By the time the analysis was completed, I realised that many of the re­
ductions which appeared random if analysed by traditional linguistic analytic cat­
egories, fell into consistent patterns when viewed within a general relevance-the­
oretic framework. In the following discussion, I therefore also implicitly argue
that relevance theory can be used as part of the teaching of subtitling, both as an
explanatory procedure and as a set of notions providing subtitlers with some
guidelines in their work. The approach I have chosen here is a discussion of brief
examples which can also be used in class work.

Relevance theory
Relevance theory (Sperber and Wilson 1986) builds on the cognitive approach
to language as a mental activity operating within certain cognitive schemata,
claiming that we can communicate because we are capable of drawing inferences
from one another's behaviour. Inferences we make as hearers in a communication
arise from the tacit expectation that the speaker follows the principle of relev­
ance: "Every act of ostensive communication communicates the presumption of
its own optimal relevance." (Sperber and Wilson 1986: 158)
Relevance is always related to a given context (and context is not defined in
the usual linguistic terms, but rather as a psychological construct, as "a subset of
the hearer's assumptions about the world"; and the context of an utterance is "the
set of premises used in interpreting it" (Sperber and Wilson 1986: 15)). "An as­
sumption is relevant in a context if and only if it has some contextual effect in
that context." (122)
The degree of relevance is determined by the extent conditions'. "Extent con­
dition 1: an assumption is relevant in a context to the extent that its contextual
effects in this context are large. Extent condition 2: an assumption is relevant in
a context to the extent that the effort required to process it in this context is
small." (125)
There are three types of contextual effects: (a) addition of contextual impli­
cations, (b) strengthening of old assumptions, and (c) elimination of false con­
textual assumptions (108-16).
In lay terms we could say that the two extent conditions define relevance as a
cost-benefit notion: we want to achieve maximum benefit (the maximum context­
ual effect) at minimum cost (the hearer's minimum effort in processing a com­
municated assumption).
 Irena Kovacič, Slovenia 247

Relevance theory and translation

Gutt (1991) presents relevance theory as THE translation theory, claiming that
it offers theoretical foundations for "accounting for translation in terms of the
communicative competence assumed to be part of our minds" as "its domain is...
mental faculties rather than texts or processes of text production" (Gutt 1991: 20).
Gutt's starting point is the notion of interpretive resemblance, which Sperber
and Wilson defined as follows:
Two propositional forms P and Q (and by extension, two thoughts or utterances with P and Q
as their propositional forms) interpretively resemble one another in a context C to the extent
that they share their analytic and contextual implications in the context C." (Wilson and Sperber
1988: 138)
The translator's objective is - to put it simply - to make a translation inter­
pretively resemble the original text as much as possible, that is to provide similar
- if not identical - contextual effects. In order to understand the nature of trans­
lation it must be kept in mind that the notion of relevance is context-dependent;
in the same way, the translator's choice of a term or structure is dependent on
the contextual effects of an utterance in a given context (of a given person, cul­
ture etc.). "The success or failure of translations, like that of other instances of
ostensive-inferential communication, depends causally on consistency with the
principle of relevance. (Gutt 1991: 189)

Relevance and subtitling reductions

Reductions in subtitling are either partial ('condensations') or total ('deletions').
Total reductions are at least relatively easy to identify, if not always to analyse,
even within more traditional linguistic frameworks, as the basic feature is the
absence of a corresponding carrier of a meaning component in the surface struc­
ture. Conversely, partial reductions are - at the current state of the available in­
ventory for semantic decomposition of language items - virtually unanalysable.

Partial reduction in English-Slovene subtitles

Relevance theory provides useful insights into the strategies and factors behind
partial reduction. The change in example 1 can be explained by a deduction
string leading to an interpretive resemblance:
EXAMPLE 1
English source: I woke up THREE MINUTES AGO.
Slovene translation: PRAVKAR sem se zbudila. [I JUST woke.]
Students can be made aware of the deduction string "three minutes ago = short
time ago = just", through which the subtitler arrived at an interpretive resem­
blance that justifies the condensed translation. The point to note is that subtitlers
frequently use such deduction strings without being aware of the cognitive nature
248 Teaching Translation and Interpreting 2

of this procedure. However, there is a significant difference between applying it

intuitively as opposed to being aware of it, knowing when to use it and what this
use entails.
Similarly, it can be shown that such an analysis may apply even to language
elements with no underlying propositional form, for instance personal and inter­
personal elements or textual connectives:
EXAMPLE 2
/ SWEAR I'm not lying.
RES, ne la'em. [Indeed, I'm not lying.]
EXAMPLE 3
LET ME REPEAT THE QUESTION,
what are we going to do?
TOREJ, kaj bova storila?
[= SO, what are we going to do?]
In examples 2 and 3 the condensed translations may not be regarded as inter-
pretively resembling the original utterances in a semantic sense, but they do
preserve a functional resemblance: (2) in the sense of interpersonal emphasis,
persuasiveness; (3) in the sense of textual cohesion.
In relevance-theoretic terms, linguistic elements with no propositional equiv­
alents can be tackled by constructing an appropriate description, which then
makes manifest further assumptions. These assumptions then allow for an ex­
tended interpretive-resemblance analysis without a propositional form. Such re­
semblances are context-dependent: two utterances (or parts of utterances) may re­
semble each other interpretively in one context, but not in another. (Gutt 1991:
39-44) In example 3, "so" may be described as implying "I am going to refer to
something already mentioned or implied", and also with a rising intonation "and
I want to get your response". In other words, in all the above examples the orig­
inal texts contained explicatures (= explicitly communicated assumptions (Sperber
and Wilson 1986: 182)) while the Slovene translations contained implicatures (=
assumptions that are communicated, but not explicitly so (Sperber and Wilson
1986: 182)).

Total reductions in English-Slovene subtitles

In cases of total reductions ('deletions'), the overall statistics in Kovačič (1992)
show a significant difference between personal/interpersonal and textual elements
on the one hand and ideational (representative) ones on the other. But any further
subcategorisation of the three functional groups requires at least partial use of
cognitive (relevance-theoretic) notions. This may be illustrated by Example 4, an
excerpt from a Sherlock Holmes film, where relevance can account for the dif­
ferent treatment of vocatives, appellatives and similar linguistic elements.
 Irena Kovacič, Slovenia 249

EXAMPLE 4
[Holmes-bookseller:] (I have) just the books you need to fill up your bookcase, DOCTOR
(a). It looks untidy, does it not? ...
[Holmes-Holmes:] WATSON (b), do you mind if I smoke a cigarette in your consulting
room? A thousand apologies, MY DEAR WATSON (c), I had no idea
you would be so affected.
[Watson:] HOLMES (d). Is it really you?
The two vocatives functioning as simple terms of address (a, c) are left out;
similarly (d), which is used only to express surprise and therefore performs the
same function as the following "Is it really you?". The only vocative preserved
in translation is (b), which is highly relevant in the given situation: by using it
and by switching from the polite "doctor" to the familiar "Watson", Holmes re­
veals his identity.
If we look only at the syntactic pattern, grammatical categorisation and macro-
functional categorisation (Halliday 1985), they do not furnish us with a criterion
for differentiating (b) from a, c, and d. Relevance theory can help: (b) has signi­
ficant contextual effects, it changes Watson's assumptions about the other man's
identity and is for that reason indispensable.
As an example (d) is not as clear, but it can still be accounted for in terms of
relevance theory. "Holmes" and "Is it really you?" perform the same function:
they express the speaker's surprise. Such double constructions are very common
in speech, as they intensify the emotive effect. Consequently, it can be argued
that (if some reduction is necessary) the subtitler is faced with an open choice.
"Holmes" could have been retained and the other part of Watson's response de­
leted. It is the second condition determining the various degrees of relevance that
is decisive in this particular case. In terms of extent condition 1, both utterances
are equally relevant as they have comparable contextual effects. However, "Is it
really you?" is more transparent and requires less effort to be processed in the
intended meaning than "Holmes", which relies more heavily on the intonational
clues.
In structural descriptive terms, deleted elements range from individual words
to whole sentences and suprasentential elements (turns or adjacency pairs).
EXAMPLE 5
She stabbed him NINE TIMES.
EXAMPLE 6
I may come around tomorrow morning AROUND 10.
EXAMPLE 7
It was the destroyers WITH THE CONVOY
shooting at the submarine.
EXAMPLE 8
I managed to obtain his courteous permission
to write the note WHICH YOU AFTERWARDS RECEIVED.
250 Teaching Translation and Interpreting 2

The above examples show, respectively, deletion of circumstantials (Examples

5, 6) and of modifiers (7, 8). Decisions about deletions are context-dependent and
hinge on the translator's judgment about whether the viewer can have access to
the intended interpretation without them. In (5) and (6) the circumstantial details
given in large letters were left out because the modifications they entail are
unimportant in the story: what matters in (5) is that the woman stabbed
somebody to death, while the number of stabs only contributes to the dramatic
effect. The relevant piece of information in (6) is the promise to come around,
not the exact time, as this is not dealt with further.
It can be argued that these elements are deleted as their absence relieves the
viewers of some processing effort, without making the processing of the rest of
the utterance (and of the preceding and following utterances) more difficult. Con­
siderations of the processing effort involved in the interpretation of such reduced
utterances are very tricky. Although a considerable part of our speech is redun­
dant, that is of lesser relevance, it is indispensable to facilitate the processing.
When the subtitler leaves out a (partly) redundant element which is present in the
original text to facilitate the identification (proper interpretation) of another part
of the utterance, one must assess what this means for the necessary processing
effort. On the one hand, the target audience is spared the effort of processing the
missing part, but, on the other hand, they may find it more difficult to process
the remaining part. It is up to the subtitler to decide which of the two prevails.
Examples (5) and (6) are cases of deletions entailing less effort. With the modi­
fication omitted, (7) and (8) probably demand more effort to access the assump­
tions necessary for interpreting the headword nouns. The subtitlers obviously
counted on the viewers' ability to retrieve the necessary information from the
previously created cognitive context: Example 7 is from a letter written by a boy
about the first days of war. The convoy is mentioned previously in the letter, and
it is part of general knowledge that a convoy is a group of (war) ships or ve­
hicles; therefore the 'destroyers' can be assumed to belong to the convoy. Conse­
quently, the deletion of the modifying phrase does not significantly increase the
effort necessary to process "the destroyers", but it eliminates the effort needed to
process "with the convoy". A similar line of reasoning applies to (8), except that
here the deleted modification is based on the speaker's and the hearer's shared
private cognitive context rather than on a common cognitive context.
Whole utterances are deleted, especially when their relationship to a preceding
one is basically that of expansion or explication. The following English originals
(which have not been subtitled (yet)) will serve for illustration:
 Irena Kovacič, Slovenia 251

EXAMPLE 9
THE AIR BITES SHREWDLY; it is very cold.
EXAMPLE 10
- How's John?
- Better. HE THREW UP LAST NIGHT, BUT HIS FEVER BROKE TODAY.
In such cases it is important for the (future) subtitler to be able to perform an
efficient analysis of the passage. In terms of relevance theory, examples 9 and
10 can be viewed as instances of deletions of those sections that would demand
more effort to interpret. In (9) the everyday version is preserved and the meta­
phorical picture is omitted; in (10) the explanatory part is longer and has no im­
mediate importance for the story. Analysis of such examples (not in isolation, as
present here, but in context) should sensitize students to the factors underlying
such decisions.

Conclusion
I have presented a few examples of how the domain of subtitling reductions
can be provided with a satisfactory explanatory framework that will account for
cases as different as (1) and (10). It should be pointed out to students (and to
practitioners) of subtitling that awareness of the relevance principles in itself will
not make a perfect, infallible subtitler. As Gutt points out, the translator "does not
have direct access to the cognitive environment of his audience... - all he can
have is some assumptions or beliefs about it" (Gutt 1991: 112). Added to this,
the subtitler's own context may be insufficient or inadequate for him to make the
right decision. Nevertheless, the theoretical framework of relevance theory does
provide a valuable explanatory tool It is also general enough to form part of
training programmes for subtitlers (and translators in general). If nothing else, it
will help them to avoid the unpleasant feeling of not knowing what they are
really doing and what they are expected to be doing, and will thus heighten their
awareness of their trade.
 TRANSCULTURAL LANGUAGE TRANSFER:
SUBTITLING FROM A MINORITY LANGUAGE

Ian Roffe and David Thome, University of Wales, United Kingdom

Background
In November 1982 the Independent Broadcasting Authority launched a new
broadcast television channel in Wales. This new television channel, called Sianel
Pedwar Cymru (S4C), was established in response to social and political concerns
in Wales regarding the future use of the Welsh language. The channel was
charged with broadcasting a full and comprehensive Welsh language service in­
cluding a considerable part of its programmes at peak hours. However, within the
reception area of these transmissions, only a minority of the population was able
to speak the language and S4C was required to provide a service for the total
Welsh audience, not for Welsh speakers alone.
Interlinguistic subtitling appeared to be a means of bridging this linguistic and
cultural divide and led to the introduction of an in-house Welsh-English subtitling
service on a trial basis in 1986. This initial experiment in language transfer
proved successful and served to convince groups in the Welsh learners' commun­
ity of the value of the service. Nonetheless the scarcity of translation resources
imposed a practical limit to the expansion of the number of broadcast subtitling
hours.
Two events stimulated subtitling output. Firstly, in 1990, a new strategic di­
rection was formulated for S4C which re-focused much of its programme output
towards the principal centres of population rather than on the rural heartland.
This reorientation meant that the broadcasting channel was now targeting market
segments where the Welsh language abilities of people were not as strong; there
were also accompanying cultural differences in this group. A second stimulus
came from the passing into law of the 1990 Broadcasting Act, which stated that
the percentage of subtitled television programmes broadcast must reach a mini­
mum of 50% by the end of 1998 with further growth in subtitling to continue
after this date. In response to the Act, S4C pledged to increase its subtitled output
to 75% by 1998.
Along with these legislative and broadcasting changes there have been technical
developments and changes in the distribution pattern of types of television re­
ceivers in Wales. Ten years ago the percentage of households which were e-
quipped with a television receiver for Teletext transmissions, by which subtitles
254 Teaching Translation and Interpreting 2

are superimposed over ordinary programmes in the United Kingdom, was negli­
gible, whereas today over 40% of homes in Wales have such television receivers.
These factors led S4C to a new approach to subtitling in order to preserve high-
standard subtitling and increase interlinguistic subtitling on television. It was
envisaged that it would be necessary to widen the pool of subtitlers so that inde­
pendent television companies and translation agencies could provide economic
and competitive services for broadcast television.
With this in mind S4C approached our College. We already had activities in
common resulting from research and educational interests in translation and from
undergraduate course options in media studies. Our interests converged in the
challenge of teaching subtitling.
Initially, we surveyed British and American databases for previous research on
interlinguistic subtitling but, at first, nothing was discovered. Until recently, most
significant information for the providers of subtitling services originated from
studies of subtitling for the hard of hearing. A research group from Southampton
University (Baker et al 1984) had studied the most effective techniques for the
production and presentation of subtitles. This early work set the standards for
current subtitling in the United Kingdom and provided detailed guidance on the
production of multi-line subtitles and the editing of text to permit logical display
in the subtitles.
Kyle (1992), who investigated preferences for style, formats, accuracy and
speed, has examined the attitudes of deaf people to subtitles. He reports (p. 10)
that the "evolution of subtitling seems to have created a de facto standard which
would be difficult to change even if more effective means of providing pro­
gramme information were found". This standard represented a first step in the
quality control of subtitles for broadcast programmes.

Programme development
In the course design stage following the survey we identified the following four
key practical difficulties.
(i) Financing the initiative.
(ii) The introduction of practical subtitling expertise into the University.
(iii) The entry requirements demanded from participants.
(iv) Establishing an acceptable standard for subtitling in translation.
The cost of equipment is a significant barrier to the start of subtitling training.
Since access to subtitling equipment is a pre-requisite for practically based train­
ing, this barrier had to be overcome. S4C hired and loaned to us subtitling equip­
ment to operate two work stations. Our University in turn provided technical
support as well as a dedicated suite in a new and purpose built Media Centre.
 Ian Roffe and David Thome, United Kingdom 255

Subtitling expertise was designed for staff in order to develop in-house ex­
perience. This initial instructor-training was organised by S4C and provided by
themselves in association with a translation agency. This led to the creation of
a cadre of internal College staff who could design and deliver an accredited train­
ing programme.
The entry qualification demanded was bilingualism. There was already a small
pool of promising translators, and the central task of an intensive programme
would be teaching them new skills associated with subtitling. This helped to
focus the strategy of the course as a bridging programme for experienced trans­
lators to commercial broadcast standards of subtitling. This conscious decision
meant that the course assessment would be based solely on subtitling ability to
broadcast standard.
Our knowledge of the potential employment market made it clear that the train­
ing initiative would be most successful if the programme was flexible. The first
programme commenced in October 1990 with eight graduates in Welsh Language
and Literature as trainees; none of them had experience of the media industry and
they were all newcomers to subtitling. The trainees' background meant that fam­
iliarisation and the development of confidence in using the technology were
prime concerns. The programme had five distinctive stages.
1. Translation of script.
2. Preparation of a script.
3. Entry of subtitle text and display parameters.
4. Synchronisation to the programme videotape.
5. Review and editing.
The ability to work accurately at speed and under pressure of time was con­
sidered to be essential in a professional subtitler, so the coursework materials
prepared provided increasing challenges for the trainee.
The pilot programme involved all aspects of subtitling. Practical tuition was or­
ganised on a 1:2 staff/student ratio in a workshop format. Each student worked
on exercise tapes with weekly attainment targets which were closely monitored
for technical competence, translation and cultural interpretation. Every trainee
prepared a professional workbook to record the development of learning on the
course.
The trainees' work was assessed by an external examiner who viewed two 15
minute compilation tapes consisting of clips from a variety of programme genres.
These were presented after completing one third and two-thirds of the course re­
spectively. Final assessment was based on a compilation tape which included
both linguistic and cultural dilemmas in addition to a variety of programme ex­
cerpts. This tape had to be completed within a three hour period which meant
256 Teaching Translation and Interpreting 2

that the examiner was also judging the standard of delivery under pressure and
consequently under conditions simulating commercial practice. Trainees who
passed the standard were then awarded a Certificate in Subtitling, an academic
certificate approved by our University and S4C.

Commentary on evaluation
The original course was organised in three-hour sessions and delivered weekly
over a six month period which met the needs of trainees for part-time study. Sub­
sequent courses have been provided in a more intensive format of 10 week part-
time and 3 week full-time immersion courses. The latter programme was devised
for experienced translators operating in organisations within the United Kingdom
and internationally, who find a longer term commitment impossible.
The central issue in the training programme was whether the objectives would
be entirely practical or composed of a combination of practical and theoretical
training. In the event the latter option was chosen as this allowed a stronger focus
on translation and helped the scheduling of practical sessions. The course pro­
gramme has been adjusted according to our experiences. It was evident that
student familiarisation with the subtitling equipment was more time-consuming
than we had anticipated. This was mainly due to the fact that a part of the course
was taught in a group format: students had relatively few opportunities to receive
individual guidance and the tutor had little indication of progress made by indi­
vidual students during the initial stage of the course. This led to difficulties as
students struggled to meet deadlines.
There has, therefore, been a shift in teaching methods away from the group for­
mat in order to concentrate more on individual tuition. Immersion weekends give
students an accelerated introduction to the machines and to the technical aspects
of subtitling. During the 1992-1993 subtitling course the students received tutor­
ing on a one-to-one basis and were then expected to work for a minimum of five
hours per week on structured exercises that were to be presented for assessment.
These amendments have proved effective; all the students following the 30 week
course were technically competent within the first four weeks, a fact which
allowed them to concentrate on effective synchronisation and on the linguistic
and cultural content of captions.
The programme is a considerable investment in time and effort for all partici­
pants. All the students have acquired more confidence from applying their lin­
guistic skills in a new context and the experience of working in a technological
environment has also been of value, since students have usually only been ex­
posed to an academic environment. The need to respond to commercial pressures
has been more deeply appreciated by the course. And finally, the requirements
 Ian Roffe and David Thome, United Kingdom 257

of live subtitling is an excellent way of introducing team work.

The programme has fulfilled the original objectives. A resource centre with fa­
cilities and a core training team has been founded at our University. It has pro­
vided services which have been well received by organisations in the media in­
dustry in Wales and beyond. The standard of successful participants is at broad­
cast level as assessed by professional commercial subtitlers.
As a course team we often ponder whether we have found the correct match
for the expectations of the trainee group and/or of their potential employers and
of the programme content. Addressing the trainees' aspirations for employment
in media companies, we have in recent courses introduced a study tour for young
graduates including visits overseas to observe and digest the work of particular
companies and facilities houses. This is meant to broaden the students' perspect­
ive of the industry. Meeting colleagues who have successfully made the transition
into commercial employment has also heightened understanding of this career
path. Moreover it has served to establish personal contacts between trainees and
potential employers in an industry which relies heavily on such links.

Transfer problems
Another feature of the work intimately associated with linguistic and cultural
transfer problems and encountered by trainees and professional subtitlers alike is
interpreting the divide between Welsh and English and anticipating audience re­
action to broadcast captions. We will recount some of our practical experiences
in order to illustrate these difficulties.
(i). As a rule, song lyrics are not subtitled, but students following the course are asked to
subtitle a song, from a popular satirical programme, lampooning recent events in a soap opera.
This lyric was broadcast with subtitles when it was screened originally, since the rendition would
be meaningless to non-Welsh speakers without captions. The target audience was, of course, fam­
iliar both with the characters portrayed and the events described in the song, and the subtitler was
constrained by the content of the original Welsh version. The task was made even more difficult
because the words of the Welsh version were set to a well known English melody, and therefore
the subtitled version had to follow the pre-set rhyming and rhythmic patterns as well as conveying
the lyrical content of the original version in order to satisfy the audience.
(ii). One of the most difficult tasks for the trainees is to appreciate the perspectives and expect­
ations of the target audience. For example a fashion programme included references to the im­
perial monetary system which was replaced by the decimal system in 1971. The programme was
aimed at youngsters who consequently would not know this system and who would therefore not
understand the references made to the steep rise in the cost of living during the last fifty years.
In this case, the original subtitler decided to convert prices from old money into decimal pounds
and pence in order to ensure that the target audience understood the content of the programme.
(iii). On occasion subtitling is made very difficult by the inclusion of English words in Welsh
programmes. The audience will expect the subtitles to include these English words. The following
situation was presented in our current subtitling course. The Welsh lexical item to express the
English spider is 'pryf copyn', a similar form means literally 'chief cop', and this was then
punned as 'chief constable' in the programme. It is very difficult for a subtitler to interpret this
nuance. Most students abandoned the set script completely in their effort to include a natural ref-
258 Teaching Translation and Interpreting 2

erence to 'chief constable' in their captions.

(iv). Forms which are uncharacteristic or inappropriate must never be attributed to subjects. In
this respect the subtitler faces difficulties when subtitling a programme which contains taboo
forms. Students were set the task of subtitling a documentary in which the pig farmer used ex­
tremely earthy language! Whilst it is important that the audience appreciates the thrust of in­
formal register, written crudeness is always more offensive than actual oral usage. In this latter
case, the would be subtitler must aim to strike a difficult balance between familiarising the target
audience with the undoubted colour of a crude character, and offending it by excessive repro­
duction of dirty talk.

Visions
In Wales we continue to be confronted by a shortfall of competent interlinguis-
tic subtitlers. The difficulty is compounded by the problem of recruiting and re­
taining high calibre specialist staff. This arises from the highly competitive nature
of television contract work tendered for by independent television producers and
the consequential uncertainty of employment. Such opportunities compete
unfavourably with more traditional and stable career routes, in the field of educ­
ation for example, which appeal to the small number of single honours graduates
in Welsh who qualify from the University of Wales each year. (A total of 63 in
1992). So we intend to continue our interlinguistic courses for the foreseeable
future to meet commercial needs.
The subtitling machines have proved valuable tools in giving learning assist­
ance in order to gain another perspective on a language. Currently we are apply­
ing the techniques to Swedish and to German in our in-house undergraduate pro­
grammes and we have received very positive feedback from our students. The
machines are providing a form of Computer Assisted Language Learning and we
consider that there is considerable scope for extending the applications to perhaps
include a machine interactive form of learning.
Finally, the factors which make subtitling an attractive means of language trans­
fer between Welsh and English are present in other language combinations. For
speakers of European languages such as Irish, Catalan, Basque and Welsh, who
form a minority linguistic set within their native countries and who have
language capability in the predominant national language, there has historically
been relatively little language transfer. We have provided a programme for
minority language subtitlers. So far, subtitlers from Ireland, Scotland, Brittany,
the Basque country, Catalonia, Galicia, Portugal, Sweden and Germany have par­
ticipated in our courses of intensive three week instructor-training programmes,
so that the know-how can be cascaded through their organisations by way of
further programmes in the home country.
It is our experience that the development of interlinguistic subtitling expertise
must be seen as a continuous process which has the full support and confidence
 Ian Roffe and David Thome, United Kingdom 259

of regional broadcasting organisations - and not merely viewed as a one-off

course attendance activity. In this respect the set of conditions which existed at
the start of the project in the United Kingdom, namely: stable pre-course links
between TV broadcasting organisations and academia, and the close specification
of a solution which makes economic and financial sense to both partners, is piv­
otal in the successful extension of our programme as well as in others that are
similarly structured.
Nyborg Slot
 SUBTITLING: PEOPLE TRANSLATING PEOPLE

Henrik Gottlieb, University of Copenhagen, Denmark

For decades, subtitling has been the prevalent mode of screen translation in the
minor Western speech communities on the shores of the Atlantic: from Portugal
via Wales to Iceland, and from Belgium and the Netherlands to the Nordic
countries.
The dominant European nations, as well as the minor countries in Central
Europe, have maintained their habit of dubbing,1 in both cinema and television.
According to the European Broadcasting Union, this pattern will change in the
near future:2 the increasing exchange of films and television across the European
language barriers, will - coupled with a growing appetite for linguistic authentici­
ty amongst the European TV audience - lead to a notable increase in the need for
a shared European subtitling strategy, and for competent subtitlers.
To cater for this situation, the Copenhagen University Center for Translation
Studies is now planning a series of international courses in subtitling, to be held
at the new European Film College in Ebeltoft, Denmark.

The Ebeltoft courses

The Ebeltoft courses will be based on the Copenhagen University Postgraduate
Course in Audiovisual Translation, taught since 1991.3 This two-module course,
focusing on TV subtitling, is aimed at Open University students at BA level.
Module 1 includes two exams:
A) an oral examination in translation theory and critique relevant to audiovisual
translation, and
B) a practical 3-hour "prima vista" subtitling of a 5-minute sequence, produced
on a professional subtitling workstation.
Module 2 includes:
C) a 20-page paper on a specific topic within audiovisual translation, illustrated
by
D) a ready-for-broadcast subtitling of a 25-minute TV program or film se­
quence of the candidate's own choice.
At Ebeltoft, we offer three different courses:
1) General course. For media critics as well as filmmakers and TV producers.
This course aims at providing an understanding of the role of subtitling in the
increasingly multilingual media world of today, emphasizing the function of sub-
262 Teaching Translation and Interpreting 2

titling as an integral part of international film and TV production.

2) Special course. For interpreters and translators, translation teachers, and - last
but not least - dubbers and other screen translators from countries hitherto unfa­
miliar with subtitling. Here, the emphasis is put on the visual nature of subtitling,
seen as a mode of language transfer that combines elements of drama translation,
interpreting and literary translation. The aim is set at obtaining, or improving,
skills in interlingual subtitling, focusing on practical, esthetic, linguistic and
journalistic aspects alike.
3) Advanced course. For experienced translation scholars and subtitlers interested
in the technical and the theoretical state of the art. On the practical side, we will
exploit all editing and time-cueing facilities of the newest equipment on the mar­
ket. On the scholarly side, we will discuss recent international research in the
field of subtitling, covering the disciplines of translation studies, media studies
and psychology. The aims are:
a) Perfecting skills in interlingual subtitling, focusing on the optimal correspon­
dence between genre-specific features, viewers, and subtitles.
b) Providing a theoretical framework for discussing subtitling at a general level,
and developing critical judgment of existing subtitling practices.
The general and the special subtitling courses (courses 1 and 2), both lasting
5 days, will follow the same schedule, with a series of plenary sessions (in Eng­
lish), interspersed with practical 4-hour workshop sessions. In all these sessions,
participants work individually, learning the nuts and bolts of subtitling. In the
plenary sesssions, however, the emphasis differs: whereas course 1 focuses on the
presence of subtitles as sound-synchronous visual elements, course 2 concentrates
on the linguistic content of subtitles. Accordingly, course 1 participants - who are
not translators, anyway - are trained in subtitling from and into their own lan­
guage (intralingual subtitling), while course 2 participants work interlingually,
from a foreign language into their own language. The advanced subtitling course
(course 3) covers two more days, and one additional exam: a "ready-for-broad-
cast" interlingual subtitling. Thus, course 1 includes exam Bl: intralingual prima
vista subtitling - converting, for example, French dialog into French subtitles.
Course 2 includes exam B2, interlingual prima vista subtitling from English -
into, for example, German. Course 3 includes exams B3 and D, interlingual
prima vista subtitling - from, say, Spanish into Swedish - and interlingual ready-
for-broadcast subtitling - from, for example, Finnish into English.
In all courses, each participant works on his or her personal subtitling work­
station. Consequently, applicants ought to have some word-processing knowledge,
a fact I took for granted when we launched our Copenhagen course in subtitling
in 1991. However, a number of our middle-aged4 Open University students found
 Henrik Gottlieb, Denmark 263

it difficult to cope with the dedicated keyboards on our subtitling workstations.

It turned out that a few of these students did not even feel comfortable with a
normal typewriter. Even some of the more keyboard-familiar students surprised
me: until the beginning of the course, I never realized that you could insert a
good old-fashioned 51/4inch diskette into the disk drive in 8 different ways! Of
course, only one of these will work, and the right one soon became obvious to
everyone in class. But, apart from a modicum of computer-literacy, no special
technical qualifications are needed to sign up for the Ebeltoft general and special
courses in subtitling. The necessary knack for language I take for granted in
everyone interested in the discipline of subtitling.

Subtitling: constraints and virtues

In the rapidly expanding literature on subtitling,5 many authors - be they prac­
titioners or theorists - refrain from defining subtitling as a type of translation. A
typical example is found in one of the few books on screen translation published
so far, titled Overcoming language barriers in television (Luyken et al. 1991).
Throughout this work, the term Language Transfer is used, instead of simply
translation.
A decade ago, a Danmarks Radio subtitler bluntly stated:
We don't translate. We subtitle. To subtitle from a foreign language is to Danicize, with due
respect to the typographical constraints, the reading speed of the viewers, the composition of
the screen image, the cutting, and the speech tempo in the foreign original.
Now, if subtitling is not translation, then what is? The standard answer will be:
"A literary translation, of course." And indeed, this type of paper-to-paper trans­
lation has been known for at least two thousand years, yielding hundreds of
running meters of theoretical works on the subject.
But the concept of translation is now widening. In their recent work, Discourse
and the translator, Hatim and Mason (1990: 2) state that "the way is open to a
view of translating which is not restricted to a particular field ... but which can
include such diverse activities as film subtitling and dubbing, simultaneous inter­
preting, cartoon translating, abstracting and summarising, etc."
As even translation scholars grow used to the electronic media, subtitling and
other types of audiovisual translation are gaining access to theoretical works on
translation. With the exception of a few non-English works on translation (Cary
1956; Soderlund 1965; Mounin 1965, 1967; and Dollerup 1978), the pre-1980
concept of translation limited its scope to 'premeditated writing': translation of
dialog was accepted only when this was not genuine, and not heard. Only filtered
through an author could spoken discourse claim any interest.
But had the audiovisual media been the original means of mass communication,
antedating printed sources, the endeavor to translate abstract representations of
264 Teaching Translation and Interpreting 2

human communication, such as books, would have been inconceivable. The

audiovisual establishment might then, with some justice, have said the following
about the prospects of translating novels, for instance: "Trying to re-create such
fictitious, printed descriptions of alien cultures, with events often taking place in
the past and the text supported by neither visuals nor sound - a foolhardy enter­
prise - hardly deserves being called translating."
All text types present the translator with constraints: unillustrated fiction (also
known as literature) as well as comic books, feature films as well as TV serials.
Printed literature just happens to have a long history in our culture, so we have
become used to interpreting the signals that the author tries to communicate via
the one, fragile channel at his disposal: the printed word.
Compared to certain types of literature, audiovisual media provide a relatively
safe ground for translations, because of the constraints involved in those media.
In an earlier article (Gottlieb 1992) I have described the time and space
constraints specific to screen translation, a set of constraints easily identified by
scholarly observers, such as Titford 1982, Fawcett 1983, and Mayoral et al. 1988.
However, a number of admittedly trivial factors are at least as important in
determining the quality and nature of the end product, as are the more spectacular
media-specific constraints. As one of the above-mentioned authors aptly puts it:
Written translation suffers like dubbing and subtitling from physical constraints, including the
economic. The most important are: poor wages ...; absurd deadlines ...; poor originals ...; and
finally, poor training of translators. (Fawcett 1983: 189, column 2)7
This does not mean that the demands put on literary translation and on sub-
titling need to be the same: the best book translator is not necessarily the best
subtitler, but the best translation is reached only by acknowledging the premises
of the media in question. Knowing these premises, the quality of a translation is
defined by the talent of the translator, not by the constraints of the specific
media.

Equivalence: ideal for technical translation

A translation can never be a clone of the original. Structural differences alone
preclude a fully equivalent transfer of verbal content and intentions between
human languages:
Perfect translation is in the best of circumstances a virtual impossibility. Languages are not
ossified nomenclatures, parallel lexical lists from which one need merely choose matching items
on the basis of a one-to-one correspondence. (Shochat and Stam 1985: 42)
All human languages express nothing but their own culture: different languages
have different semantic fields and different usage-governed rules for collocation
and cohesion between elements. And not only do languages differ in terms of
what can be said; they also differ in terms of what is likely to be said in specific
 Henrik Gottlieb, Denmark 265

situations.
To modify the initial statement of the previous paragraph, certain written text
types which are culture-neutral, explicit and impersonal may indeed be translated
"equivalently". In technical texts, as for instance instruction manuals, neither
people, nor language, nor culture, is in focus. Such texts are purely informative,
and to a large extent predictable, which allows them to be translated mechani-
cally, either by computers, or by people, with or without computer assistance.
The aim is, generally, consistency in terminology and denotative precision. But
when we talk about text types dealing with human beings, their thoughts, their
behavior, and their interpersonal relations - their nature and culture - mechanical
transfer of the discourse involved is impossible.

Authenticity: the obligation of subtitling

As the realm of film and television is people and their world, TV programs and
films will have to be translated organically: by an interpreting person, in this
case a subtitler. Mechanical translation in the audiovisual media is - even without
considering media-specific obstacles - quite unfeasable. Hence, the notion of
equivalent translation is an illusory ideal for film and TV dialog.
In trying to get the message through to the target audience, across language and
culture barriers, a more realistic ideal would be achieving the same effect on the
audience as the one the original audience experienced; the same text they cannot
get.8 But perhaps the ultimate result a (screen) translator can opt for is simply
giving the target audience the experience they would have had if they already
knew the foreign language in question.9
Any translated text must function in the communicative situation around it. In
the case of monosemiotic texts, such as (unillustrated) books, the translator is in
control of the entire expression. In polysemiotic texts, the translator is constrained
by, and in some situations supported by, other communicative channels present.
Film and TV, being of a polysemiotic nature, force the translator to consider four
such simultaneous channels:
1) The verbal audio channel: dialog, background voices; sometimes lyrics
2) The non-verbal audio channel: music and sound effects
3) The verbal visual channel: captions and written signs in the image
4) The non-verbal visual channel: picture composition and flow
In dubbing, where foreign-language dialog is replaced by domestic-language
dialog, the balance of the individual film or TV program is maintained: The four
semiotic channels each hold the same semantic load as in the original version.
In subtitling, however, the balance is shifted from channel (1) to channel (3),
the latter normally the one with the lowest semantic content in original-language
266 Teaching Translation and Interpreting 2

films and TV. Although subtitling retains the original dialog, with the target audi­
ence thus enjoying the voice quality and intonation of the original, the authen­
ticity gained this way is partly lost when it comes to reconstructing the polysem­
antic puzzle. The reception work going on in the minds of the audience differs
considerably from the way in which the original was perceived, and this brings
us back to discussing the "equivalence of effect" notion.
How can a film that is partly read convey the same impression as the "same"
film listened to, with hardly any visual verbal signs? In trying to answer such
questions we clearly leave translation studies proper and enter the realm of psy­
chology. And in fact, psychologists have considered this issue. For nearly a dec­
ade, the Department of Experimental Psychology at the Catholic University of
Leuven, in Flemish-speaking Belgium, has conducted studies in viewer reception
of subtitles, interlingual as well as intralingual (see for instance d'Ydewalle et al.
1987 and 1989). However, most studies so far deal with the more behaviorist
side of the issue, notably subjects' eye movements when reading subtitles on the
screen. Genuinely cognitive, linguistically founded research is still very rare,
indeeed. With often conflicting results in the field of subtitle reception, establish­
ing a "scientific" ideal for polysemiotic texts, let alone testing this, is a risky
enterprise. One might simply conclude that "it is hardly fair to ask more of a
translation than we demand from monolingual communication, i.e. functional
adequacy." (Pedersen 1988: 15)

Ideals and facts of subtitling: the foundations of teaching

Unlike traditional literary translation critique, modern translatology does not
play the game of "spotting errors" in a particular translation. Although it is evi­
dent that not all possible renderings of a word, a phrase, or a text, are equally
suited, in non-technical texts, there is always more than one acceptable solution.
Discarding translatological fundamentalism, based on a black-and-white, absolute
attitude to language, a relative approach makes it possible to deal with the intri­
cate pragmatics of polysemiotic translation. The subject of study is what trans­
lators do, rather than what they should do. But this analytic, descriptive approach
on behalf of the translation scholar is not irreconcilable with a more prescriptive,
pedagogical stance. In teaching translation, the scholar-as-teacher may use the
'George Orwell principle': "Some renderings are more equal than others!"10
I believe that to fully understand the relativity of human communication, in this
case translation, one must know all the rules and roles of the game:
1) How to translate;
2) How to analyze and evaluate the translations of yourself and others;
3) Teaching others (1) and (2).
 Henrik Gottlieb, Denmark 267

In this way, creative, descriptive, and prescriptive activities go hand in hand.

In evaluating our students' achievements in practical subtitling, as tested in
exams B and D (see this article: p. 261), we look at the subtitles as they present
themselves on the screen. On the Copenhagen course, we apply professional
criteria, as we will in the Ebeltoft courses to come. Accordingly, every assign­
ment is evaluated in terms of
A) Understanding the spoken dialog in its audiovisual context: What has the
student heard and seen in the sequence subtitled?
B) Interpretation of extra-textual, genre-specific and historical qualities of the
program: What are the central themes, and what is the point of view behind
the dialog?
C) Authenticity and correctness in the rendering of the dialog: Is this what this
kind of person might have said in the source language - with due respect to
the standards of written language - and is the spelling all right?
D) Segmentation and layout of the subtitles: How filmically adequate is the con­
tinuing dialog 'cut up' into subtitle blocks, and how reader-friendly are these
blocks being structured?
E) Subtitle cueing: How elegantly are the individual subtitles presented on the
screen, and how well are the expected viewers able to perceive them? Do
they get enough reading time for each subtitle?
The teaching of screen translation thus involves the basic concerns of written
translation, and of good usage in general, but adds several dimensions, as ex­
pressed here under (D) and (E), and - to a certain extent - (A), since on our
courses we rarely use manuscripts. The students, like simultaneous interpreters,
must rely on their eyes and ears in order to receive the message correctly. In
practice, this has proven not to be all that difficult, whereas segmentation, layout,
and cueing - work procedures that demand a well-developed sense of timing, and
some flair for film esthetics - seem to be insurmountable obstacles to some.
The quest for authenticity, treated under (C), is something subtitling has in
common with other 'people-oriented' translations, as opposed to translations of
technical texts. However, as a consequence of a relative outlook on translation,
it must be admitted that sometimes even a 'human' translator may have good
reasons for choosing very source-text flavored solutions. These may be cases of
what Nida called formal equivalence, with the intention of transferring certain
lexical elements of the original, rather than merely the semantic and stylistic
content.11 When found in subtitles, the strategy of formal correspondence may be
consciously used, to emphasize culture-specific elements in the dialog, etc. But
quite often, the subtitler is simply lured by the original phrasing, especially when
working from a language related to the target language. Even more often, form-
268 Teaching Translation and Interpreting 2

ally correspondent renderings in subtitles are triggered by expected audience re­

actions: cinemagoers or TV and video viewers do not always want the "same"
impression as the original audience. The feedback-effect from the original -
whether that consist of recognizable words, prosodic features, gestures, or back­
ground visuals - may be so strong that a more idiomatic, 'functional' rendering
will be counterproductive. In many cases, people want a direct translation of what
is being said, not a complete reconstruction of the dialog in their own language.
This means that a consistently target-language oriented, 'idiomatic' translation
may backfire. This happens when the distance between the effect of the subtitle
and the total effect of the multi-channel original text - in the poly-semiotic sense
described above - becomes too great In such cases the friction between original
and subtitle causes noise, and the illusion of the translation as the alter ego of
the original is broken.
An example of this, taken from the TV genre political satire, will illustrate the
problems encountered when trying to domesticize the objects of ridicule:
When we subtitled the [British] series 'Not the Nine O'Clock News' we had this brilliant idea
of giving Danish names to the people commented on, for instance Jan Bonde Nielsen, in the
case of a big-time entrepeneur - or Allan Simonsen [a well-known Danish footballer in the
1980s]. They were used in the same context, and then we realized that the threshold of
credibility had been crossed. Well, we were not as brilliant as we thought. ("Der er altid nogen
der ved bedre", press release by Danmarks Radio 1989: 4. My translation)
Even in printed translations, a too markedly target-language oriented translation
may backfire: In his polemic against the present doctrine of idiomatic translation
aimed at rendering the meaning behind the words, Brian Mossop mentions the
problem of rendering people - not their names, this time, but their statements -
in other languages than the one they speak themselves. In his critique of a pos­
sible, idiomatically correct, translation of the resignation speech of a premier of
the French-speaking Quebec province, Mossop suggests a more 'rough', source-
language oriented translation, for this reason:
... most English Canadians would have heard the former Quebec premier being interviewed on
television and radio, and [the communicative] translation just does not sound like him. Unlike
former prime minister Pierre Trudeau, he was not known as someone who spoke English in the
same way as a native speaker, and even though this is writing, not speech, readers might
wonder whether the 'voice' they are 'hearing' is that of the same person they have heard on
TV. (Mossop 1989: 13)
Both examples show the problems that may occur when two or more semiotic
channels collide. In the Danish example, the references in the TV subtitles clash
with what can be seen on the screen, and what is heard, even by non-English
speaking Danes. In the Canadian example, the readers' knowledge of how the
original spoken discourse might sound runs contrary to the 'correct' written trans­
lation. Translating in a bilingual society may easily cause problems of authen­
ticity. Even though the target audience is not bilingual, strictly speaking, and thus
 Henrik Gottlieb, Denmark 269

in need of a translation, everybody in a speech community like Quebec knows

the peculiarities, especially the spoken features, of 'the other language'. A trans­
lation devoid of elements pointing to that language may present itself as sterile,
losing its credibility. The parallel to the English-Danish subtitling of TV satire
is evident. In any audiovisual translation from English in a country to a large ex­
tent (passively) bilingual, the same problem of authenticity enters the picture, lit­
erally speaking. In such cases, because of situational factors, the translation is
forced to retain some of the linguistic features or culture-specific references that
an ideal communicative rendering would have modified.

Subtitling as cross-cultural communication

Any translation is an adaptation of the original message to a culture outside the
original speech community:
According to the Sapir-Whorf hypothesis, language is culture, and those who 'inhabit' different
languages might be said to inhabit different worlds. (Shochat and Stam 1985: 36)
Neither film nor language can be transferred in total from one culture to an­
other. Attitudes and ideas, as well as tangible items of daily life, may be specific
to the original speech community. In some cases, the constraints of subtitling
make things even more complicated:
A drama or a film deriving from another culture will be in part based on assumptions and
concepts which may not exist in that form outside that culture, and which cannot be adequately
summed up in a four second subtitle. (Manzoufas 1982: 18)
Culture-specific elements in a film or TV dialog need not be extra-lingual, as
exemplified above. Intra-lingual features peculiar to the source language may be
just as difficult to tackle for the subtitler. For any translator, a crucial question
will always be: "Should I bring the source culture to my audience, or vice
versa?" In answering this, intra- and extra-linguistic features may have to treated
differently:
Only references to the surrounding culture should have local colour. Linguistic formulations
should preferably be exclusively target language wordings, no matter what they refer to.
(Nedergaard-Larsen 1993: 235)

Types of screen translation: the time factor as distinctive feature

Earlier in this paper, I classified subtitling, or rather, the subtitled film or TV
program, as a polysemiotic text type. In the following, I will list a number of
such multi-channel types juxtaposed by some central monosemiotic text types.
Within polysemiotic translation, the relevant distinctive feature is time. This
term covers two phenomena: time of text production, and time of text presenta-
tion to the target-language audience. In this context, 'time' is seen as a point in
the continuum from the past to the present. For the notion of continuous time
(time as a line rather than a point) I will reserve the term duration.
270 Teaching Translation and Interpreting 2

Any translation type is defined by two factors: (A) time and (B) semiotic
composition.
A) Three points in time will suffice:
T1, the time for production of the original verbal element
T2, the time for presentation of the original verbal element
T3, the time for presentation of the translated verbal element
If Tl precedes T2, and T3 is simultaneous with T2, the translation is synchro-
nous. If Tl precedes T2, and the original verbal channel is not perceived by the
target audience, the translation is non-synchronous. Finally, if Tl is simultaneous
with T2, and T2 precedes T3, the translation is delayed.
In other words, whereas at a word-to-word level non-synchronous translation
is not concerned with synchrony between original and translation - as in the case
of book translations - delayed types of translation, such as the so-called 'simul­
taneous' interpreting, can be seen as less lucky varieties of synchronous trans­
lation. Here, as with the synchronous types, simultaneity is relevant.
As opposed to non-synchronous translations, where the receptor - reading a
book, for instance - controls both time and duration for reception, synchronous
and delayed translations are both immediate. Here, the translated product - say,
a TV film - defines the time and duration available to the receptor.
B) The basic distinctions of semiotic composition are:
7) Monosemiotic text types, with only one channel of communication, versus
polysemiotic types, with two or more channels.
2) Isosemiotic text types, where the translation communicates via the same
channel - or set of channels - as the original, versus
3) diasemiotic text types, with different channel(s) used.
Coupling the tripartite time-defined distinction with the two-times-two distinction
referring to semiotic composition, a total taxonomy of translation types can be
established. In the table on the opposite page, I have included the intralingual
simultaneous subtitling, used in news broadcasts for the deaf and hard of hearing,
and non-electronic polysemiotic types as (translated) comic books.
As is clearly seen, subtitling differs from other types of verbal transmission by
virtue of its additive nature. In adding written text to speech, subtitling earns its
diasemiotic status. Unlike subtitling, the three isosemiotic types of screen
translation listed above all work with voice replacement, or - to use a more
common term - revoking}2 In a TV program with voice-over, the original dialog
is still partly audible, 'drowned' by the target-language speak. In Russia, this is
still the common way to transmit foreign TV films, whereas lip-synchronous
 Henrik Gottlieb, Denmark 271

dubbing is dominant in the German, French, Italian, and Spanish speech com­
munities. The method of commentary deletes the original speak and replaces this
off-screen narration with target-language narration.

TYPOLOGY OF TRANSLATION

Semiotic composition Time-defined categorization

Synchronous Delayed Non-synchronous

Mono- & isosemiotic
Speech Radio

interpreting

Writing — Written translation

Mono- & diasemiotic

Speech Book translation
on audiotape
Writing - Interpreting Minutes
for the deaf (from a meeting)
Poly- & isosemiotic
Writing + Image Translation of
comic books and
advertisements

Speech + Image ___ Simultaneous -—

interpreting

Speech + Image Dubbing TV voice-over TV commentary;

+ Music & Effects Performance of
translated drama
Poly- & diasemiotic
Speech + Image Subtitling Simultaneous
+ Music & Effects subtitling
+ Writing
272 Teaching Translation and Interpreting 2

Interlingual translation: a comparison of types

In the following table, interlingual subtitling and four other major types of
interlingual communication are juxtaposed, and compared with reference to ten
communicative parameters. The types are placed in the table according to degree
of 'naturalness', ranging from the abstract monosemiotic printed translation in
the far left column, to the more simple polysemiotic consecutive interpreting.

CHARACTERISTICS OF SELECTED TYPES OF TRANSMISSION

Media: BOOK PLAY FILM & TV ENCOUNTER

Type: Printed Acted Sub- Dub- Consecutive

transl. drama titling bing interpreting
Parameters:
Polysemiotic No Yes Yes Yes Yes

Isosemiotic Yes Yes No Yes Yes

Impromptu No No No No Yes

Immediate No Yes Yes Yes Yes

Spoken original No Yes Yes Yes Yes

Spoken translation No Yes No Yes Yes

Condensed translation No No Yes No Yes

Acting translator Yes No Yes No Yes

Known receptor No No No No Yes

Open for two-way No (No) No No Yes

communication

All ten qualities are typical features of natural communication,13 while the
counterparts of these represent what could be called symbolic communication.
Following from this, the 'naturalness' of the different types of transmission can
be expressed on a scale from 10 to 1, depending on the number of 'yes'answers:
 Henrik Gottlieb, Denmark 273

Interpreting 10 "points"
Dubbing 5
Subtitling 5
Drama translation 5
Printed translation 2

Reduction: the evil spirit of subtitling?

Throughout the history of subtitling, the necessity of dialog condensation has
always been stressed. Many scholars and critics lament this seemingly inescap­
able fact,14 as does this editor of a Canadian film journal:
... scripts must be heartlessly abridged to be accommodated on a screen.
(Moskowitz 1979: column 1)
But the late directeur fondateur of UNESCO's International Journal of Transla­
tion, Babel, saw the demand for reduction in subtitling as a gift:
En condensant des phrases on s'aperçoit qu'on peut presque tout dire en si peu de mots que
l'exercise du langage parait une fonction humaine pour ainsi dire superflue. (Caillé 1960: 109)
An experienced subtitler expresses a more matter-of-fact attitude:
The shortening of the text for subtitling purposes is nothing more than deciding what is padding
and what is vital information. (Reid 1987: 28, column 3)
Not only does the immediate nature of subtitling - controlled by speech tempo
and cutting - press for brevity. Two other, less obvious, factors are at work:
1) Intersemiotic redundancy, which enables the viewer to supplement the semiotic
content of the subtitles with information from other audiovisual channels -
notably the image, and prosodic features in the dialog
2) Intrasemiotic redundancy in the dialog. Especially with spontaneous speech,
not only the informative content, but also the verbal style and characterization of
the speaker are better served with some reduction in the subtitles.15 However,
even deliberate speech, including script-based narration, often contains so much
redundancy that a slight condensation may enhance the effectiveness of the in­
tended message. Take this excellent statement by Rafael Nir (1984: 85):
It seems indisputable that the translation of dialogue in films belongs to the category of trans-
lations which can be defined as highly contextualized. [148 characters]
No great harm is done by trimming this heavyweight into the following 'subtitle':
The translation of dialogue in films
is highly contextualized. [61 characters]
Many scholars would do their readers, lay and learned alike, a favor by 'subtit­
ling' themselves. Or, to condense myself: "Scholars in the world, be brief!"

Notes
1. In dubbing, the original dialog is replaced by postsynchronized target-language dialog.
2. By the late eighties, the EBU Review Editor-in-Chief was already admitting that "if we
absolutely must discern a trend, let us say that subtitling is gaining ground". (Derasse 1987: 10).
274 Teaching Translation and Interpreting 2

Or, citing a more recent source: "As educational standards have, and are likely to continue to rise
over time in Europe, subtitled audiovisual versions of foreign language productions might
therefore become increasingly more acceptable to wider parts of the peoples in Europe." (Luyken
etal. 1991: 187)
3. As of August 1993, this course has produced 21 candidates. A more detailed description of the
course, as well as a theoretical presentation of the concept of subtitling, is found in Gottlieb 1992.
4. Looking at the achievements of my first two subtitling classes, containing a total of 58 stu­
dents, the only statistically significant factor guiding good results in subtitling is 'age'. Younger
Open University students, aged 25-39, fare better than their would-be colleagues in their forties
or fifties. In traditional practical translation, age is often considered a plus, but in the case of
subtitling neither previous (literary) translating experience nor academic title(s) are safe indicators
of success.
5. By the end of 1993, I had 544 titles on interlingual subtitling - most of them dating less than
five years back - listed in my bibliography (see Gottlieb, forthcoming). A large number of these
items were unpublished university papers, in-house material produced by TV subtitling companies,
etc.
6. My translation of a statement by Danish TV subtitler Jesper Kjaer in a letter to the editor
(Politiken, April 1984), answering a critique by a TV reviewer of the "translation" of the
American comedy series M.A.S.H.
7. In Scandinavia, interpreters and technical translators are far better paid than literary translators,
with subtitlers occupying a middle position. In terms of deadlines and workload, subtitlers - with
their efficient dedicated equipment - often fare better than literary translators.
8. This ideal, sometimes called equivalence of effect, is in keeping with Nida's concept of func-
tional equivalence (see Nida and de Waard 1986).
9. This no-nonsense attitude, in spite of its irrefutable logic, is rarely encountered in scholarly lit­
erature. However, it was expressed in a recent Danish undergraduate paper (see Lind and Sestoft
1992).
10. In the article Subtitling: Diagonal Translation, (Gottlieb 1994) I suggest nine 'pedagogical
pillars' to consider when creating and evaluating subtitles. Partly prescriptive, partly descriptive,
this article complements the present paper.
11. See Nida 1964: 159. This strategy is often used involuntarily: in his article The traps of
formal correspondence, Sándor Albert (1993) illustrates how any translation not considering the
text as a whole runs the risk of pragmatic defectiveness.
12. To my knowledge, the invention of these collective terms is rather recent: the term 'revoicing'
was introduced by the European Institute for the Media (see Luyken et al. 1991), whereas 'voice
replacement' is found in Watching your language: Foreign version issues in Screen Digest, July
1992.
13. 'Natural' in the sense 'phylogenetically and ontogenetically original': Before human beings
began expressing themselves through written signs, people mastered spoken discourse, and before
children are taught how to read, they are able to talk.
14. Recent experiments at Leuven University, Belgium, (d'Ydewalle et al. 1989) showed that
young and old subjects spent "an average of 55.4% of the presentation time on the normal two-
line subtitles". That is to say that people seem to have almost half of their television viewing time
left for watching the action on the screen. If this is generally true, it raises the possibility of
speeding up normal subtitling, abandoning the present European standard of 6 seconds per two-
liner. This would eventually lead to a fuller rendering of up-tempo dialog (see Gottlieb 1992: 164-
165).
15. The transition from spoken dialog to written subtitles is treated in Gottlieb 1994, where a
number of redundant features of spoken discourse are listed.
 AUDIO-VISUAL COMMUNICATION: TYPOLOGICAL DETOUR

Yves Gambier, University of Turku, Finland

In this article, I shall focus mainly on subtitling. Subtitling is a force to be

reckoned with in subtitling countries, like Portugal, Greece, Wales, the Nether­
lands, the Nordic Countries. For instance, in Finland some 3,000 foreign TV pro­
grammes are shown per year, up to seven per evening. This would correspond
to reading approximately 200 novels of 300 pages each, a total of 60,000 pages
or a complete novel every other day. We are not only watching but also reading
TV! There is, however, more to it than that.
In the modern world there is a multiplicity of audiovisual messages: documentary
films, short-length films, cinema films, TV broadcasts, children's programmes,
radio interviews, business videos and home videos. Cinema, video and TV do not
call for precisely the same type of subtitling, for reasons which fall under two
headings.
There are technical and economical reasons:
* they are imported by different firms
* they require different working conditions
* the speed of the picture is different
* the definition of the screen is different
There are linguistic reasons: in Finland, for instance, bilingual subtitling (into
Swedish and Finnish) is used in the cinema and monolingual subtitling into Finnish
alone on TV.
Up till now, research has mainly been concerned with the subtitling and dubbing
of fiction films. In view of the enormous variety of audiovisual communication,
this may seem somewhat surprising; but it reflects the prevailing orientation in
translation theory, which is still strongly dominated by literary translation.
In this article, I shall discuss three points: the types of multilingual transfer in
audiovisual communication, two effects of subtitling, and finally some challenges
to translator training.

Types of multilingual transfer in audiovisual communication

We are not concerned here with intralingual subtitling (such as that provided
for deaf or hard-of-hearing people). The tentative typology which follows is a first
stage going beyond the current prescriptive standpoint.
There are various means of conveying the linguistic meaning of a foreign-Ian-
276 Teaching Translation and Interpreting 2

guage message to a target audience:

1. subtitling: of films, commercials, TV series and programmes
2. subtitling in real time or simultaneous subtitling, for instance live performance
interviews
3. dubbing (lip-synchronisation)
4. interpreting, in four possible modes:
- consecutive, often shortened interpreted renditions, on the radio (interview
with a singer or politician, telephone calls), on TV (cultural debates such as
'Apostrophes' in France);
- prerecorded consecutive interpreting, close to voice-over;
- consecutive interpreting in duplex, long-distance communication, for instance
during teleconferences
5. voice-over: simultaneous interpreting, characterised by the faithful translation
of original speech and approximately synchronous delivery. This means was ex­
tensively used during the Gulf War (1990-1991): it allows scoops and news
flashes in different languages, mixing orality, audibility and a more or less high
degree of information.
In cases (4) and (5), the original voice/sound is either heard, fading away slowly,
retained at a low level or reduced entirely.
6. Narration. The difference between (5) and (6) is linguistic. With narration,
the original speech is prepared, translated and possibly condensed in advance
and is then read by a journalist or an actor, while voice-over is applied mainly
to spontaneous speeches. In both cases, delivery remains synchronous, especially
if the narrator appears on the screen. The verbal sense/content works together
with the visual information being presented.
7. Commentary is a way of adapting a programme to a new target audience -
not literally duplicating the original speech but adding new information, like
creating a new work, with the identity of the commentator (actor) distinct from
that of any of the programme's participants. The synchronisation is with on-
screen images rather than with the original voices.
(6) and (7) are used with children's programmes, documentary films, business
videos, slide shows, industrial short-length films; they fall in between translation
and interpreting, because of the condensation, the reshaping of the original, and
the oral output. Modes (3) to (7) constitute different types of revoking, spanning
from quite free expression to renditions involving extremely tight technical and
linguistic constraints to quitefreeexpression.
8. Revoking or multilingual broadcasting: the receiver selects a sound track with
an appropriate language.
9. 'Surtitles' or 'supratitles': for instance in opera houses or theatres. These are
 Yves Gambier, Finland 277

normally presented on a line-screen using digital print.

10. Simultaneous translation is a type of sight translation from a script or a sub­
title in a foreign language taken from a written source text; thus the term 'simul­
taneous translation'. It is used during film festivals and in film libraries (cinema­
thèques). If no script is given in advance, the work becomes genuine simulta­
neous interpreting (or voice-over).
The ten modes confuse and obscure the usual borders between translation and
interpreting on the one hand, and between written and oral codes on the other.
Subtitling, for instance, is a kind of written simultaneous interpreting; voice-over
is close to drama translation and resembles 'oral subtitling'; dubbing resembles
interpreting because of the oral synchronisation with the original. Subtitling (from
the oral to the written form) seems to be the converse of drama translation (written
cues to be acted out) but both work within a split communication between the
original, actors and an audience. Could there be analogies between sight translation
(written/oral) and subtitling (oral/written), especially in the neurolinguistic schema­
ta or processes?
In short, the audio-visual translation in the media is a new genre, still largely
unexplored in the field of translation studies.
Even subtitling is not an homogeneous activity; one may work with or without
a script, with or without a planned schedule, with or without the pictures. The
working conditions are quite different. Subtitling with the script entails different
translating strategies than working without the script.
What are the factors determining the choice of a method of audio-visual language
transfer made ? Even though it is easy to see that specific methods are preferred
according to the type of programme (drama, cartoon, educational or entertainment
programme, children's film, science or art programme) and target audience, it is
more difficult to determine how TV broadcasting companies and film importers
have made decisions in favour of one or the other method in the past, and the role
which linguistic norms and conventions have played in these decisions.

What are the effects of subtitling?

Oral vs. written discourse
The traditional approach to translation and interpreting has often relied on the
following conceptions:
- the original product is uttered by a single 'speaker'
- the output is seen as a homogeneous, stable, and finished product
- the translation aims at an interlinguistically 'equivalent' transfer of the
'original'.
In the case of media (TV, cinema) using different semiotic systems, it is better
278 Teaching Translation and Interpreting 2

to speak of an 'adjustment', resulting from both the complexity of the media (text
and image), and from the immediate conditions of reception and interpretation
of the message.
With a film, in fact, we must take into account at least four phases, each produc­
ing a meaning in collaboration between interlocutors:
- there is a transformation of a linear text (fiction or non-fiction) into a scenario
(or the reverse, which is less common)
- there is a switching/transformation/conversion from a scenario or a script to
a dialogue articulated by the actors (semiotic translation, which also involves
interlingual translation if the language of the script and that of the actors is not
the same)
- iconic and linguistic features are related to one another (another kind of
semiotic translation)
- finally, there is an interlingual transfer of dialogue in subtitles, a process which
can be regarded as semiotic and interlingual since it involves a) a change of code
- the oral code is converted into a written code - and b) a change of language.
It should be stressed in this context that this double mutation is not confined
to the cinema and TV alone. To mention just a few examples: at numerous inter­
national conferences the debates are transcribed and stenographed and then reform­
ulated in another language, while operas may be 'surtitled'.
In the case of subtitling, the original message is delivered by various enunciators
with different voices and different personalities. But what is more, in passing from
the script writer to the director and the actors, the message has undergone changes
and transformations. Before reaching the spectator's eye, the message has thus
in fact already been changed and adjusted several times. These changes are deter­
mined by different factors.
This being the case, it is too limited to view subtitling as a mere 'condensation'
of a so-called 'original'. To regard subtitling as reducing the number of words
is not applicable as a method of comparative analysis of intercultural communica­
tion. Of course, it is naturally easy to perceive and count the number of omissions,
deletions and substitutions (for instance, of phrases of politeness, exclamations
and interjections). But to do so is also to be under the spell of a quantitative, math­
ematical theory of information (information entropy) which considers cross-lin­
guistic communication in terms of losses or additions and sees translation as a
process of mimetically copying a literary work, a duty to repeat. In subtitling, it
is important to study what is transformed and why. Reduction may be one of its
components; it is not, however, a property of subtitling alone. It is also characteris­
tic, for example, of interpreting and cartoons. What is unique in subtitling as a
form of selective translation is the fact that it operates at two levels simulta-
 Yves Gambler, Finland 279

neously: (1) the change in code from the temporally organized oral code to the
linear written code and (2) the switch from one language to another.
In writing, there is normally at least some delay between production and recep­
tion, and thus no immediate feedback. Furthermore, no prosodic or pragmatic
means are available. In the cinema, the recoverability of the written code is
weakened. Under normal conditions, the spectator does not have the opportunity
of going back in order to reinterpret the subtitling, and in this respect it is close
to the oral code. But whereas the multiform, co-produced oral message is intimate­
ly intertwined with the image, the interpretation of the written message (subtitling)
creates a certain delay and distances the moment of interpretation and that of the
appearance of the original complete message. The oral message is fused with vari­
ous semiotic systems and thus activates the visual and auditive sensitivity of the
spectator. In writing, the utterance is detached from other sense-making systems
which are separated from one another, since the spectator's attention is now caught
mainly monosensorially (the eye) and in one direction (from left to right). In addi­
tion, the written code is valued differently from the oral code in our culture. The
anthropology of writing has shown the extent to which power is based on writing,
and the amount of power the written word possesses. In fact, when a written dis­
course is processed, it fascinates readers. This also goes for readers of subtitling,
even if they understand the original language behind the subtitles.
In processing subtitles, the spectator is caught by the various spontaneous mean­
ings involved in visual, aural and paralinguistic signs. He becomes himself an
enunciating subject of the utterance, thus adding his interpretations to those of
the actors and the translator. He reads and interprets a text which itself is a double
interpretation, from oral to written and from one language to another.
Seeing, hearing and reading are three different skills and their social and sym­
bolic values are also different. If we 'interpret/ a speech or 'translate' a text, what
do we do with subtitling? Is subtitling 'translation' merely because the end product
is a written 'text', even though it deals both with a transcript and with the audible
dialogue which the translator also interprets, and which the spectator can hear even
if he does not understand it?

Norms
The difference between oral and written codes is not clearcut but forms a conti­
nuum. We have, for example, literary written discourse, educated written discourse,
written discourse imitating the oral, cultivated oral discourse, and spontaneous oral
discourse; the last of these has its own registers, such as ordinary, familiar and
vulgar, which are in fact not easily defined. The stress on oral or written discourse
varies also in literary texts (dialogue, interior monologue, sociolect markers). The
280 Teaching Translation and Interpreting 2

two codes are integrated in different ways in different literary genres and also vary
from epoch to epoch, according to the status of written and spoken language
during a given period and the prevailing stylistic norms.
This has a bearing on the relationship between subtitling and literary translation.
What is said on the screen (the dialogue) has its origin in a written scenario.
However, this does not mean that the actors merely recite the scenario. They also
add to the written lines their own interpretation and their presence on the scene.
Naturally they are guided by the director, but even he cannot control, for example,
the actors' voice quality. Consequently, we have spoken about oral discourse in
a communication situation which is partly limited by the script writer and the
writer. This is in contrast with the procedure employed by an author of fiction
in which the output can be developed and polished endlessly and the message be
delivered by a single person, the author.
The translator is thus not in the same position as the author; the relationship be­
tween oral and written discourse is not the same in these cases, nor are the norms.
An author elaborates a written code and rhetorical conventions, in order to perpetu­
ate, modify and/or go beyond them. The reception of his work and the underlying
expectations are purely literary. A subtitler does not have the same rights. He must
render in writing what has been formulated orally, but at the same time respect
both technical exigencies and a certain sanctity attached to written discourse in
our culture. While one of the writer's tasks might be said to be to transgress
certain taboos (thematic, stylistic or discoursal), the translator must respect (1)
norms of good usage (avoiding elements considered extremely vulgar or offensive
if they appear in written discourse), (2) readability (textual coherence being
dependent on phenomena such as word order, repetition, a certain amount of re­
dundancy, discourse markers and pragmatic connectors that are very frequent in
oral discourse (e.g.in French 'bon', 'ben', 'alors', 'à propos', 'tu sais'). This is
one of the reasons why strong sociolinguistic variation or particular linguistic
features typically characterizing the protagonists are often neutralized, making them
unrecognisable and unmarked.
At this point it is worth posing the question: what kinds of norms, if any, are
followed in subtitling? It seems that broadcasting companies and film distributors
give translators no explicit norms. On the other hand, it is apparent that some of
the countries where subtitling is used (for instance Austria and Flemish-speaking
Belgium) have adopted standardised linguistic norms which have in fact been de­
fined elsewhere, for example in Germany and in the Netherlands. Does their relat­
ive linguistic insecurity contribute to their desire to stick more closely to certain
traditions of written discourse? And how about the situation in countries like
Sweden and Norway or in bilingual Finland? So far there have been no studies
 Yves Gambler, Finland 281

investigating possible similarities between the conventions of subtitling in Sweden

and those of Swedish subtitling in Finland. Does the concept of bilingualism in­
clude other factors than those discussed here or do we simply have to look at these
factors in a different way?
Having described the perspective from which I see subtitling, I wish once more
to emphasise my conviction that it is useless to reflect on the process as a mere
difference in lexical quantity (amount of words). Such a reductionist standpoint
means ignoring the functional differences between oral and written discourse,
between psycho-auditive, if not audiovisual, perception (comprehension-interpreta­
tion) and linear perception (comprehension-readability), and the culture-bound dif­
ferences involved in both oral and written discourse.
An interesting analogy arises at this point between subtitling and simultaneous
interpreting. In simultaneous interpreting, a discourse which is too colloquial, or
which follows the conventions of written discourse too strictly, tends to be trans­
formed into 'cultivated' oral discourse, that is, to be neutralized by either 'de-
oralising' or 'oralising' the speech. As in subtitling, there is a tension between
the codes, but interpreters have a certain margin within which they can operate,
thus striking a balance between the heaviness of the formal, controlled discourse
and the more or less normative expectations of the audience.

Training in audiovisual language transfer

Subtitling comprises:
- language conversion from longer units to shorter ones,
- transfer from spoken language to written text,
- transfer from one language to another, and
- interpretation of verbal speech combined with numerous other cultural and
socio-symbolic signs or with other types of semiotic systems.
Subtitling is, however, only one of many new types of multilingual mediation
in the audiovisual media. So we are facing an educational challenge. Most studies
dealing with interlinguistic aspects of audiovisual communication have come to
fairly negative conclusions. As has been stressed several times in this article, it
is easy to enumerate omissions, condensations, résumé translations, etc. This is
a one-dimensional approach which reinforces the illusion that visual communica­
tion is unproblematic and that iconic and aural means most readily overcome se­
miotic and cultural barriers.
In the discussion, the importance and nature of multilingual audiovisual commu­
nication is too often underestimated. It is considered either as one aspect of literary
translation (or similar to it) or as a technical problem, more or less irrelevant to
training courses.
282 Teaching Translation and Interpreting 2

I believe that a degree course in language transfer within the media needs to be
developed. Such courses have been organised in Finland, Denmark (Copenhagen),
France (Lille, Strasbourg), Wales (Lampeter), and Ireland (Dublin). These courses
have usually focused on a single method (subtitling or dubbing). But the need for
collaboration - between schools of translation studies and departments of mass
communication, and journalism studies, between professional translators and tele­
vision, between translators and script writers or producers, between translators,
interpreters and representatives of the film industry and of TV companies (deci­
sion-makers, importers), between translators of dominant languages and translators
of minor or minority languages - still needs to be discussed. And what can we
expect from computers and machine translation? Do they or will they have any
implications for self-learning?
In Turku (Finland), our university training programme in television subtitling
is offered in cooperation with the national TV broadcasting company. It is based
on the use of a floppy disk and a software called 'scantitling'. With such a tool,
it is relatively easy to generate the timing of insertion and removal of the subtitles,
the duration of the two-line texts on the screen (exposure time) and the display
and format of the subtitles (font type and size, box-framing of the lines, position
on the screen, sentence division and/or any combination of sentences). The inten­
sive course (14 hours, with 12 students) includes exercises in transcription, practice
in condensation, elaboration of subtitles on a diskette with and without pictures,
and a comparison between the student's product and the actual TV output. Among
problems still to be solved, are the need for 'progression' in learning in order to
avoid repetitive work, the types of film and programme to work with, the criteria
of assessment in order to improve critical knowledge and quality, and the possibil­
ities of learning in small groups.
In conclusion, I want to stress three main characteristics of subtitling and two
potential lines of research:
1) As with other forms of translation, the subtitler has to choose certain strategies
so that the message conveyed fulfils certain social, moral, didactic, aesthetic and
linguistic functions, which go beyond the subtitler himself acting as an individual.
2) Subtitling is permanently and simultaneously constrained by two factors: first
by the demands of coherence (to make comprehensible a content which is inter­
twined with the visual side), and, secondly, by the need to appeal to and captivate
the spectator's senses, that is the ability to hear and to read and to stimulate the
spectator's interest.
3) Subtitles are constantly juxtaposed with the original - the oral discourse. This
is rare with other forms of translation, except for interpreting and bilingual publica­
tion. This co-presence of two codes and two languages will hopefully make us
 Yves Gambier, Finland 283

more tolerant towards multilingualism, if not multiculturalism.

4) Technological developments mean not only satellites, cables, high definition
TV, etc. but also digital pictures, virtual imaging, overcoming the differences be­
tween reality and reflection, resemblance and questioning the concept of probabili­
ty. What will be the function of language in these contexts?
5) Language acquisition through viewing subtitled foreign programmes and films
has not yet been closely studied. In a multilingual Europe in search of an identity,
it might be worth looking at the media as a stimulating way of learning not only
the languages but also the cultures of our neighbours.

Vor Frue Kirke, domkirken i Haderslev

Vallø Slot
TOOLS
 TEACHING LINGUISTS TRANSLATION

M.K.C. Uwajeh, University of Benin, Nigeria

Rationale
Translation studies, although somewhat recent in linguistics, are actually recog­
nised as part of linguistic inquiry by many linguists. Since Eugene Nida's pion­
eering study in the 1940s, linguists have increasingly appreciated the importance
of translation as a fitting subject for linguistics. Works by other linguists, such
as Catford (1965), Mounin (1976), Kittredge (1983), Kittredge and Grishman
((eds) 1986), etc., amply attest to the ever-increasing interest of linguists in the
nature of translation. Nowadays, translation is generally recognised as an
important part of Applied Linguistics - whether this is interpreted restrictively as
"the application of linguistics to the domain of language teaching only/essen­
tially" (Corder 1973: Preface), or defined broadly to subsume "the application of
linguistics to any domain whatsoever of human endeavour where it is applicable."
(Perren and Trim 1971)
All the same there remains in certain academic circles, especially among trans­
lation experts themselves, a persistent mistrust of the relevance of linguistics to
translation theory and practice. The general thrust of the underlying argument -
as deducible from, for example, Le Feal (1992:23: 26) or Lederer (1992: 29) -
goes more or less like this: linguistics can be used in translation, but its role there
is quite limited because translation, by its nature, goes well beyond Saussure's
langue or 'language as such' (which is understood to be the one and only preoc­
cupation of linguistics) and also deals with the particularities of Saussure's parole
or 'language use' (which some linguists erroneously believe lies outside the sub­
ject matter of modern linguistics). By and large, however, non-linguists have in­
creasingly appreciated the contribution of linguistics to translation theory and
practice since the Second World War. It takes the linguist's grammar, for in­
stance, to demonstrate the limitations of machine translation - discussed in
Coughlin (1988) - and the linguist's grammar is surely needed to develop those
'super-intelligent' language using machines of the post-1990 era, the fifth-gener­
ation computers discussed in Doi et al. (1987).
If the importance of linguistics for translation is obvious nowadays to linguists
and non-linguists alike, it is generally not appreciated by linguists that linguistics
itself could benefit immensely from the findings of translatology. Many linguists
would in fact not hesitate to assert that expertise in translation is not essential to
288 Teaching Translation and Interpreting 2

the practice of their science; yet, since in his characterisation of language (the
central preoccupation of modern linguistics) the linguist now and then has to deal
with the structural features of foreign languages, he must know how to translate
properly in order for his grammatical pronouncements to be valid. Therefore, the
rationale for a translation course designed specifically for linguists is that ideally
linguists should also be trained translators.

Contents of the course

I propose that a translation course for linguists should include all aspects of
translation - its definition, history, application(s), etc. - and that, in particular, its
central structure should be based on the tenets of 'Uwajeh's Four-Level Model
of Translation' (1992).
The notion of 'level' in my model needs clarification, since it is crucial to
understanding its tenets, and differs radically from that associated with, for ex­
ample, Catford's (1965) classical model of translation. The fundamental problem
of translation theory being that of equivalence, a 'level of translation' in my
model constitutes a particular degree of 'finesse' within an ascending order of
target language equivalents for specific source language texts, so that the first and
'lowest' translation level presents the crudest, while the fourth and 'highest'
translation level conveys the finest possible rendition. The 'level of translation'
in my model is similar to Catford's (1965: 24-26) classical model 'rank of trans­
lation' (and not to his own 'level of translation'). The four levels of translation
in my model (instead of the traditionally recognised three ranks), are presented
with their respective orthographic conventions, explained, and illustrated as
follows:
'Lexical Translation' = Lex. Tr., refers to lexical item-by-lexical item trans­
lation: it aims at an approximate conceptual equivalence of the source language
text, through the use of appropriate target language symbols (that is, form
units).
'Literal Translation' = Lit. Tr. , refers to a rough-rendition translation: at this
level, the overall propositional equivalence of the source language text, as sug­
gested by the conceptual units of the corresponding lexical-level translation, is
now conveyed with some judicious concatenation of the target language
symbols.
'Free Translation' = Fr. Tr., is a smooth-flowing translation: at this level both
the semantic and symbolic aspects of the chosen target language seek the them-
atic equivalence of the source language text.
'Figurative Translation' = Fig. Tr., where applicable, and if feasible, refers to
a special-effects translation: here, the translator makes every effort to choose
 M.K.C. Uwajeh, Nigeria 289

that target language text which, over and above expressing directly the informa­
tion conveyed with the source language text, also captures the contextual equiv-
alence of the source language text by manifesting as far as posssible the pecu­
liar, culture-specific and contextual factors which are only communicated
indirectly.
It is necessary to expatiate on the above presentation: the first and 'lowest'
(lexical) level of translation is 'crudest' in the sense that it is farthest removed
from an acceptable target language-specific rendition of the source language com­
munication (and therefore closest to source language-specific texture features),
while the fourth and 'highest' level of translation, the figurative one, is 'finest'
in the sense that it constitutes the closest target language-specific textual ren­
dition feasible (and therefore also farthest removed from source language-specific
structural particularities). It follows from the above not only that 'equivalence'
does not refer to exactly the same notion at the different translation levels (that
is, each level deals with a different type of intrinsic equivalence) but also that the
different equivalences characteristic of the different levels of translation collect­
ively make an ascending order of refinement in target language equivalents for
the given source language texts - ranging, then, from the lowest equivalence at
the first, lexical, level, to the highest equivalence at the fourth, figurative, level
of translation.
Furthermore, the four levels of translation are abstractions from the repertoire
of the translator's linguistic judgements. It should be noted that all four levels are
not needed for every translation exercise, but depend on the text and the goal of
the translation. In some cases the fourth level may thus be inapplicable and there­
fore unnecessary; the first and second levels, or the second and third, may some­
times be so similar to one another that it is unreasonable to differentiate between
them in a given context; and, occasionally, just one level may suffice for the
translator's implicit or explicit objective(s). However, it is obvious from the
nature and function of each of the levels that these clarifications do not detract
from the theoretical and practical validity of each of the translation levels' sepa­
rate identity.
Below are three sample presentations, together with appropriate conventions of
orthography stipulated for the different levels, for three different sentence ex­
amples of English, Igbo, and Nigerian Pidgin English respectively - with French
as the target language for the English sentence example and English for the Igbo
and Nigerian Pidgin sentence examples.
EXAMPLE 1
It's raining cats and dogs
Lex. Tr.: Il-est pleut-ant chat-s et chiens.
Lit. Tr.: II est pleuvant des chats et des chiens.
290 Teaching Translation and Interpreting 2

Fr. Tr.: II pleut des chats et des chiens.

Fig. Tr.: II pleut à verse.
EXAMPLE 2
Áwo àsé íbewe "wókòm".
Lex. Tr.: Frogs are-tell peers-they "wokom".
Lit. Tr.: Frogs are telling their peers "wokom".
Fr. Tr.: Frogs mock their peers for saying "wokom".
Fig. Tr.: The pot is calling the kettle black.
EXAMPLE 3
Trouble dey sleep, nyanga go wake am.
Lex. Tr.: Trouble is sleep, pride go wake him.
Lit. Tr.: Trouble is sleeping, pride goes to wake him.
Fr. Tr.: Trouble is asleep, but pride would wake him.
Fig. Tr.: Let sleeping dogs lie.
Of the four levels of my model, the first two - those of lexical and literal trans­
lations - are of special interest to linguists taking a course in translation, since
these levels are source language-oriented, in the sense that they emphasise source
language text structural peculiarities. As such, these are the levels whereby the
linguist draws attention to source language grammatical peculiarities by means
of the target language. Correspondingly, the third and fourth levels of the model -
those of free and figurative translations - are target language-oriented in the sense
that they emphasise target language-specific features. These higher levels of
translation are obviously complementary for a linguist communicating source lan­
guage grammar: they help a target language user compare and contrast the source
language with his target language by making reasonable sense of what the source
language community members are expressing directly and/or indirectly with their
different textual structures. This is done by confronting him with a rendition us­
ing his own target language-specific particularities at the lexical and literal trans­
lation levels.

Method
From the above discussion, it is fairly evident that the objective of the course
in translation for linguists which I envisage is to teach linguists to translate
properly. In other words, the course has a practical bias. This does not mean, of
course, that theory should be excised from the course - the central structure of
its content, as outlined above in 'Contents of the course', is theory-laden - but
that such theory should be geared towards the pragmatic function of making good
translators of linguists. Theory is no doubt also indispensable in the following
non-trivial respect: since translation theory is part of general linguistic theory,
every linguist should be interested in whatever insights the nature of translation
gives us about the nature of language, but that is not the concern of the present
article.
 M.K.C. Uwajeh, Nigeria 291

Some other possible pedagogical issues of the course, besides the question of
bias or focus, are those of who is to teach linguists translation, and when this
should be done. It is here recommended that, as far as possible, teachers of trans­
lation to linguists should be translation experts who are professional translators
themselves and have had sufficient training in linguistics to understand intimately
the need of linguists for a practical translation course. The best period to teach
linguists translation is, I think, in the early formative years - say, by the second
semester of their first year or the first semester of their second year of undergra­
duate studies, before they have been initiated into deep grammatical analyses; in
Nigeria, these students would be about twenty years of age. Professional linguists
could get their dose of translation practice in specially designed crash-programme
translation workshops or seminars.
In teaching linguists translation, special attention should be paid to pitfalls in
translation which are (most) likely to invalidate the linguist's scientific results.
There are several such (possible) pitfalls; but here I shall examine only two - one
at the lexical level, and one at the literal level of translation.
At the lexical level of translation, a fairly common source of error for the un­
wary linguist is due to a target language/source languge 'categorial interference'.
In these cases the linguist's translation reveals that he has ascribed to the source
language a lexical category which it does not have in that context, but which the
target language normally does have in free usage. This is illustrated in the
following examples.
EXAMPLE 4
Polínà èjí Njí.
Fr. Tr.: Paulina is black.
EXAMPLE 5
Akpáná jò Njó.
Fr. Tr.: Akpana is ugly.
EXAMPLE 6
Óbyé-Nkwuzí lí ùlá.
Fr. Tr.: The teacher is asleep.
EXAMPLE 7
Wà jí áká-ódo tigbú ónnyé-óshi.
Fr. Tr.: They beat the thief to death with a pestle.
EXAMPLE 8
Ukàdíke si Káduná jé Josi.
Fr. Tr.: Ukadike went from Kaduna to Jos.
Although 'black' is the English Free Translation equivalent of the Igbo 'Njf
of sentence 4 above, for example, the English lexical item here is an adjective
while the corresponding Igbo element is a noun. An inappropriate approach to
translation leads to the grammatical mistake that 'Njf is also an adjective merely
because 'black', an adjective in English, can be translated freely. 'Njó' and 'úlá'
292 Teaching Translation and Interpreting 2

(in 5 and 6), which can similarly be substituted with the English adjectives 'ugly'
and 'asleep', respectively, in Free Translation, are both, again, nouns in Igbo.
Also the Igbo elements 'Jì' and 'sì' (in 7 and 8), which appear to be freely trans­
latable with the English 'with' and 'from' respectively are not prepositions in
Igbo but verbs. Appropriate lexical-level translations avoid possible grammatical
blunders by underscoring any source language/target language categorial differ­
ences in these examples:
EXAMPLE 4
Pòlínà èjí Njí.
Lex. Tr.: Paulina is-'black' blackness.
Lit. Tr.: Paulina 'is blacking' blackness.
Fr. Tr.: Paulina is black.
EXAMPLE 5
Ákpáná jò Njó.
Lex. Tr.: Akpana 'ugly' ugliness.
Lit. Tr.: Akpana 'uglies' ugliness.
Lit. Tr.: Akpana is ugly.
EXAMPLE 6
Óbyé-Nkwuzí lì úlá.
Lex. Tr.: One-teaching is sleep.
Lit. Tr.: Teaching-person is sleep.
Fr. Tr.: The teacher is asleep.
EXAMPLE 7
Wà jì áká-ódo tigbú ónnyé-óshi.
Lex. Tr.: They held hand-mortar beat-kill one-theft.
Lit. Tr.: They used hand-mortar to beat-kill theft person.
Fr. Tr.: They beat the thief to death with a pestle.
EXAMPLE 8
Úkàdke s Kddúndje Jòsì.
Lex. Tr.: Ukadike went-from Kaduna go Jos.
Lit. Tr.: Ukadike went from Kaduna go to Jos.
Fr. Tr.: Ukadine went from Kaduna to Jos.
For its part a good literal translation draws attention to certain facts and mys­
teries about a source language which would otherwise elude the linguist in a
careless translation. In the constructions which follow below, the relationship be­
tween the verb and indirect object may be quite unclear in Igbo (of the Onicha-
Ugbo dialect). It is not clear, indeed it is a mystery, what exactly the Igbos in
question are reporting with the verb-indirect object construct in the sentences.
EXAMPLE 9
ícho mádù ókwú
Lit. Tr. to seek a person word
Fr. Tr. to seek a confrontation with someone,
Here the question is: is one seeking the 'word' off someone or of him?
 M.K.C Uwajeh, Nigeria 293

EXAMPLE 10
íjì mádù úgwo
Lit. Tr. to hold a person debt.
Fr. Tr. to be indebted to someone.
It is not clear as to whether one is 'holding the debt' from someone or to him?
And in EXAMPLE 11,
ísó mádù ámú
Lit. Tr. to 'sweet' a person laughter
Fr. Tr. to amuse someone,
it is not obvious whether the thing is 'sweeting' someone with or by laughter, or
is the thing 'sweeting' laughter to someone.
It could well be that the questions I have asked in the above paragraph are the
wrong (kinds of) questions for the Igbo source language - a reflection of present
ignorance of the facts involved, and only the result of feeble attempts to capture
the subtlety of the Igbo world picture through a radically different mode of ap­
prehension peculiar to the English language community. One cannot be sure what
is what until linguists have conducted the necessary semantic investigations satis­
factorily. One thing that linguists should certainly not do - as I have seen done
too often in modern linguistics - is to pretend that no mysteries really exist in the
examples of my illustration above and innumerable other similar ones, by simply
imposing target language semantic structures on the source language with bad lit­
eral translations, and thereby wiping out summarily any source language/target
language inherent differences.
Conclusion
I have examined the modalities for teaching translation to language scientists
in linguistics like myself, who need help from translatologists. I have tried to
explain the rationale for such a course of study, to determine what its content
should be, and how it could be conducted so as to be effective.
The proposals made in this study are meant to be tentative in the short term,
but in the long run they should provoke reactions which will lead to the estab­
lishment of a comprehensive programme for the novel field of translation ped­
agogy with which I am concerned in this article.
 TECHNICAL TRANSLATION:
PUTTING THE RIGHT TERMS IN THE RIGHT CONTEXT

Peter Baumgartner, Flensburg Polytechnic, Germany

In its translation studies program, the Flensburg Polytechnic specializes in the

training of technical translators, therefore the strategies and insights discussed in
this article refer exclusively to such training.

Teaching components in the technical translation program

At the Flensburg Polytechnic the teaching of translation centers on English and
German. It takes place in three units. The first unit - teaching of the subject
matter - in the case of Flensburg mechanical engineering, electrical engineering
and computer science - lays the foundation and provides the knowledge base for
the other two units, which are translation classes. Unit two covers the translation
classes conducted by engineers and/or natural scientists, and focuses on why an
engineer or scientist would express the technical facts contained in a source text
in only one specific way and no other way in the target language. Unit three
covers translation classes which investigate and concentrate on the language used
to express these technical facts. The first two units deal mainly with the "what"
and "why" and are technical and subject-oriented, whereas the third unit is pre­
dominantly concerned with the "how" and is language-oriented.

Typical difficulties encountered in technical translation

This article concentrates on the third unit. My former experience in translation
services in industrial companies and my present experience at the polytechnic
show that the major problems technical translators have to grapple with lie in the
following areas: (a) translators may have difficulty in understanding the technical
information in the source text, and (b) they may have difficulty in formulating
the technical facts in the target language by appropriate linguistic means so that
specialist readers will not realize that they are reading a translated text. In
extreme cases the translators understand the technical information in the source
text perfectly and even know the proper technical terms in the target language,
but still cannot render this information in the target language, simply because
they do not know "how to say things" in a certain field of knowledge, no matter
whether the language involved is the mother tongue or a foreign language.
296 Teaching Translation and Interpreting 2

Stages in the technical translation process

Technical translators have to understand the technical information as far as pos­
sible, extract this information from the text and dress it in the appropriate words
of the target language. On the language side, this means that they have to know
and master not only the special terms forming a network throughout the texts
dealing with the subject area in question, but also the typical contexts in which
these terms occur. That is, the technical translator must convert the source text
into abstract knowledge and subsequently this abstract knowledge must be ma­
terialized again by converting it into a text in the target language. In order to be
able to do this the translator must understand the text and at the same time must
have at his or her fingertips the linguistic means required for text building in the
target language. When building the new text, he or she first needs the terms
representing the technical concepts. This one may regard as the more technical
knowledge-oriented aspect, because in order to decide which term represents
which concept, the translator must draw on his or her technical knowledge base.
Once the correct term has been found, it must be incorporated in a syntagma of
some sort. Finally, the syntagmas have to be incorporated into larger units to con­
stitute a text (Baumgartner 1990: 7). Finding the appropriate syntagmas, that is
packaging the terms properly, is one major problem area of our students. This
problem exists irrespective of whether they translate into their mother tongue or
into a foreign language, in the case of our polytechnic, exclusively English.

Translation students vs. engineering students

In this light, it is interesting to compare our translation students with the engin­
eering students. In principle, the engineering students have the same problem.
They, too, have to learn a new language, namely a sub-language of their mother
tongue. The difference between the translation students and the engineering stu­
dents, however, is that the latter acquire the means for depicting technical subject
matter in language (in their mother tongue) while dealing with the subject matter
proper. This is both to their advantage and disadvantage. It is to their advantage
that on the one hand they use the language tool to acquire knowledge of the
subject matter and in so doing they unconsciously learn the sub-language of their
field. They pick up the subject matter and in the process the language. It is to
their disadvantage, however, that this process takes place unconsciously, instinct­
ively and automatically and, therefore, to a certain degree, arbitrarily and in an
uncontrolled way, which means that to a large extent the level of competency in
their technical sub-languages is left to chance. Using it merely as a tool, most of
them are not aware of the language they use and see no need to reflect on it.
Whereas the engineering students learn the typical technical terms and syntagmas
 Peter Baumgartner, Germany 297

of the sub-languages of their mother tongue while learning the technical subject,
the students in the technical translation program have to learn the terms and their
usage in syntagmas consciously and by a separate and systematic effort.

Theoretical assumptions for teaching work in the third unit

Work in the third unit is based on the assumption that a translator working in
a particular field has to deal with a finite number of technical facts which come
up again and again and which are translated into language in a finite number of
ways and thus constitute a major part of the source texts encountered in actual
practice (Baumgartner and Kraus 1992: 57). In translation the standard technical
facts expressed in the source text can be translated into the target language again
only in a finite number of ways. Therefore, one could consider them as the static
portion of technical texts. The dynamic portion of the texts would be determined
by the text category, the purpose of the writer, etc. The standard facts represent­
ing the static portion of technical texts are, so to speak, fixed modules of thought
which can be transformed into a limited number of modules of language both in
the source and the target language. These modules are integrated into larger units
so as to form a text. This means that students have to learn the modules of
thought and the various ways of converting them into the sub-language of the
subject areas of specialization. It also means that when our students graduate as
technical translators from the polytechnic they must master a certain number of
these modules of thought and language which will serve as a basis for their
future translation work.
Ideally, translators must master the technical subject matter of the field in
which they do translation work, but on the other hand they must also master the
linguistic means used to describe these facts.

Some approaches tested in the teaching of technical translation

As mentioned before, it is a common experience in our translation classes that
students have problems in doing technical translations not only in their foreign
language but also in their mother tongue due to a lack of familiarity with the
technical texts of a special field. Very often they are insecure, for example, about
what verbs, adjectives and prepositions can be used as partners of a given term
(Arntz and Picht 1991: 34) and do not know the building blocks required for the
construction of the strictly technical parts of their texts.
To remedy these deficiencies, I have been and still am looking for ways which
help overcome these difficulties. One approach has been to start translation ex­
ercises in a particular technical field by analyzing technical texts within the same
field in the target language before actually starting any translation work. Hohn-
298 Teaching Translation and Interpreting 2

hold (1990: 24) calls these texts 'Gegentexte' (similar texts in the target lan­
guage). Ideal for this purpose are texts in which the typical technical facts of the
area in question are linguistically described. In the process of such an analysis,
the terms representing typical concepts are isolated and their textual environment
is investigated, that is, a closer look is taken at the collocational partners of the
terms. Not only the terms, but also the larger modules in which the terms occur
must be studied (Warner 1966: 9). In this phase, I also use 'cloze' texts in the
target language to demonstrate to students that knowing the partners of the terms
is not a matter of course and cannot be left to the creative imagination of the
translator. It also makes them aware of the special difficulties in finding the valid
word combinations in a particular field and that only a limited number of them
is available.
In another step, I use overheads on which the texts to be translated are reduced
to sets of terms and text fragments, each set representing one item of information.
From these sets of terms and text fragments in the source language, the students
have to build a text in the target language. In this way they no longer have syn­
tactic patterns in the source text which may serve as a crutch and are often mis­
leading. They have to identify the concepts behind the terms, think of possible
relationships between these concepts and find ways of representing this informa­
tion linguistically. In order to be able to do this they have to resort to their tech­
nical knowledge base which means that their translation approach is no longer
language-, but subject-oriented. In a first run, for example, we move from term
to term, and the students have to be able to make minimal true statements using
these terms. To facilitate the task I usually give directions of some sort. I may,
for instance, ask them to use the broader or narrower generic term in a statement,
cite special qualities of a technical object, describe typical movements, functions
and so on. Technical texts are, among other things, characterized by the fact that
special typical relationships exist between the terms occurring in the texts (Sager
1990: 29-37). Therefore, when we start relating the terms and text fragments to
one another to arrive at more complex statements, I may ask my students to es­
tablish a generic, partitive or causative, etc. relationship between two terms, to
express certain facts nominally or verbally, etc. Thus we move from one set of
information to the next, each set being linguistically depicted by one or more
sentences. In this way we obtain a number of true statements about all kinds of
technical facts within a given field of knowledge. In itself, this accumulation of
sentences, however, does not constitute a text. While in the first run we look at
the terms and text fragments and find out what their relationships are, in a second
run, we look at the statements produced and look for possibilities of whether or
how they can be related to each other. In order to obtain coherence and cohesion,
 Peter Baumgartner, Germany 299

we may have to switch sentences around, split single sentences up into two or
more, combine simple sentences into more complex ones, supplement with logi­
cal connectors to make things clearer, or even add new sentences. These are the
finishing touches applied in the second run, resulting, hopefully, in a text in the
target language which is a true picture of the source text.
One might argue that exercises of the type described above apply more to the
domain of technical writing. But then it could also be said that good technical
translation is nothing but "technical rewriting" of the source text in the target
language.

Viborg Domkirke
Bregentved Slot
 COMPUTER-ASSISTED TRANSLATION:
THE STATE OF THE ART

Robert Clark, Praetorius Limited, United Kingdom

I am not an information scientist, not a computational linguist, and I am defi­

nitely not a salesman. I am not writing this to sell anything to anybody, but as
Software Editor of Language International I am constantly reviewing and evalu­
ating translation-related software products as they appear on the market. As a
result, I am in a position to recognise the crucial role that technology plays in the
translation process.
On this background, I shall:
- give you a brief overview of the state of the art of commercially available
Machine Translation systems and computer-based translation tools. I stress
the term commercially available, because people can claim what they like
when systems are still at the developmental stage, but it does not mean
much to us until we can actually go out and purchase them.
- discuss prevailing attitudes of translators to computer-based translation,
especially Machine Translation.
- and, finally, discuss the impact that language technology is having on the
translation industry as a whole and the implications this should have on
translator training.
The career of a new translator entering a university today will take place
mainly in the 21st century. We have to concern ourselves with what sort of
knowledge this translator needs to meet the challenges and survive the competi­
tion of that century. Advances in language technology are already starting to
produce changes in translation practice, but the major impact is yet to come. The
importance of language engineering is illustrated by the fact that the European
Commission is planning a 667-million ECU programme to promote language
technology in the period 1994-98. This programme signals the shift away from
a situation where a handful of isolated developers was working producing
systems. It is hoped that this initiative, and others like it, will result in greater
development cooperation and the sharing of information on natural language
processing research.
In this discussion we must take into account the fact that translators often have
their reservations about using computers in the translation process and are less
302 Teaching Translation and Interpreting 2

than totally accepting of the idea of computer-based translation.

The reservations volunteered from an informal survey produced the following
list:
1. We might be out of a job.
2. Computers do not have enough knowledge to produce good translation.
3. We are put off by extravagant claims.
4. Machine Translation has no soul.
5. A computer has no idea of sense/meaning.
6. A computer cannot produce nuances of meaning.
7. Computers are difficult to master.
The list can serve as a point of departure for discussion.
In his Introduction to Machine Translation, John Hutchins defines it as "a
computerised system which produces translations from one natural language to
another, with or without human assistance" (Hutchins and Somers 1992:3). By
qualifying his definition with "human assistance" he is implying that fully auto­
matic, high quality translation actually exists. That was certainly the dream over
forty-five years ago, but even today the output of the best systems is still unpol­
ished. As early as 1951 Yehoshua Bar-Hillel insisted that translation of a stan­
dard comparable to that produced by human translators was not only an unrealis­
tic aim for research but also impossible in principle. He was not implying that
Machine Translation should be scrapped, but that efforts should be redirected to
more realistic aims. In his opinion, translation involved certain human abilities,
such as real-world knowledge, which no computer could ever assimilate. Recent
advances in Artificial Intelligence indicate that real-world knowledge can, in fact,
be incorporated into Machine Translation systems. But it has not happened yet.
So the best we can expect is human assisted machine translation (= HAMT). In
other words, currently, no fully automatic systems can translate to the same high
standard one would expect from a human translator. There must be human assist­
ance at some stage or some kind of restriction on the source text, so the human
translator is still very much in the frame. Although Machine Translation is less
than perfect, it is here to stay. Given these premises, we can freely discuss its
role in the translation industry.
Machine translation is discussed in-depth at many conferences. At a conference
of the American Translators Association in November 1992, L. Chris Miller
presented the results of a survey that she had done of commercially available
Machine Translation systems in North America and Europe, but, unfortunately,
not in Japan and the rest of the world.
The mainframe/workstation systems were, for instance, SYSTRAN, LOGOS,
METAL and TOVNA. The personal computer translation systems were PC
TRANSLATOR and GLOBALINK. A few of these systems have been around
for quite a while. Although they have not reached perfection, they have been
 MT User Sample
USER SYSTEM DOC TYPE STAFF YEARLY
OUTPUT

US Air Force SYSTRAN Scientific/ 13 Full-Time Post-Editors 50,000 pp

Technical Articles 1 Half-Time Post-Editor
& Documents 2 Half-Time Terminologists

XEROX Corporation SYSTRAN, Service & Customer 50 Full-Time Translators 36,000 pp

ALPS Documentation

Pan American Health SPANAM, Texts on Public 3 Full-Time Post-Editors, 10,000 pp

Organisation ENGSPAN Health, Agriculture, 23 Part-Time Post-Editors,
Management, etc 1 Contract Terminologist
Robert Clark, United Kingdom

Siemens METAL, In-House Technical 2 Full-Time Post-Editors, 10,000 pp

Stromberg-Carlson Documentation 1 Full-Time Terminologist

Lexi-Tech LOGOS Technical Manuals 45 Post-Editors, 100,000 pp

1 Terminologist

Hartmann SYSTRAN Technical Manuals 1 Full-Time Post Editor, 12,000 pp

International 1 Full-Time Terminologist,
303

43 Part-Time Post-Editors
304 Teaching Translation and Interpreting 2

steadily fine-tuned and more and more language pairs are being offered. Four of
the mainframe systems are also available on desktop models. This 'down-sizing',
to use the current jargon, is probably the most significant development in Ma­
chine Translation. Systran is now available on an IBM PS2 desktop computer
using an AST 3270 emulation card. The PC system does exactly the same job as
the mainframe - only quicker! SYSTRAN are currently porting their code to the
C programming language which should allow the system to operate on an ordi­
nary 486 PC. They also hope to offer a scaled-down version which would enter
the marketplace at about the same price as Globalink GTS or PC Translator.
One can get an impression of annual translation volume that is machine pro­
cessed from the sample shown on the opposite page. It lists a few major users
of Machine Translation in the US and Canada, and it was compiled by Joann
Ryan of the Systran Corporation.
It may be that the figures are not impressive in, say, a European context, but
it allows for making a few general points.
Except from the SPANAM and ENGSPAN systems, all others employ the old
'workhorse' systems, SYSTRAN, LOGOS and METAL. It is also evident that
technical documents with very limited domains are the type of text best suited
for processing by Machine Translation. As mentioned, people are sceptical about
Machine Translation because it has "no soul". But there is little "soul" in most
technical translation work. To describe it as boring would be charitable. If a com­
puter can eliminate that kind of drudgery, then roll-on Machine Translation! It
will also be noted that there is a high number of post-editors. Post-editing is
rapidly becoming a profession in its own right. In the US, SYSTRAN started to
sell finished machine-processed translation directly to the consumer in 1990. In
the first year of the experiment, they processed 7,000 pages. In 1992 they pro­
cessed 17,000 pages. Their post-editing work is subcontracted to translation com­
panies. So, machine translation, instead of depriving human translators of work,
is creating new jobs.
To summarise, Machine Translation developers have not broken the linguistic
equivalent of the sound barrier - Fully Automatic High Quality Translation. How­
ever, significant progress has been made. Development has reached a third gener­
ation of systems. The increase of performance coupled with the enormous de­
crease in the price of personal computers is putting Machine Translation within
the reach of everyone.
Of course, the most important spin off from Machine Translation development
is the ever-growing supply of tools that are being produced to assist the trans­
lation process. Not long ago you could count the number of translation tools
commercially available on one hand. Today there are so many that it is becoming
 Robert Clark, United Kingdom 305

a full-time job just to keep informed of what is being released. At present the list
comprises:
Word Processors
Spell, Grammar & Style Checkers
Electronic Dictionaries
Terminological Databases
Text Retrieval Packages
Modems
BBS Services
Workbench-style Integrated Packages
Word processing packages have become very powerful. The new version of
WordPerfect includes a spreadsheet feature and will soon incorporate a grammar
and style checker. On-line electronic dictionaries are now commonplace and
powerful terminological database programs such as MutliTerm and Termex are
an essential part of a translator's reference material. Sophisticated searches of
previously translated documents are now possible with text retrieval packages
such as WordCruncher. Terminology exchange regularly takes place using mo­
dems with CompuServe or the worldwide dedicated translator bulletin board ser­
vices that are being created. The most ambitious of these tools are the translation
memory-based integrated packages. At Language International we have gained
hands-on experience with three systems which have come on the market during
the past year, namely:
- IBM's Translation Manager/2, which we reviewed when it was still being
tested internally at IBM European Language Services, Birker0d, Denmark.
- TRANSIT, developed by STAR AG, a software company based in Stein am
Rhein, Switzerland.
- Translator's Workbench II, developed by Trados, a Stuttgart-based
company.
I shall not discuss the differences between these systems, such as language and
file format support, hardware requirements, etc., but concentrate on what they
have in common.
- Terminology management
The key to consistency of terminology use in any large-scale translation project
is the ability to identify terms beforehand and ensure their consistent use by the
translator. As projects become larger and deadlines become shorter, it is not
unusual for more than one translator to work on the same project. Therefore,
it is essential that a common terminology database is used for each project and
maintained throughout the process.
306 Teaching Translation and Interpreting 2

- Word processing
An efficient, user-friendly word processor with powerful editing features is the
core of any translation project.
- Automatic dictionary look-up
There is little point in going to the trouble of ensuring that terminology is
agreed beforehand and stored in a database if translators choose not to look up
the term. The choice is, therefore, taken out of their hands by having terms
automatically checked against the database.
- Tag protection
It is time-consuming to reformat texts when formatting tags have been lost as
a result of exporting and importing files. It is, therefore, vital that all formatting
tags remain intact and are protected from accidental deletion during the trans­
lation process. This is done by the three systems.
- Translation memory
This is probably the most revolutionary feature of the programs we are dis­
cussing. A database stores previously translated material in such a way that
each segment of source text is linked to a corresponding segment of target text,
thereby producing a synchronised bilingual text file. Each segment is usually
a sentence. During the translation process each segment of the source text is
matched against the translation memory database. If a match is found, the
previous translation is made available for use in the current translation. After
each segment of the source text is translated, it, in turn, is stored on the data­
base and becomes part of the translation memory.
- Automatic translation
During the analysis stage, that is when the source text is being processed and
checked against the terminology database and the translation memory database,
an option is available to have all exact matches against the translation memory
automatically translated.
Until now, all translations had to go through the program in order to become
part of the translation memory database, but IBM now supply an algorithm with
their product that allows you to create a translation memory using source and tar­
get texts that have not been processed within Translation Manager/2. This implies
that any previously translated text in electronic form is potentially translation
memory material.
The other thing these products have in common is that they are not cheap.
Prices range from three to five thousand dollars, so most of us would have to be
convinced that, by purchasing one of these packages, our output and quality
would improve. It is obvious that the translation-memory approach is best em­
ployed for medium to large-scale projects that involve a fair amount of repetition.
 Robert Clark, United Kingdom 307

Under these circumstances, once the translator has become familiar with the new
environment, our experience indicates that speed and consistency are definitely
improved.

Impact on translation work

Translation buyers have always had to consider delivery time, cost and quality.
They always want the translation yesterday. It seems that translation is always
the last thing considered in the production of documentation. The salesman has
promised the product on a certain date and that is when the product will be de­
livered. Quite often the translation is not even commissioned until the product is
sold to the customer. Obviously, if two translation suppliers are competing for
the same contract and one is equipped with tools that allow for faster production
of the translation, the job will be awarded to that company, for, the translation
buyer does not care about "soul"! Next in priority to delivery time is cost. Trans­
lation is always too expensive. Ideally, the translation buyer would like the trans­
lation to be free! Time is money, so it follows that if the translation supplier can
produce a translation faster than the competition, there is a good chance that pro­
duction is also cheaper. If the translation buyer requires quality in the traditional
sense, the advantage of ensuring consistency of terminology is obviously attract­
ive. If one of the ways that you plan to guarantee quick turnaround is to use a
team on a translation project, the translation buyer is certainly going to like the
idea of all translators using the same terminological database containing pre-
agreed terminology. On the other end of scale there is translation for information
purposes: The translation buyer does not always require a perfect translation. He
might have a stack of documents that fill a room that would never get translated
in the first place if traditional translation was the only option. Unfortunately, the
translator is not only reluctant to produce a less than perfect translation, it would
probably not be possible for him or her to do so. A human translator is either
"off" or "on". This is where Machine Translation comes in. If a translation is
only required for information purposes, it does not really matter how many
howlers are produced.

Conclusion
I have discussed the impact that language technology is having on the trans­
lation industry out in the real world. The question is, how prepared is the average
language graduate for survival in this real world? Dieter Walterman's article on
the programme for translation students at Carnegie Mellon discusses what is
happening at one end of the scale (below, pp. 307-317). Many other institutions,
the University of the Saar at Saarbrucken to name one, are offering advanced
308 Teaching Translation and Interpreting 2

courses in language technology to their translation students. At the other end of

the scale, there are people graduating from universities around the world, without
ever having even touched a word processor. I am not saying that every language
graduate should be able to leave university with enough knowledge to enable him
or her to build their own translation system, but they should be acquainted with
the basics. You may also feel that all this technology has no relevance to literary
translation. If you can earn a living translating James Joyce, then maybe there is
no problem. However, the bulk of commercial translation work is technical. It is
the responsibility of educators to arm students with the skills that will be ex­
pected by potential employers.
It is not fair to send them out into the real world without these skills.
 MACHINE TRANSLATION SYSTEMS
IN A TRANSLATION CURRICULUM

Dieter Waltermann, Carnegie Mellon University, USA

Background
The need and demands for translations have in recent years increased consider­
ably, primarily due to an ever-growing internationalization and globalization of
existing markets. This need and demand for translation has also outpaced the pro­
duction capability of human translators. While it is true that more and more
translators rely heavily on a wide variety of basic computational means to speed
up their translation tasks in order to meet the increased demands for translation
services, their computational means tend to be limited to basic word processing
equipment, dictionary packages and, possibly, higher-end computer systems. But
this use of more or less basic computational means must not be confused with
machine translation (= MT) as it is used in the context of the following
discussion.
The constantly growing demand for translations has also led to a tremendous
increase in activities centering on various forms of machine translation systems.
The European Union, with its nine working languages, is one of the major parties
interested in expanding MT technology; others include the US government and
major international companies, such as IBM and Siemens. Depending on the type
of MT system, researchers differentiate between:
■ Machine-Aided Human Translation (= MAHT),
■ Human-Aided Machine Translation (= HAMT), and
■ Fully Automatic Machine Translation (= FAMT).
In the past, accuracy of translations produced solely by MT systems has fre­
quently proved to be a problem. Integrating human translators into the machine
translation process has, therefore, been viewed by some as the necessary step to
augment the existing chain in order to produce fast and accurate translations.
Today's FAMT systems are neither a suitable pedagogical environment nor
suited for training new translation students. Yet, machine-aided human translation
and human-aided machine translation (henceforth referred to as Machine-Aided
Translation [= MAT]) can serve the purpose of educating students in the use of
such systems as well as training the students in the fields of scientific-technical
translations. However, the process of integrating human translators into the MT
environment has not received the pedagogical attention it deserves.
310 Teaching Translation and Interpreting 2

Future translators must be brought up to a level of sophistication where they

are routinely able to use state-of-the-art computational environments. Institutions
training translators must, therefore, develop symbiotically the pool of highly
trained human translators and the technological base for machine-aided trans­
lation. Today's state-of-the-art MT systems can be successfully deployed only if
a human translator is present in the translation loop. This presence depends, in
essence, on the different types of approaches to MT, the type of system in use,
and the quality of the system. The human translator must, therefore, take on the
role of a pre-editor, an interactive editor, or a post-editor. One reason why MT
systems have not been accepted in actual translation environments is due to a se­
vere shortage of formally trained specialists who can perform these tasks. This
situation persists even though, from the purely economic standpoint, the use of
MT already makes sense today, as evidenced by several commercially available
MT systems, such as InterLingua, LOGOS, and Metal. However, training in the
application and maintenance of such MT systems is still rudimentary. This situa­
tion can be remedied by training translation students in the fields applicable to
MAT.
In 1992, the Department of Modern Languages at Carnegie Mellon University
realized the shortcomings outlined above and initiated first steps to establish an
innovative translation program by collaborating with the institution's Center for
Machine Translation which has long been recognized as one of the leading MT
research centers. The new program is technology-based and concentrates on
teaching translation skills using MT tools. The unique aspect of the program is
best witnessed in the fact that no other translation program in the United States
is currently utilizing MT technology in its teachings leading to a degree or certifi­
cate in translation.
In addition to high levels of proficiency, the education of a modern-day trans­
lator must centrally involve a technological angle, including the knowledge of
up-to-date computer-based word processing equipment, document processing, and
desktop publishing software, among others. Specialized translators' work stations
will enhance the productivity of any translator. Yet, it is essential to train trans­
lators in the use of such workstations from the very beginning.
This training must, however, do more than merely expose students to machine-
based translation tools and present them with introductions to MT systems. In an
attempt to classify the current needs of translators and to establish a comparative
database, an extensive survey of translation degree and certificate programs avail­
able in the United States was conducted during the fall 1991 and spring 1992 se­
mesters (Waltermann 1992b). Based on this survey, it was decided to provide
technology training using a translation-based curriculum in the Department of
 Dieter Wältermann, USA 311

Modern Languages. This means that the students work with high-end UNIX-
workstations to perform translation task - they are not merely being taught the
use of computational facilities. Such a technology-oriented translation program
is also extremely beneficial from a research and development perspective. The
program integrates, and relies on, expertise of researchers at CMU's CMT and
experienced translators in the Department of Modern Languages. It also integrates
existing and future translation tool research and development projects in the
curriculum.
Parallel investigations into the possible group of students interested in the trans­
lation courses revealed a variety of undergraduate and graduate students at CMU
and the University of Pittsburgh: students who are double majors in a language
and a scientific field (computer science, engineering, chemistry, physics, graphics
design, and others), and language-proficient students who are enrolled in science
courses. All students are computer-literate since they had to take required intro­
ductory computer workshop classes lasting several weeks during their first seme­
ster (if they had no computer skills), so that there is no need to allot any course
time for teaching computer skills.
Students interact and react with available MAT systems. This interaction greatly
benefits the students by giving them an opportunity to work in a state-of-the-art
computational translation environment and by acquainting them with MT testing
and evaluation procedures. Another benefit is the valuable feedback from the stu­
dents to developers of MAT technology, which will guide improvements to the
existing technology. The importance of this program is evident in the fact that
it addresses the existing lag between technological advances and pedagogical en­
vironments with sound, high-quality computational translation teaching pro­
gram-environments, and in the fact that it teaches actual translation skills in a
classroom setting using an MAT system, the Translator's Workstation (= TWS).1
The TWS consists of a number of application modules integrated through the
central user interface. Each application uses the facilities of the user interface for
display and input, each module uses a standard window to interact with the user,
and each window has standard menus which allow the user to invoke any other
module. Each module also has special menus. Text editing can take place in any
window, and text may be moved between windows with the help of the global
kill-ring facility (for a detailed view, see Cohen et al. 1993; and Wältermann
1992a).

The TWS in the translation curriculum

The translation program survey identified several gaps in existing translation
curricula (see also Benton 1989; Vasconcellos 1990), uncovering the following
312 Teaching Translation and Interpreting 2

needs:
■ post-editing machine output,
■ training, coordinating, and supervising teams of translators using MT
systems, and
■ learning the skills to augment the lexicons and other rule bases of MT
systems.
Training students in these translation-related skills plays a central role in our
translation courses. Besides giving the students the linguistic and cultural skills
required of translators, they are being trained in maintaining and updating MT
systems. This skill will be increasingly in demand as an ever-growing number of
MT systems is deployed. The development of a curriculum to train such special­
ists is predicated on the use of advanced computerized translator's tools. A curri­
culum concerned with these issues must rest on a firm technological structure.
Technology will not only allow the students in such a program to reach new skill
levels. It will also lead to an intensification of the basic translation curriculum
itself. As the first step, we established a two-course sequence at the undergradu­
ate level teaching scientific and technical translation skills (fourth and fifth-year
studies).
The first course serves as an introduction to scientific and technical translations
and to the MT system with which the students will be working throughout the
first and second course. The second course presents students with an actual trans­
lation project to be performed using MAT tools. Both courses are served by the
technology available at CMU. Existing networking capabilities allow instructor
and students alike to share their translations for immediate comparison and edit­
ing. TWS allows students to see their own translations next to the translation of
one (or more) of their peers or the instructor. Students are encouraged to act as
peer editors which will help them learn to improve their writing and translating
abilities. The first course provided not only a testing ground for the new curricu­
lum, but also for the technological structure. Recommendations and improve­
ments resulting from the classroom evaluation of TWS were then applied in
further developing the system for use in the second course.
Since TWS supports multiple editing windows, students may view and edit se­
veral translations simultaneously. In order for students to translate using the TWS
as well as to perform editing and consolidating tasks, it will be necessary to in­
corporate appropriate scientific and technical texts and to prepare this material
for translation purposes, in addition to augmenting dictionaries and termbanks to
accommodate the new technology presented in the text materials. Part of this
work is done by the translation students as outside assignments. The students
generally work in small groups according to their background and their area of
 Dieter Wältermann, USA 313

studies (major or minor) in order to take advantage of their terminology skills.

Contents of the courses

The contents of the courses are as follows:

SCIENTIFIC-TECHNICAL TRANSLATIONS: COURSE 1

A. Historic and Linguistic Background to Technical and Machine Translation

a. Introduction to Scientific-Technical Translation and Machine Translation
* History of (scientific and technical) translation; Survey of machine translation; Influence of
linguistics and computer science on machine translation
b. Language Knowledge in Scientific-Technical and Machine Translation
* Text types: analysis, role of pre-editing and post-editing; Characteristics and limitations of
scientific-technical texts; Linguistic analysis of scientific-technical texts (word, sentence, text
level); Basic concepts and system characteristics of machine translation systems; Importance
of terminology

B. Introduction to the Translator's Workstation

c. The Translator's Workstation
* Underlying rationale (Machine-Aided Translation vs. Machine Translation); Basic
functionalities and special features
d. TWS Practicum
* Role of text features and variance; Role of linguistic aspects
e. Evaluation and Discussion of TWS: Language Coverage & Ergonomic Issues

SCIENTIFIC-TECHNICAL TRANSLATIONS: COURSE 2

A. Setting up the translation project

* Previewing translation project: Introduction to translation management; Establishing
terminology needs; Creating terminology and dictionary databases; Coordination and
management of translator's tasks; Introduction to the Pedagogical TWS (= PTWS)
B. Performing the translation project: Pre-Translation Stage
* Time management; Resource management; Task hierarchy management; Group assignments
and group/task coordination; Training in the use of PTWS (continued); Preparation of
project-wide terminology bank
C. Document production: Translation Stage
* Translation of document; Desktop Publishing skills applicable to translation task; On-going
terminology and translation archive management using the PTWS
D. Post-Editing of Translation Document
* Assessing editing needs; Coordination and management of post-editing tasks; Post-editing
using PTWS of both human and machine output
E. Maintenance of MT Systems
* Introduction to LAI; Updating English lexicons; Updating German lexicons
F. Discussion and Evaluation
* Set-up stage, Pre-translation stage; Document production; Post-editing stage; Overall course
and courseware

Development of the Pedagogical TWS (PTWS)

The previous section identified three major areas in which translation instruc­
tion must be modernized. The current TWS supports most of the activities con-
 314 Teaching Translation and Interpreting 2

nected with these tasks. For instance, TWS is used to simulate post-editing by
manually presenting the original in the left-hand-side window of the TWS editor
and a translation (not necessarily generated by a computer) in the right-hand-side
window. In order to implement TWS in the translation curriculum and to provide
adequate technical support for teaching the above skills, the following three
developments are presently underway to transform the TWS into the PTWS:
■ a translation-by-example facility to help in teaching post-editing,
■ a translation archive database to support the translation team management
task, and
■ a lexicon acquisition interface for training in maintaing and updating MT
systems.

Figure 1: The different components of the

Pedagogical Translators Work-Station

At present, most work related to system maintenance and MAT carried out by
translation students centers on the following system components: statistical text
analysis tools, augmentor, lexicon acquisition interface, glossary acquisition,
 Dieter Wältermann, USA 315

Ontos, and user glossaries. The system environments with which students perform
most of the system-related work concentrate on the three developments outlined
above.
The translation-by-example environment increases the efficiency of translators
and translation editors by providing candidate translations based on identity or
similarity of current text material to previously translated text material. This
module also relies on the availability of a large multilingual corpus of texts in
which passages in the source and target language are aligned. Such an environ­
ment is important because in many practical translation tasks, documents (e.g.,
manuals and handbooks) which have to be translated into a number of languages
undergo regular changes such as revisions and updates.
The translation archive is but one instrument that is used both for Translation-
By-Example (= TBE) and for instruction in translation management. This facility
is constantly updated with new translations, a task which is carried out primarily
by the translation students since their translation assignments are added to the
translation archive. An alignment algorithm for translation passages in each text
is currently under development at the Center to increase the efficiency of the
TBE and the Translation Archive (= TA). The TWS interface is used in both in­
stances, so that very little interface-related work is involved. To train translators
in the maintenance and update of MT systems, it is necessary to teach them the
skills needed for updating the static knowledge sources of an MT system - such
as the system-oriented lexicon, grammars and (for knowledge-based machine
translation) domain models. In the framework of the CMU program, researchers
offer translation students training in lexicon building. The technological structure
for this task is provided by the Lexicon Acquisition Interface (= LAI), consisting
of a set of questionnaires organized in decision trees and presented to the user
by means of TWS dialog boxes and an entry browser window. The questionnaires
guide the user through the list of morphological, syntactic, semantic and other
properties required for a particular lexicon entry format. The system records in­
formation entered by the user in a specified formalism and displays the appro­
priate lexicon entry in a browser window. Although a version of LAI exists in
the current TWS, it supports the acquisition of lexicons only for English and is
supposed to be used exclusively by expert computational lexicographers. CMT
intends to develop a version devoted to German and a version of the interface
intended for non-expert users. The statistical text analysis tools provide students
with a concordance facility which can be used to look up the translation of any
given expression while specifying the depth or narrowness of the context (up to
7 words before and after the term in question). The search results are displayed
in a special window, and the students can view each occurrence found in the
316 Teaching Translation and Interpreting 2

database in its full context in a separate window, thereby examining the different
or identical translation decisions for any given word or multi-word expression.
Most text material used in both translation courses comes from state-of-the-art
documentation in scientific and technical fields. Some of the texts may present
material which is not only innovative but also challenging linguistically so that
students may encounter new terms which are not yet present in the MT diction­
aries and customized user glossaries, and which may not be found in any general
dictionary or termbank. Whenever students encounter such terms or expressions
during the translation task, they can be added to a user glossary by applying the
menu option 'ADD-A-TERM'. The glossary acquisition tool invokes the appro­
priate user glossary (e.g. engineering or electronics glossary) and automatically
places the user at the correct point within the glossary for the term in question.
Thus, students can build task-specific or domain-specific glossaries in consulta­
tion with on-line reference materials, fellow students, and the instructor.

The Pedagogical Translator's Workstation in the translation instruction

curriculum
The PTWS provides the framework for the second course in the new translation
program. This course centers on a real-life translation task which will be spon­
sored by two translation agencies, one in Pittsburgh, Pennsylvania, and the other
in Austin, Texas. By providing an interaction with an industrial partner students
will become directly involved with the needs and difficulties provided by actual
translation tasks. The translation survey showed that several of the major trans­
lation programs offer translation projects in which students interact with an advi­
sor, either a faculty member or a translator affiliated with the individual
institution.
Our translation project differs from previous ones by allowing students to per­
form actual translation work in collaboration with a translation agency and to use
MAT tools throughout the translation and editing stages of the translation pro­
cess. In addition, this project differs with respect to the overall translation task:
instead of offering individual translation projects, the CMU course involves all
students in all the various tasks associated with the production of a single large
translation project, thereby giving all students a chance to learn firsthand about
each stage of a real-life translation project. The students will not only benefit
from translating a larger document, but also learn the various aspects involved
in coordinating several translators working on a single project. During the whole
project, the PTWS is the common denominator to all tasks and to all partici­
pants.2
 Dieter Wältermann, USA 317

Notes
1. TWS is implemented on a UNIX-based workstation using the X-11 window system, the
C-based X-11 toolkit called MOTIF (OSF/MOTIF 1991) and its CommonLisp interface called
CLM (Babatz et al. 1991). MOTIF provides a high-level interface to X-11 by defining various
types of widgets, e.g., windows, scroll bars, menus, and buttons, and it also allows multitasking.
Finally, CLM uses a control-flow discipline known as callbacks which help to enforce good
system modularity. Before deciding to use the above environment, other substrates had been taken
into consideration, such as GARNET (Myers et al. 1990) and SERPENT (Bass et al. 1990).
2. I would like to acknowledge the help and support of my colleague, G. Richard Tucker, whose
comments on earlier versions of this paper have been extremely valuable. I would also like to
thank the participants of the Second 'Language International' Conference - their comments and
suggestions in response to the presentation of a shorter version of this paper have been very
helpful. Finally, I owe a great debt of gratitude to my friend, Esther Sydow, for her suggestions,
her time and her endless patience.

Jens Bangs Stenhus i Ålborg

WORKS CITED
 WORKS CITED
Editors' notes
The present list is compiled from individual contributions to the volume. The
'Works Cited' largely follows the MLA style sheet. Italics are used for titles of
journals, magazines and books. Authors are listed alphabetically by surnames.
Collective books (e.g. Proceedings) are marked with 'In:' before the title and
listed immediately after the article, if there is only one reference in the present
book. In case there are references to two contributions or more, the collective
book is listed alphabetically by its editor(s). 'u', 'a', '5' are listed as Danish 'y',
'æ' and 'ø' at the end of the alphabet.
AIIC. 1982. Practical Guide for Professional Interpreters. Geneva: Association
Internationale des Interprètes de Conférence.
ALBERT, SAND6R. 1993. The traps of formal correspondence. Perspectives:
Studies in Translatology 1, 11-21.
ALEXIEVA, BISTRA. 1985. Semantic Analysis of the Text in Simultaneous
Interpreting. In: BUHLER, HILDEGUND (ed). 1985. Proceedings of the Xth
World Congress of FIT. Vienna: Braumuller. 195-198.
ALLIONI, SERGIO. 1989. Towards a Grammar of Consecutive Interpretation.
In: The Theoretical and Practical Aspects of Teaching Conference Inter-
pretation. Udine (Italy): Campanotto. 191-197.
ANDERSEN, ERIK SK0TT & CLAUS DETLEF & JENS RAAHAUGE. 1991.
Sœt ord på din verden. Copenhagen: Dansklærerforeningen.
ANDERSON, R.B.W. 1976. Perspectives on the role of interpreter. In: BRISLIN,
RICHARD W. (ed). 1976.
ARNTZ, REINER & HERIBERT PICHT. 1991. Einfuhrung in die Terminologie-
arbeit. Hildesheim & Zurich & New York: Georg Olms Verlag.
BABATZ, R. & A. BACKER & C. BEILKEU & T. BERLAGE & M. SPEN-
KE. 1991. CLM: A Language Binding for Common Lisp and OSF/Motif User
Guide and Reference Manual Version 2.0. Technical report. German National
Research Center for Computer Science.
BAKER, R.G. & A. D. LAMBOURNE & G. ROWSTON. 1984. Handbook for
Television Subtitlers. IBA/Oracle Ltd, Southampton University.
BARADUC, JEAN. 1972. La dénotation dans les annonces publicitaires. Commu-
nications et langages 14, 105-115.
BARTLETT, FREDERICK C. 1932. Remembering. A Study in Experimental and
Social Psychology. Cambridge: Cambridge University Press.
BASS, L. & B. CLAPPER & E. HARDY & R. KAZMAN & R. SEACORD.
1990. Serpent: A User Interface Management System. Proceedings of the
Winter 1990 USENIX Conference.
BASSNETT-McGUIRE, SUSAN. 1980. Translation Studies. London: Methuen.
322 Teaching Translation and Interpreting 2

BASSNETT-McGUIRE, SUSAN. 1988 & 1991. [Revised Edition]. Translation

Studies. London & New York: Routledge.
BAUMGARTNER, PETER. 1990. Sprachmittel zum Ausdruck des Zwecks in
der englischen Fachsprache der Elektrotechnik. In: REKTORAT der Fach-
hochschule Flensburg (eds). 1990. Fachhochschule Flensburg - Forschung und
Technologietransfer. 7-16.
BAUMGARTNER, PETER & ROLAND KRAUS. 1992. Der Sachverhalt "Auf-
wand" und seine sprachliche Realisierung in deutschen und englischen techni-
schen Fachtexten. Lebende Sprachen 38, # 2, 57-60.
BEAUGRANDE, ROBERT de. 1978. Factors in a Theory of Poetic Translating.
Assen: Van Gorcam.
BEAUGRANDE, ROBERT de. 1980. Text, Discourse and Process. Toward a
Multidisciplinary Science of Texts. Norwood, New Jersey: Ablex.
BEAUGRANDE, ROBERT de & WOLFGANG DRESSLER. 1981. Introduction
to Text Linguistics. London: Longman.
BEAUGRANDE, ROBERT de. 1984. Text Production. Toward a Science of
Composition. Norwood, New Jersey: Ablex.
BELL, ROGER T. 1991. Translation and Translating. Theory and Practice.
London: Longman.
BENJAMIN, WALTER. 1977 [1968]. The Task of the Translator. In:
ARENDT, HANNAH (ed). 1977. Illuminations. Translated by Harry Zon. New
York: Schocken Books. 69-82.
BENTON, P. M. 1989. A practical test of machine assisted translation systems
and agencies. In: HAMMOND, DEANNE L. (ed). 1989. 443-448.
BERGER, PETER L. & THOMAS LUCKMANN. 1989 [orig. 1967]. The Social
Construction of Reality. A Treatise in the Sociology of Knowledge. Garden
City: Doubleday.
BERNE, ERIC. 1964. Games People Play. New York: Ballantine Books.
BOHLEN, CHARLES E. 1973. Witness to History 1929-1969. New York:
W.W. Norton.
BOWEN, DAVID AND MARGARETA. 1989. Aptitude for Interpreting. In:
GRAN, LAURA & JOHN DODDS (eds). 1989. 109-125.
BOWEN, DAVID AND MARGARETA (eds). 1990. Interpreting - Yesterday,
Today and Tomorrow (= American Translators Association Series on
Translation and Interpretation IV). Binghamton: State University of New York.
BOWEN, MARGARETA. 1989. Language Learning Before Translator/Interpre­
ter Training. In: KRAWUTSCHKE, PETER W. (ed). 1989. 51-64.
BRISLIN, RICHARD W. (ed). 1976. Translation: Applications and Research.
New York: Gardner Press.
BRITISH ASSOCIATION OF COMMUNITY INTERPRETERS. 1989. Guide
to Good Practice. Cambridge.
BROWN G. & G. YULE 1983. Discourse Analysis. Cambridge: Cambridge
University Press.
BRUNDAGE, D.H. & D. MACKERACHER. 1980. Adult Learning Principles
and Their Application to Programme Planning. Ontario: Ontario Institute for
 Works Cited 323

Studies in Education. 21-31.

BUTTERFIELD, HERBERT. 1955. Man on his Past. Cambridge: Cambridge
University Press.
BUHLER, HILDEGUND. 1986. Linguistic (semantic) and extra-linguistic (prag­
matic) criteria for the evaluation of conference interpretation and interpreters.
Multilingua 5, 231-235.
BUHLER, HANNS HERMANN. 1987. Die Diktatur des Dolmetschers? Bemer-
kungen zu den "Grundregeln für das Dolmetschen". Mitteilungsblatt des
Österreichischen Übersetzer- und Dolmetscherverbandes 'Universitas', # 2,
15-18.

CAILLE, P.F. 1960. Cinéma et traduction. Babel 6, 103-109

CARR, EDWARD HALLET. 1962. What is History? New York: Alfred A.
Knopf.
CARTELLIERI, CLAUS. 1983. The inescapable dilemma. Quality and/or quan­
tity in interpreting. Babel 29, 209-213.
CARY, EDMOND. 1956. La traduction dans le monde moderne. Geneva: Li-
brairie de 1'Université.
CASSE, PIERRE. 1981 [2nd ed]. Training for the Cross-Cultural Mind: A Hand-
book for Cross-Cultural Trainers and Consultants. Washington, DC: Sietar.
CATFORD, J.C. 1965. A Linguistic Theory of Translation. Oxford: Oxford Uni­
versity Press.
CHAFE, WALLACE. 1977. Creativity in verbalisation and its implication for the
nature of stored knowledge. In: FREEDLE. L. O. (ed). 1977.
CHAFE, WALLACE. 1987. Cognitive Constraints on Information Row. In:
TOMLIN, R.S. (ed). 1987. 21-51.
CHAO Y. R. 1976. Dimensions of Fidelity in Translation, with special reference
to Chinese. In: Aspects of Chinese Sociolinguistics. Essays by Yuan Ren Chao.
Selected and introduced by Anwar S. Dil. Stanford University Press. 148-169.
CHESTERMAN, ANDREW (ed). 1989. Readings in translation theory. Hel­
sinki.
CHESTERMAN, ANDREW. 1993. Theory in Translation Theory. The New
Courant (Department of English, University of Helsinki) 1, 69-79.
CHOMSKY, NOAM. 1965. Aspects of the Theory of Syntax. Cambridge, Massa­
chusetts: M.I.T. Press.
CHUKOVSKY, K. 1984. A High Art. The Art of Translation. Translated by L.
G. Leighton. Knoxville: University of Tennessee Press.
COHEN, A. & P. COUSSEAU & D. GRANNES & C. McNEILLY & S. NI-
RENBURG & P. SHELL & D. WALTERMANN. 1993. Translators Worksta-
tion: User's Manual. Pittsburg: Carnegie Mellon University. School of Com­
puter Science.
COLLINI, STEFAN (ed). 1992. Interpretation and Overinterpretation. Cam­
bridge: Cambridge: Cambridge University Press.
CONNINE, CYNTHIA M. & DAWN G. BLASCO & MICHAEL HALL.
1991.Effect of Subsequent Sentence Context in Auditory Word Recognition:
324 Teaching Translation and Interpreting 2

Temporal and Linguistic Constraints. Journal of Memory and Language 30,

234-250.
CORBETT, EDWARD P. J. 1971 [2nd ed]. Classical Rhetoric for the Modern
Student. New York: Oxford University Press.
CORDER, S.P. 1973. Introducing Applied Linguistics. Penguin Books.
COUGHLIN, J.M. 1988. Artificial Intelligence and Machine Translation: Pre­
sent Developments and Future Prospects. Babel 34, # 1.
CULLER, JONATHAN. 1992. In Defence of Overinterpretation. In: COLLINI,
STEFAN (ed). 1992. 109-123.
CUNNINGHAM, JAMES W. 1987. Toward a Pedagogy of Inferential Compre­
hension and Creative Response. In: TIERNEY, ROBERT J. & al. (eds). 1987.
229-253.
DAMIDA, A. 1977. L'Éducation en Afrique a la lumiere de la conference de
Lagos. Paris: UNESCO.
DAVIDSON, P.M. 1992a. Segmentation of Japanese Source Language Discourse
in Simultaneous Interpretation. The Interpreter's Newsletter. Special Issue 1,
2-11.
DAVIDSON, P.M. 1992b. Simultaneous Interpreting Research, Past, Present
and Future. Interpreting Research (Journal of the Interpreting Research Asso­
ciation of Japan) 3, # 2, 23-44.
DELISLE, JEAN. 1980. L'analyse du discours comme méthode de traduction.
Ottawa: Presses de I'université d'Ottawa.
DELISLE, JEAN. 1981. De la théorie à la pédagogie: réflexions méthodo-
logiques. In: DELISLE, JEAN (ed) 1981. 135-152.
DELISLE, JEAN (ed). 1981. L'enseignement de la traduction et de l'interpré-
tation. Ottawa: Éditions de l'université d'Ottawa.
DELISLE, JEAN. 1991. Projet d'Histoire Thématique de la Traduction. In:
JOVANOVIC, M. (ed). 1991. Proceedings of the 12th World Congress of FIT.
Belgrade: Prevodilac. 63-68.
DERASSE, BERNARD. 1987. Dubbers and subtitlers have a prime role in the
international distribution of television programmes. EBU Review 38, # 6, 8-13
DETWEILER, RICHARD A. 1980. Intercultural Interaction and the Cat­
egorization Process: A Conceptual Analysis and Behavioral Outcome. Inter-
national Journal of Intercultural Relations 4, 275-293.
DOI, N. & K. FURUKUWA & K. FUCHI. 1987. Fifth-generation Computer
Systems and Their Impact on Society. Impact of Science on Society. No. 146.
UNESCO/Taylor & Francis.
DOLLERUP, CAY. 1978. Omkring sproglig transmission (= Anglica et Amer
icana 3). Copenhagen: Department of English, University of Copenhagen.
DOLLERUP, CAY. 1982. An analysis of some mechanisms and strategies in the
translation process based on a study of translations between Danish and
English. The [Incorporated] Linguist 21, 162-169.
DOLLERUP, CAY & ANNE LODDEGAARD (eds). 1992. Teaching Trans-
lation and Interpreting. Training, Talent and Experience. Amsterdam & Phila-
 Works Cited 325

delphia: John Benjamins.

DOLLERUP, CAY. 1993. Interlingual transfers and issues in translatology.
Perspectives: Studies in Translatology 1, 137-154.
DOLLERUP, CAY & IVEN REVENTLOW & CARSTEN ROSENBERG
HANSEN. 1993. Identity in Practical Translation: Conducting Cross-cultural
Studies. Le Langage et l'homme (Brussels) 28, 11-25.
DOLLMANN, EUGEN. 1967. The Interpreter. Memoirs of Dr. Eugen
Dollmann. Translated from the German by J. Maxwell Brownjohn. London:
Hutchinson.
DOWNING, BRUCE T. & KATE HELMS TILLERY. 1992. Professional
Training for Community Interpreters. Minneapolis: Center for Urban and Reg­
ional Affairs.
DUMBLETON, C.W. 1982. Language Services in West Africa. Geneva:
UNCTAD/ECDC/140.
ECO, UMBERTO. 1990. The limits of interpretation. Bloomington and Indian­
apolis: Indiana University Press.
ECO, UMBERTO. 1992a. Interpretation and Overinterpretation. In: COLLINI,
STEFAN (ed). 1992.
ECO, UMBERTO. 1992b. Les limites de l'interpretation. Paris: Grasset.
EKVALL, ROBERT B. 1960. Faithful Echo. New York: Twayne Publishers.
ERVIN, S. & C. OSGOOD. 1954. Second language learning and bilingualism.
Journal of Abnormal Social Psychology 49, 139-146.
FAWCETT, PETER. 1983. Translation modes and constraints. The Incorporated
Linguist 22, 186-190
FURNHAM, ADRIAN & STEPHEN BOCHNER 1986. Culture Shock. Psycho-
logical Reactions to Unfamiliar Environments. London & New York.

GEMAR, JEAN-CLAUDE. 1983. De la pratique à la théorie, l'apport des

praticiens à la théorie générale de la traduction. Meta 28, 323-333.
GENTILE, ADOLFO. 1991. The Application of Theoretical Constructs from a
Number of Disciplines for the Development of a Methodology of Teaching in
Interpreting and Translating. Meta 36, 344-351.
GERVER, D. & W. H. SINAIKO (eds). 1978. Language Interpretation and
Communication. New York: Plenum Press.
GILE, DANIEL. 1983. Aspects méthodologiques de l'évaluation de la qualité du
travail en interprétation simultanée. Meta 28, 236-243.
GILE, DANIEL. 1985. La sensibilité aux écarts de langue et la sélection d'in-
formateurs dans l'analyse d'erreurs. The Incorporated Linguist 24, 29-32.
GILE, DANIEL. 1990a. Basic Concepts and Models for Conference Inter-
pretation Training. Paris: INALCO & CEEI (ISIT).
GILE, D. 1990b. Scientific Research vs. Personal Theories in the Investigation
of Interpretation. Paper given at the Scuola Superiore di Lingue Moderne per
Interpreti e Traduttori, Università degli Studi, Trieste.
326 Teaching Translation and Interpreting 2

GILE, DANIEL. 1990c. L'évaluation de la qualité de l'interprétation par les

dé1égués: une étude de cas. The Interpreters' Newsletter 3, 66-71.
GILE, DANIEL. 1991a. The processing capacity issue in conference interpreta­
tion. Babel 37, 15-27.
GILE, DANIEL. 1991b. A Communication-Oriented Analysis of Quality in Non-
literary Translation and Interpretation. In: LARSON, MILDRED L. (ed). 1991.
188-200.
GILE, DANIEL. 1992a. Les fautes de traduction: une analyse pédagogique.
Meta 37, 251-262.
GILE, DANIEL. 1992b. Basic theoretical components in interpreter and transla­
tion training. In: DOLLERUP, CAY & ANNE LODDEGAARD (eds). 1992.
185-193.
GOMES DE MATOS, FRANCISCO. 1991. Human Rights Applied to Transla­
tions. In: LARSON, MILDRED L. (ed). 1991. 254-259.
GOTTLIEB, HENRIK. 1992. Subtitling - A new university discipline. In: DOL­
LERUP, CAY & ANNE LODDEGAARD (eds). 1992. 161-170.
GOTTLIEB, Henrik. 1994. Diagonal Translation. Perspectives: Studies in Trans-
latology 2, 101-123.
GOTTLIEB, HENRIK. Forthcoming. Literature on interlingual subtitling since
1932 (= Copenhagen Studies in Translation 4).
GRAHAM, JOSEPH F. 1985. Difference in Translation. Ithaca: Cornell Uni­
versity Press.
GRAN, LAURA. 1988. Interaction between Memory and Note-Taking in Conse­
cutive Interpretation. In: Ubersetzungswissenschaft und Sprachmittleraus-
bildung, Proceedings. Berlin: Humboldt Universitat.
GRAN, LAURA & JOHN DODDS (eds). 1989. The Theoretical and Practical
Aspects of Teaching Conference Interpreting. Udine (Italy): Campanotto.
GRICE, H.P. 1975. Logic and conversation. In: COLE, PETER & JERRY
MORGAN (eds). 1975. Syntax and Semantics, Speech Acts. New York:
Academic Press. 41-58.
GROSMAN, META. Forthcoming. Huckleberry Finn for Non-American
Readers. Proceedings of the Conference "American Literature for Non-
American Readers". Bellagio, June 1-5 1992.
GUMPERZ, JOHN J. 1982. Discourse Strategies (= Studies in Interactional So-
ciolinguistics I). Cambridge: Cambridge University Press.
GUTT, ERNST-AUGUST. 1990. A Theoretical Account of Translation - Without
a Translation Theory, Target 2, 135-164.
GUTT, ERNST-AUGUST. 1991. Translation and Relevance: Cognition and Con-
text. Oxford: Basil Blackwell.
GOHRING, HEINZ. 1978. Interkulturelle Kommunikation: Die Uberwindung
der Trennung von Fremdsprachen- und Landeskundeunterricht durch einen inte-
grierten Fremdverhaltensunterricht. In: Kongressberichte der 8. Jahrestagung
der GAL. Stuttgart. 4.9-14.
GOHRING, HEINZ. 1980. Deutsch als Fremdsprache und interkulturelle Kom­
munikation. In: WIERLACHER, ALOIS (ed). Fremdsprache Deutsch. Grund-
 Works Cited 327

lagen und Verfahrender Germanistik als Fremdsprachenphilologie I. Munich:

Wilhelm Fink. 70-90.

HACK, ANNE-CAROLINE. 1992. Interferenzen beim Simultandolmetschen.

Heidelberg: Institut fur Ubersetzen und Dolmetschen der Universitat Heidel­
berg. Unpublished diploma thesis.
HALASZ, LASZLO. 1991. Understanding Short Stories: An American-Hungarian
Cross-Cultural Study. Empirical Studies of the Arts 9, 143-163.
HALL, DONALD. 1982. The Weather for Poetry. Michigan: University of
Michigan Press.
HALLIDAY, MICHAEL A.K. 1967. Notes on Transitivity and Theme in English.
Journal of Linguistics 3 (Part 2), 199-244.
HALLIDAY, MICHAEL A.K. & RUQAIYA HASAN. 1976. Cohesion in Eng-
lish. London: Longman.
HALLIDAY, MICHAEL A.K. 1985. An Introduction to Functional Grammar.
London: Edward Arnold.
HAMMOND, DEANNE L. (ed). 1989. Coming of Age. Proceedings of the 30th
Annual Conference of the American Translators Association. Medford, N.J.:
Learned Information.
HARRIS, RICHARD JACKSON & LAWRENCE M. SCHOEN & DEANA L.
HENSLEY. 1992. A Cross-cultural Study of Story Memory. Journal of Cross-
cultural Psychology 23, 133-147.
HATIM, BASIL & IAN MASON. 1990. Discourse and the Translator. London:
Longman.
HEWSON, LANCE & JACKY MARTIN. 1991. Redefining Translation. The
Variational Approach. London & New York: Routledge.
HILLOCKS, Jr., GEORGE. 1986. Research on Written Composition. Urbana,
Illinois: NCTE/ERIC.
HILVERKUS, BARBARA. 1991. Monitoringstrategien beim Simultandolmet-
schen. Heidelberg: Institut für Übersetzen und Dolmetschen der Universitat
Heidelberg. Unpublished diploma thesis.
HOFSTADTER, DOUGLAS R. 1980. Gödel, Escher, Bach: An Eternal Golden
Braid. A Metaphysical Fugue on Minds and Machine in the Spirit of Lewis
Carroll. Middlesex: Penguin.
HOHNHOLD, INGO. 1990. Ubersetzungsorientierte Terminologiearbeit. Stutt­
gart: InTra.
HOLMES, JAMES S. 1988. Translated! Papers on Literary Translation and
Translation Studies. Amsterdam: Rodopi.
HOLZ-MANTTARI, JUSTA. 1984. Translatorisches Handeln. Theorie und Me-
thode (= Annales Academiae Scientiarum Fennicae B 226). Helsinki: Suoma-
lainen Tiedeakatemia.
HOLZ-MANTTARI, JUSTA. 1986. Translatorisches Handeln - theoretisch fun-
dierte Berufsprofile. In: SNELL-HORNBY, MARY (ed). Übersetzungswissen-
schaft - eine Neuorientierung. Zur Integrierung von Theorie und Praxis.
Tubingen: Francke (= Uni-Taschenbucher 1415).
328 Teaching Translation and Interpreting 2

HOLZ-MÄNTTÄRI, JUSTA & CHRISTIANE NORD (eds). 1993. Traducere

navem. Festschrift für Katharina Reiss zum 70 Geburtstag. Tampere: Tampere
University Press. Distributed by University of Tampere, Sales Office, P.O. Box
617, SF-33101 Tampere/Finland, Fax: +358 31 157 150.
van HOOF, HENRI. 1986. Petite histoire de la traduction en Occident. Lou-
vain-la-Neuve: Cabay.
HUTCHINS, W. JOHN & HAROLD L. SOMERS. 1992. An Introduction to
Machine Translation. London: Academic Press.
HYMES, D. 1977. Models of Interaction of Language and Social Life. In:
GUMPERZ, JOHN & DELL HYMES (eds). 1977. Directions in Sociolinguist-
ics: the ethnography of communication. New York: Holt, Rinehart and
Winston, 35-71.
HÖNIG, HANS G. 1992. Von der erzwungenen Selbstentfremdung des Uber-
setzers - Ein offener Brief an Justa Holz-Mänttäri. TEXTconTEXT 7, 1-14.

IBRAHIM, HASNAH. 1992. OH BABEL! Problems of translating Malay verse

into English. Unpublished PhD dissertation.
IBRAHIM, HASNAH. 1991. Translation Spectrum: Problem in Description. In:
NOOR, NOOREIN MOHD & ATIAH HJ. SALLEH (eds). Pragmatik Penter-
jemahan. Prinsip, amalan dan penilaian menuju ke abad 21 [The Pragmatics
of Translation: Principles, Practice and Evaluation. Moving towards the 21st
Century]. Kuala Lumpur.
IHENACHO, A. 1991. Planning translation studies in Nigeria. In: ANYAEHIE,
E. & A. IHENACHO (eds). Language studies and relevance. Uturu: LC
Publications.
ILG, GERARD. 1980. L'interprétation consecutive. Paralleles (Geneva) 3,
109-136.
ILG, GERARD. 1982. L'interprétation consecutive. La pratique. Paralleles
(Geneva) 5, 91-109.
INHELDER, B. & J. PIAGET. 1966. La psychologic de l'enfant. Paris: Presses
Universitaires de France.
IVIR, VLADIMIR. 1989. Procedures and Strategies for the Translation of Cul­
ture. In: TOURY, GIDEON (ed). 1989. 35-46.

JACKENDOFF, RAY. 1985. Semantics and Cognition. Cambridge, Massa-

chusetts: M.I.T. Press.
JACOBSEN, HENRIK GALBERG & PEDER SKYUM-NIELSEN. 1988 [2nd
edition]. Erhvervsdansk. Copenhagen: Sch0nberg.
JAKOBSEN, ARNT LYKKE. 1993. Translation as textual (re)production. Per-
spectives: Studies in Translatology 1, 155-166.
JAKOBSON, ROMAN. 1959. On Linguistic Aspects of Translation. In:
BROWER, R. A. (ed). On Translation. New York: Harvard University Press.
232-239.
JACOBY, LARRY L.A. 1991. Process Dissociation Framework: Separating Auto­
matic from Intentional Uses of Memory. Journal of Memory and Language 30,
 Works Cited 329

513-541.
JIASHUI, WU S. 1991. Task-Oriented and Comprehensive Training of Transla­
tors and Interpreters. In: Proceedings, Asia-Pacific Conference on Translation
and Interpreting: Bridging East and West.
JUMPELT, R. W. 1961. Die Ubersetzung naturwissenschaftlicher und techni-
scher Literatur. Berlin-Schoneberg: Langenscheidt.
KADE, OTTO. 1968. Zufall und Gesetzmäβigkeit in der Übersetzung. (=
Beihefte zur Zeitschrift Fremdsprachen I). Leipzig: VEB Verlag Enzyklopadie.
KALINA, SYLVIA. 1991. Zur Rolle der Theorie in der Dolmetscherausbildung.
TEXTconTEXT 6, # 2/3, 101-113.
KALINA, SYLVIA. 1992. Discourse processing and interpreting strategies - an
approach to the teaching of interpreting. In: DOLLERUP, CAY & ANNE
LODDEGAARD (eds). 1992. 251-257.
KANG, HEE-WON, 1992. Cultural Inference in Second Language Reading.
International Journal of Applied Linguistics 2, 95-119.
KARKER, ALLAN. 1993. Dansk i EF - en situationsberetning om sproget (=
Nordisk SpråksekretariatsSkrifter 16). Copenhagen & Oslo: Gad.
KEMPSON, RUTH (ed). 1988. Mental Representations: the Interface between
Language and Reality. Cambridge: Cambridge University Press.
KENDON, ADAM (ed). 1981. Nonverbal Communication, Interaction, and Ges-
ture. Selections from Semiotica (= Approaches to Semiotics 41). The Hague:
Mouton.
KENDON, ADAM. 1982. The organization of behavior in face-to-face interac­
tion: observations on the development of a methodology. In: SCHERER,
KLAUS R. & PAUL EKMAN (eds). 1982. Handbook of Methods in Nonverbal
Behavior Research (= Studies in Emotion and Social Interaction). Cambridge:
Cambridge University Press. 440-505.
KING, JONATHAN & MARCEL ADAM JUST. 1991. Individual Differences
in Syntactic Processing: The Role of Working Memory. Journal of Memory
and Language 30, 580-602.
KINTSCH, WALTER & EDITH GREEN. 1978. The Role of Culture-Specific
Schemata in the Comprehension and Recall of Stories. Discourse Processes 1,
1-13.
KITTREDGE, R. 1983. Sublanguage - Specific Computer Aids to Translation -
a survey of the most promising areas. Université de Montréal & Bureau des
Traductions.
KITTREDGE, R. & R. GRISHMAN (eds). 1986. Analysing Language in Re­
stricted Domains: Sublanguage Descriptions and Processing. Hillsdale, N.J.:
Lawrence Erlbaum.
KOCK, CHRISTIAN & BIRTHE TANDRUP. 1989. Skriv kreativt. Copenhagen:
Gyldendal.
KOHN, KURT. 1990. Dimensionen lernersprachlicher Performanz. Theoretische
und empirische Untersuchungen zum Zweitsprachenerwerb. Tubingen: Gunter
Narr.
330 Teaching Translation and Interpreting 2

KOLLER, WERNER. 1979. Einführung in die Übersetzungswissenschaft.

Heidelberg: Quelle & Meyer (= Uni-Taschenbucher 819)
KOLLER, WERNER. 1993. Zum Begriff der 'eigentlichen' Ubersetzung. In:
HOLZ-MÄNTTÄRI, JUSTA & NORD, CHRISTIANE (eds). 1993. 49-63.
KOMISSAROV, VILEN. 1985. The practical value of translation theory. Babel
31, 208-212.
KOPCZYNSKI, ANDRZEJ. 1994. Quality in conference interpreting: Some prag­
matic problems. In: SNELL-HORNBY, MARY & FRANZ POCHHACKER &
KLAUS KAINDL(ed). 1994. 189-198.
KOVAČIČ, IRENA. 1992. Jezikoslovni pogled na podnaslovno prevajanje tele-
vizijskih oddaj. [A linguistic approach to subtitling television programs].
Ljubljana: University of Ljubljana. PhD dissertation.
KRAWUTSCHKE, PETER W. (ed). 1989. Translator and Interpreter Training
and Foreign Language Pedagogy (= American Translator Association Series
III). Binghamton: State University of New York.
KRISHNAMOORTHY, K. 1985. Dhvani or Suggestion: A Study in Perspective.
Indian Literature. July-August, 113-122 .
KRUSCHE, DIETRICH. 1990. Vermittlungsrelevanten Eigenschaften literari-
scher Texte (1985). In: KRUSCHE, DIETRICH & ALOIS WIERLACHER
(eds). 1990. Hermeneutik der Fremde. Munich: Iudicum. 103-126.
KUBLER, C. 1985. A Study of Europeanized Grammar in Modern Written
Chinese. Taipei: Student Book Co.
KURZ, INGRID. 1989. Conference Interpreting: User Expectations. In:
HAMMOND, DEANNE L. (ed). 1989. 143-148.
KURZ, INGRID. 1992. Simultandolmetschen als Gegenstand der interdisziplin-
ären Forschung. Vienna: University of Vienna. Unpublished Habilitations-
schrift.
KYLE, K. 1992. Switched On: Deaf people's views on television subtitling. A re­
port for the ITC and BBC by the Centre for Deaf Studies, University of
Bristol.

LADMIRAL, J.-R. 1990. La traduction proligère? - Sur le statut des textes

qu'on traduit. Meta 35, # 1.
LAKOFF, GEORGE. 1987. Women, Fire, and Dangerous Things. What Cat-
egories Reveal about the Mind. Chicago: University of Chicago Press.
LAMBERT S. 1988. Information Processing among Conference Interpreters: A
Test of the Depth-of-Processing Hypothesis. Meta 33, 377-387.
LAMBERT, W. E. & J. HAVELKA. 1958. The Influence of language acquisition
contexts on bilingualism. Journal of Abnormal Social Psychology 56, 239-244.
LAROSE, ROBERT. 1992. La théorie de la traduction: à quoi ça sert? Meta 37,
405-407.
LARSON, MILDRED L. 1984. Meaning-based Translation: A Guide to Cross-
Language Equivalence. Lanham: University Press of America.
LARSON, MILDRED L. (ed). 1991. Translation: Theory and Practice. Tension
and Interdependence (= American Translators Association Series V).
 Works Cited 331

Binghamton, New York: State University of New York.

LASZLO, JANOS & STEEN FOLKE LARSEN. 1991. Cultural and Text Vari­
ables in Processing Personal Experiences While Reading Literature. Empirical
Studies of the Arts 9, 23-34.
LE FEAL, K.D. 1992. Linguistique et traduction. TURJUMAN 1, # 1.
LEDERER, MARIANNE. 1976. Synecdoque et traduction. Etudes de Lingui-
stique Appliquee 24.
LEDERER, MARIANNE. 1978. Simultaneous Interpretation: Units of Meaning
and Other Features. In: GERVER, D. & W. H. SINAIKO (eds). 1978. 323-332.
LEDERER, MARIANNE. 1981. La traduction simultanee - experience et
theorie. Paris: Minard-Lettres Modernes.
LEDERER, MARIANNE. 1992. Principes et méthodes de l'enseignement de
1'interprétation. TURJUMAN 1, # 1.
LEECH, GEOFFREY. 1983. Principles of Pragmatics. London: Longman.
LEFEVERE, ANDRE. 1977. Translating Literature: The German Tradition.
Assen: Van Gorcum.
LENNEBERG, E. H. 1967. Biological foundations of language. New York:
Wiley & Sons.
LIND, JENS N0RMARK & MORTEN SESTOFT. 1992. Tv-tekstning af "El
Quijote". Term paper, Department for Romance Languages, University of
Copenhagen.
LIU, C.C. (ed). 1991. Fanyi xin lun ji. Kong Kong: Commercial Press.
LOSA, EDITH F. (ed). 1993. New Frontiers. Proceedings of the 33rd Annual
Conference of the American Translators Association. Medford, New Jersey:
Learned Information.
LOTMAN, JURI & B.A. USPENSKY. 1978. On the Semiotic Mechanism of
Culture. New Literary History 11, 211-232.
LUYKEN, G.-M. & THOMAS HERBST & JO LANGHAM-BROWN &
HELENE REID & HERMAN SPINHOF. 1991. Overcoming language barriers
in television. Dusseldorf: The European Institute for the Media.
LVOVSKAJA, Z. D. 1985. Teoreticheskie problemy perevoda. Moscow: Vy-
shaja Shkola.
MACKINTOSH J. 1991. Language Enhancement for Interpreters. In: LIU, C.
C. (ed). 1991. 389-402.
MADARIAGA, SALVADOR DE. 1974. Morning Without Noon. Memoirs. West-
mead, Farnborough, Hampshire: Saxon House, D.C. Heath Ltd.
MAHMOODZADEH, KAMBIZ. 1992. Consecutive interpreting: its principles
and techniques. In: DOLLERUP, CAY & ANNE LODDEGAARD (eds). 1992.
231-236.
MALONE, JOSEPH L. 1986. Trajectional Analysis: Five Cases in Point. Babel
13.
MANZOUFAS, MARENA. 1982. The art of the subtitle: The subtitling unit at
Channel 0/28. Media Information Australia # 25, 17-18; 25.
MARKUS, MANFRED. 1992. Rhythm, stress and intonation in English and
332 Teaching Translation and Interpreting 2

German seen contrastively. In: MAIR, CHRISTIAN & MANFRED MARKUS

(eds). New Departures in Contrastive Linguistics. Innsbruck: Universität
Innsbruck. I, 21-36.
MARTINS, MARCIA DO AMARAL PEIXOTO. 1992. For a Process-oriented
Methodology in Translation Teaching. Paper given at the Translation Studies
Congress: "Translation Studies - An Interdiscipline", 9-12 September 1992,
Vienna, Austria.
MAYORAL, ROBERTO & DOROTHY KELLY & NATIVIDAD GALLARDO.
1988. Concept of constrained translation. Non-linguistic perspectives of trans­
lation. Meta 33, 356-367.
McDAVID, JOHN & HERBERT HARARI. 1967. Social Psychology. New
York: Harper & Row.
MEAK, LIDIA. 1990. Interprétation simultanée et congrès médical: attentes et
commentaires. The Interpreters' Newsletter 3, 8-13.
MINNESOTA DEPARTMENT OF HUMAN SERVICES, OFFICE OF REFU
GEE MENTAL HEALTH SERVICES. 1988. Assessment of Refugee Mental
Health Needs, Part 2: Survey of Menial Health Service Providers. St. Paul:
Department of Human Services.
MOSER, B. 1978. Simultaneous Interpretation: A Hypothetical Model and its
Practical Application. In: GERVER, D. & W. H. SINAIKO (eds). 1978.
353-368.
MOSER, B. 1991. Paradigms Gained, or the Art of Productive Disagreement.
Bulletin XIX-2 (Association Internationale des Interprètes de Conférence,
Geneva).
MOSKOWITZ, KEN. 1979. Subtitles vs. dubbing: Debate goes on. Take One
(Toronto) 7, # 5. 58.
MOSSOP, BRIAN. 1989. "Write idiomatically and translate ideas not words".
Three defects of the prevailing doctrine of translation. In: SEGUINOT,
CANDACE (ed). The translation process. H.G. Publications, School of
Translation, York University, Canada. 1-20.
MOUNIN, GEORGES. 1965. Teoriae Storia della Traduzione. Torino: Einaudi.
MOUNIN, GEORGES. 1967. Die Ubersetzung. Geschichte, Theorie, Anwen-
dung. Munich: Nymphenburger Verlag.
MOUNIN, GEORGES. 1976. Linguistique et traduction. Brussels.
MYERS SCOTTON, C. 1983. The negotiation of identities in conversation: a
theory of markedness and code choice. International Journal of the Sociology
of Language AA, 115-136.
MYERS, B. & D. GIUSE & R. DANNENBERG & B. ZANDEN & D. KOS-
BIE & E. PERVIN & A. MICKISH & P. MARCHAL. 1990. Garnet:
Comprehensive Support for Graphical, Highly-Interactive User Interfaces. IEEE
Computer 23, # 11.
MÜLLER, BERND-DIETRICH. 1980. Zur Logik interkultureller Verstehenspro-
bleme. Jahrbuch Deutsch als Fremdsprache 6, 102-119.
MÜLLER, BERND-DIETRICH. 1986. Interkulturelle Verstehensstrategien - Ver-
gleich und Empathie. In: NEUNER, GERHARD (ed). 1986. Kulturkontraste
 Works Cited 333

im DaF-Unterricht (= Studium Deutsch als Fremdsprache - Sprachdidaktik 5).

33-84.
NEDERGAARD-LARSEN, BIRGIT. 1993. Cultural factors in subtitling. Per-
spectives: Studies in Translatology 1, 207-241.
NEUBERT, A. 1985. Text and Translation. Leipzig: VEB Verlag Enzyklopadie.
NEWMARK, PETER. 1981. Approaches to Translation. Oxford: Pergamon.
NEWMARK, PETER. 1988. A Textbook of Translation. Cambridge & New York:
Prentice Hall.
NEWMARK, PETER. 1991. About Translation. London: Multilingual Matters.
NIDA, EUGENE A. 1945. Linguistics and Ethnology in Translation Problems.
Word 1, 194-208.
NIDA, EUGENE A. 1964. Toward a Science of Translating. Leiden: E. J. Brill.
NIDA, EUGENE A. & C.R. TABER. 1982. The Theory and Practice of Trans-
lation. Leiden: E. J. Brill.
NIDA, EUGENE & J. de WAARD. 1986. From one language to another. Nash­
ville: Thomas Nelson Publishers.
NIDA, EUGENE A. 1990. The role of rhetoric in verbal communications. Babel
36, 143-154.
NINTAI, MOSES. 1993. African Literature in European Languages: Major Feat­
ures and Implications for Translation. In: Proceedings of the 13th FIT World
Congress. London: ITI. I, 564-572.
NIR, RAFAEL. 1984. Linguistic and sociolinguistic problems in the translation
of imported TV films in Israel. International Journal of the Sociology of
Language 48, 81-97.
NORD, CHRISTIANE. 1988. Textanalyse und Ubersetzen. Theoretische Grund-
lagen, Methode und didaktische Anwendung einer ubersetzungsrelevanten Text-
analyse. Heidelberg: Groos.
NORD, CHRISTIANE. 1991. Text Analysis in Translation: Theory, Methodo-
logy, and Didactic Application of a Model for Translation-Oriented Text Ana-
lysis. (= Amsterdamer Publikationen zur Sprache und Literatur 94). Translated
by C. Nord and P. Sparrow. Amsterdam & Atlanta: Rodopi. Translation of
Nord 1988.
NORD, CHRISTIANE. 1992. Text Analysis in Translator Training. In: DOL-
LERUP, CAY & ANNE LODDEGAARD (eds). 1992. 39:48.
NORD, CHRISTIANE. 1993. Einfuhrung in das funktionale Ubersetzen: Am Bei-
spiel von Titeln und Überschriften. Tübingen: Francke (= Uni-Taschenbücher
1734).
NUNAN, DAVID. 1984. The learner-centred curriculum, a study in second lan-
guage teaching. Cambridge: Cambridge University Press.
OITTINEN, RIITTA. 1992. Teaching Translation of Fiction - A Dialogic Point
of view. In: DOLLERUP, CAY & ANNE LODDEGAARD (eds). 1992. 75-80.
OJO, ADE. 1986. The Role of the Translator of African Written Literature in
Inter-cultural Consciousness and Relationships. Meta 31, 291-299.
334 Teaching Translation and Interpreting 2

OSF/Motif Programmer's Guide, Revision 1.1. 1991. Open Software Foundation

Engelwood Cliffs, N.J.: Prentice Hall.

PATTANAYAK, D.P. 1981. Multi-Lingualism and Mother Tongue Education

Delhi: Oxford University Press.
PEDERSEN, VIGGO HJØRNAGER 1988. Essays on Translation. Copenhagen:
Nyt Nordisk Forlag Arnold Busck.
PERREN, G.E. & J. L. M. TRIM (eds). 1971. Applications of Linguistics,
Cambridge: Cambridge University Press.
POPPER, KARL R. 1972. Objective Knowledge. An Evolutionary Approach,
Oxford: Clarendon Press.
POYATOS, FERNANDO. 1983. New Perspectives in Nonverbal Communication.
Studies in Cultural Anthropology, Social Psychology, Linguistics, Literature,
and Semiotics (= Language and Communication Library 5). Oxford: Pergamon.
POYATOS, FERNANDO. 1987. Nonverbal communication in simultaneous and
consecutive interpretation: A theoretical model and new perspectives. TEXT-
conTEXT 2, 73-108.
POYATOS, FERNANDO (ed). 1988. Cross-Cultural Perspectives in Nonverbal
Communication. Toronto & Gottingen: C.J. Hogrefe.
POYATOS, FERNANDO (ed). 1992. Advances in Nonverbal Communication.
Sociocultural, Clinical, Esthetic and Literary Perspectives. Amsterdam &
Philadelphia: John Benjamins.
PYM, ANTHONY. 1992a. Translation and Text Transfer. Frankfurt-am-Main:
Peter Lang.
PYM, ANTHONY. 1992b. The Relations Between Translation and Material Text
Transfer. Target 4, 171-189.
POCHHACKER, FRANZ. 1992a. The role of theory in simultaneous interpret-
ing. In: DOLLERUP, CAY & ANNE LODDEGAARD (eds). 1992. 211-220.
POCHHACKER, FRANZ. 1992b. Simultandolmetschen als komplexes Handeln.
Ein Theorie- und Beschreibungsrahmen, dargestellt an einer Fachkonferenz.
Vienna: University of Vienna. Unpublished PhD dissertation.
POCHHACKER, FRANZ. 1993. From knowledge to text: Coherence in simul­
taneous interpreting. In: GAMBIER, YVES & JORMA TOMMOLA (eds).
1993. Translation and Knowledge. Turku: University of Turku. 87-100.
POCHHACKER, FRANZ. 1994. Simultaneous interpretation: "Cultural transfer"
or "Voice-over text"? In: SNELL-HORNBY, MARY & FRANZ POCH­
HACKER & KLAUS KAINDL (eds). 1994. 169-178.

RADICE, WILLIAM. 1986. The Rules of Translating Tagore. Indian Literature,

May-June 1986, 33-40.
von RAFFLER-ENGEL, WALBURGA, 1988. The Impact of Covert Factors
in Cross-Cultural Communication. In: POYATOS, FERNANDO (ed). 1988.
RATH, RAMAKANT. 1985. Sri Radha. Cuttack: Agradut.
REGE, M.P. 1990. Editorial. New Quest, March-April 1990, 65-67.
REID, HELENE. 1987. The Semiotics of Subtitling, or Why don't you translate
 Works Cited 335

what it says? EBU Review # 6, 28-30.

REISS, KATHARINA. 1983. Texttyp und Übersetzungsmethode. Heidelberg:
Groos.
REISS, KATHARINA & HANS J. VERMEER. 1984. Grundlegung einer allge-
meinen Translationstheorie (= Linguistische Arbeiten 147). Tubingen: Max
Niemeyer.
REUTER, E. & H. SCHRODER & L. TIITTULA. 1989. Deutsch-Finnische Kul-
turunterschiede in der Wirtschaftskommunikation. Fragestellungen, Methoden
und Ergebnisse eines Forschungsprojekts. Jahrbuch Deutsch als Fremdsprache
15, 237-269.
RIEDMÜLLER, THOMAS. 1989. Strategien beim Simultandolmetschen. Heidel­
berg: Institut fur Ubersetzen und Dolmetschen der Universitat Heidelberg. Un­
published diploma thesis.
ROBERTS, RODA P. 1992. The Concept of Function of Translation and Its Ap­
plication to Literary Texts. Target 4, 1-16.
ROBINSON, DOUGLAS. 1991. The Translator's Turn. Baltimore & London:
Johns Hopkins University Press.
RODITI, EDOUARD. 1982. History in a Nutshell. Washington, DC: National
Resource Center for Translation and Interpretation, Georgetown University.
ROSE, MARILYN GADDIS (ed). 1981. Translation Spectrum. Albany: State
University of New York.
ROY, DEBI. 1992. Maa. Desh (India). 17 October 1992.
RSMB Television Research Limited. 1993. Language, Community and Leisure:
a study of the understanding and use of Welsh in Wales. Prepared for SAC,
Cardiff.
RØNN-POULSEN, ANNI & FINN BRANDT-PEDERSEN. 1986. Skrivevcerk-
stedet. Copenhagen: N0gleforlaget.
SAGER, JUAN. 1990. A practical course in terminology processing. Amster­
dam & Philadelphia: John Benjamins.
SAGER, JUAN. 1992. The translator as terminologist. In: DOLLERUP, CAY
& ANNE LODDEGAARD (eds). 1992. 107-122.
SAINZ, MARÍA JULIA. 1992. Developing Translation Skills. In: DOLLERUP,
CAY & ANNE LODDEGAARD (eds). 1992. 69-73.
SAVORY, THEODORE. 1957. The Art of Translation. London: Jonathan Cape.
SCHMIDT, PAUL. 1954. Statist auf diplomatischer Buhne. Bonn: Athenäum.
SCHWEDA NICHOLSON, NANCY. 1987. Linguistic and Extralinguistic As­
pects of Simultaneous Interpretation. Applied Linguistics 8, 194-205.
SCHWEDA NICHOLSON, NANCY. 1990a. Consecutive Note-Taking for Com­
munity Interpretation. In: BOWEN, D. & M. (eds). 1990. 136-145.
SCHWEDA NICHOLSON, NANCY. 1990b. A New Look at the Cognitive Flexi­
bility Factor in Interpreter Training. In: WILLSON, A. L. (ed.). 1990. 107-117.
SCHWEDA NICHOLSON, NANCY. 1991. Self-Monitoring Strategies in Simul­
taneous Interpretation. In: PICKEN, C. (ed). 1991. Proceedings of the Institute
of Translation and Interpreting Conference 5. London: The Association for In-
336 Teaching Translation and Interpreting 2

formation Management (Aslib). 46-51.

SCHWEDA NICHOLSON, NANCY. 1993a. An Introduction to Basic Note-Tak­
ing Skills for Consecutive Interpretation. In: LOSA, E. (ed). 1993. 197-204.
SCHWEDA NICHOLSON, NANCY. 1993b [In press]. The Constructive Criti­
cism Model for Post-Interpretation Analysis. The Interpreters' Newsletter.
SCHWEDA NICHOLSON, NANCY. 1994 [Forthcoming]. Professional Ethics
for Court and Community Interpreters. In: HAMMOND, DEANNE L. (ed).
Professional Issues in Translation and Interpretation (= American Translators
Association Series on Translation and Interpretation VII). Binghamton: State
University of New York.
SCHWEITZER, A. D. 1973. Perevod i lingvistika. Moscow: Vojenizdat.
SCHWEITZER, A. D. 1988. Teorija perevoda: Status, problemy, aspekty. Mos­
cow: Nauka.
SELESKOVITCH, DANICA. 1981. L'enseignement de l'interprétation et de la
traduction: de la theorie a la pedagogie. Ottawa: University of Ottawa Press.
SELESKOVITCH, DANICA. 1989. Teaching Conference Interpreting (=
American Translators Association Scholarly Monograph Series HI). Bing­
hamton: State University of New York.
SELESKOVITCH, DANICA & MARIANNE LEDERER. 1984. Interpreter pour
traduire. Paris: Didier.
SELESKOVITCH, DANICA & MARIANNE LEDERER. 1989. Pedagogie
raisonnée de l'interpretation (= Collection "Traductologie" 4 ) . Paris: Didier.
SETTON, ROBIN. 1991. Training Conference Interpreters with Chinese: Cultural
Factors. Paper given at the Third International Conference on Cross-Cultural
Communication: East and West. National Chengkong University, Tainan, Re­
public of China.
SETTON, ROBIN. 1993a. Is non-intra-IE interpretation different? European
models and Chinese-English realities, Meta 38, 2.
SETTON, ROBIN. 1993b. Speech in Europe and Asia: Discourse Processes in
the Training of Cross-Cultural Conference Interpreters. Interpreting Research
(Journal of the Interpreting Research Association of Japan) 3, # 2, 2-11.
SHACKMAN, J. 1987. The Right to be Understood. Cambridge: National Ex­
tension College.
SHLESINGER, MIRIAM. 1989. Monitoring the Courtroom Interpreter. Paralleles
(Geneva) 11,29-34.
SHLESINGER, MIRIAM. 1992a. Lexicalization in translation: an empirical study
of students' progress. In: DOLLERUP, CAY & ANNE LODDEGAARD (eds).
1992. 123-128.
SHLESINGER, MIRIAM. 1992b. Intonation in the Production and Perception of
Simultaneous Interpretation. Paper given at the Translation Studies Congress:
"Translation Studies - An Interdiscipline", 9-12 September 1992, Vienna,
Austria.
SHOCHAT, ELLA & ROBERT STAM. 1985. The Cinema after Babel: Lan­
guage, Difference, Power. Screen 26, # 3-4, 35-58.
van SLYPE, G. & J.F. GUINET, F. SEITZ, E. BENEJAM, 1983. Better
 Works Cited 337

Translation for Better Communication. Oxford: Pergamon.

SNELL-HORNBY, MARY. 1988. Translation Studies - An Integrated Approach.
Amsterdam & Philadelphia: John Benjamins.
SNELL-HORNBY, MARY, & FRANZ POCHHACKER & KLAUS KAINDL
(eds). 1994. Translation Studies - An Interdiscipline. Amsterdam & Phila­
delphia: John Benjamins.
SPERBER, DAN & DEIRDRE WILSON. 1986. Relevance: Communication and
Cognition. Oxford: Basil Blackwell.
STACKELBERG, JURGEN & BURKHART KROEBER. 1986. Alter Monch -
strenger Greis [eine Kontroverse]. Der Ubersetzer 22, # 3, 1-4.
STEINER, GEORGE. 1975. After Babel. London: Oxford University Press.
STENZL, CATHERINE. 1983. Simultaneous interpretation: Groundwork towards
a comprehensive model. London. Unpublished M.A. thesis.
STUBBS, M. 1983. Discourse Analysis. Oxford: Blackwell.
SVANHOLM, FLEMMING. 1992. The Happy Triad - The Human, the MAT,
and the MT. In: Translating and the Computer 14, Quality Standards and the
Implementation of Technology in Translation. London: Aslib. 15-20.
SODERLUND, BORJE. 1965. Att översätta. Stockholm: Bonniers Ugglebocker.

TIERNY, ROBERT J. & AL. (eds). 1987. Understanding Readers' Under-

standing. Hillsdale, N. J.: Lawrence Erlbaum.
TIRKKONEN-CONDIT, S. 1992. A Theoretical Account of Translation With­
out Translation Theory? Target 4, 237-245.
TITFORD, CHRISTOPHER. 1982. Sub-titling - Constrained Translation. Le-
bende Sprachen 27, # 3, 113-116.
TOMLIN, R.S. (ed). 1987. Coherence and Grounding in Discourse. Amsterdam
& Philadelphia: John Benjamins.
TOURY, GIDEON. 1979. Interlanguage and its manifestations in translation.
Meta 24, 223-231.
TOURY, GIDEON. 1989. Integrating the Cultural Dimension into Translation
Studies: An Introduction. In: TOURY, GIDEON (ed). 1989. 1-8.
TOURY, GIDEON (ed). 1989. Translation across Cultures. Delhi: Bahri Pub­
lications. Book edition of special issue of Indian Journal of Applied
Linguistics (1987)
TRANI, EUGENE P. 1969. The Treaty of Portsmouth. An Adventure in Amer-
ican Diplomacy. Lexington: University of Kentucky Press.
TYTLER, ALEXANDER FRASER. 1978 [original: 1791]. Essay on the
Principles of Translation. Amsterdam & Philadelphia: John Benjamin.

UNESCO. 1962. Conference des ministres de l'éducation des pays africans

participant a l'exécution du plan d'Addis-Abeba: rapport final, 26-30 mars,
1962. Paris: UNESCO.
UWAJEH, M. K. C. 1992. Uwajeh's 'Four-Level Theory of Translation' and
the Study of African Languages. Paper prepared for the 23rd Annual Confer-
338 Teaching Translation and Interpreting 2

ence of African Linguistics, USA.

VARGO, EDWARD. Forthcoming. (Non)American Readers "Writing"
(Un)American Literature. Proceedings of the Conference "American
Literature for Non-American Readers". Bellagio, June 1-5 1992.
VASCONCELLOS, MURIEL. 1990. Practical and linguistic strategies for post­
editing machine translation. In: WILLSON, A. L. (ed). 1990. 339-351.
VERMEER, HANS J. 1978. ein rahmen für eine allgemeine translationstheorie.
In: VERMEER, HANS. 1983. Aufsätze zur Translationstheorie. Heidelberg:
Julius Groos.
VERMEER, HANS J. 1989. Skopos and commission in translational action. In:
CHESTERMAN, ANDREW (ed). 1989. 173-187.
VERMEER, HANS J. 1989. What does it mean to Translate? In: TOURY,
GIDEON (ed). 1989. 25-33.
VERMEER, HANS J. & HEIDRUN WITTE. 1990. Mögen Sie Zistrosen?
Scenes & frames & channels im translatorischen Handeln (= TEXTcon-
TEXT, Beiheft 3). Heidelberg.
VERMEER, HANS J. 1990. The CERA Lectures 1990. Unpublished manuscript
for the CERA Chair for Translation, Communication and Cultures, Katho-
lieke Universiteit Leuven, Belgium, 25 June-19 July 1990.
VERMEER, HANS J. 1992 [orig. 1989]. Skopos und Translationsauftrag -
Aufsdtze. Heidelberg.
VERMEER, HANS J. 1992a. Describing Nonverbal Behaviour in the Odyssey:
Scenes and Verbal Frames as Translation Problems. In: POYATOS,
FERNANDO (ed). 1992. 285-299.
VESTER, FREDERIC. 1980. Neuland des Denkens. Stuttgart: Deutsche
Verlags-Anstalt.
VESTER, FREDERIC. 1978. Denken, Lernen und Vergessen. Munich: Deut-
scher Taschenbuchverlag.
VIAGGIO, S. 1992a. Translators and Interpreters: Professionals or Shoemakers?
In: DOLLERUP, CAY & ANNE LODDEGAARD (eds). 1992. 307-312.
VIAGGIO, S. 1992b. Cognitive Clozing to Teach Them to Think. The Inter-
preters' Newsletter 4, 40-44.
VIAGGIO, SERGIO. 1992c. Few Ad Libs on Communicative and Semantic
Translation, Interpretation and Their Schools. Meta 37, 278-288.
VINAY, J.-P. & J. DARBELNET. 1958. Stylistique comparee du frangais et
de l'anglais. Méthode de traduction. Paris: Didier.
VINAY, J.-P. & DARBELNET, J. 1977 [orig. 1957]. Stylistique comparée du
frangais et de l'anglais. Methode de Traduction. Quebec: Beauchemin.
WALTERS, VERNON. 1978. Silent Missions. Garden City, N.Y.: Doubleday.
WARNER, ALFRED. 1966. Internationale Angleichung fachsprachlicher Wen-
dung en der Elektrotechnik. Versuch einer Aufstellung phraseologischer
Grundsätze für die Technik. Beihefte der ETZ # 4.
WEBER, W. 1991. Improved Ways of Teaching Consecutive Interpretation. In:
 Works Cited 339

LIU, C.C. (ed). 1991. 403-412.

WILLSON, A. L. (ed). 1990. Looking Ahead, Proceedings of the 31st Annual
ATA-Conference, Medford, N. J.: Learned Information.
WILSON, DEIRDRE & DAN SPERBER. 1988. Representation and Relevance.
In: KEMPSON, RUTH (ed). 1988. 133-53.
WILSS, WOLFRAM. 1977. Übersetzungswissenschaft - Probleme und Me-
thoden. Stuttgart: Klett.
WILSS, WOLFRAM. 1978. Syntactic Anticipation in German-English Simul­
taneous Interpreting. In: GERVER, D. & W. H. SINAIKO (eds). 1978.
323-332.
WILSS, WOLFRAM. 1992. Übersetzungsfertigkeit. Annäherung an einen kom-
plexen übersetzungspraktischen Begriff, Tübingen: Gunter Narr.
WITTE, HEIDRUN. 1987. Die Kulturkompetenz des Translators - Theoretisch-
abstrakter Begriff oder realisierbares Konzept? TEXTconTEXT 2, 109-136.
WITTE, HEIDRUN. 1992a. Zur gesellschaftlichen Verantwortung des Transla­
tors - Anmerkungen. TEXTconTEXT 7, 119-129.
WITTE, HEIDRUN. 1992b. El traductor como mediador cultural. Fundamentos
teoricos para la ensenanza de la Lengua y Cultura en los estudios de Traduc­
tion. In: Actas del II Congreso Internacional de la Sociedad de Didáctica de
la Lengua y la Literatura, Las Palmas de Gran Canada, Diciembre 2-3-4,
1992 (= El Guiniguada 3). 407-414.
WITTE, Sergei. 1921. The Memoirs of Count Witte, translated and edited by
Abraham Yarmolinsky. Garden City, N.Y.: Doubleday.
WALTERMANN, D. 1992a. Integrating Machine Translation in a Translation
Curriculum. In: LOSA, EDITH F. (ed). 1992. 167-177.
WALTERMANN, D. 1992b. A Survey Report on Translation Programs in the
United States. Pittsburg: Carnegie Mellon University. Modern Languages Pro­
gram. Unpublished report.

d'YDEWALLE, GERY & JOHAN VAN RENSBERGEN & JORIS POLLET.

1987. Reading a message when the same message is available auditorily in
another language. The case of subtitling. In: REGAN, J. O. & A. LEVY-
SCHOEN (eds). 1987. Eye Movements: from Physiology to Cognition,
Amsterdam: Elsevier Science Publishers. 313-321.
d'YDEWALLE, GERY & LUK WARLOP & JOHAN VAN RENSBERGEN.
1989. Television and Attention. Differences between young and older adults
in the division of attention over different sources of TV information. Medien-
psychologie: Zeitschrift fur Individual- und Massenkommunikation 1, 42-59
ZHUANG, M.L. 1991. Han-ying tongsheng chuanyi de jiqiao [Skills in Simul­
taneous Interpretation from Chinese into English]. Zhongguo fanyi (Chinese
Translators Journal), # 2, 24-26.
ZIMMAN, L. 1989a. Interview with Elena Pollard. London. Mimeo.
ZIMMAN, L. 1989b. Interview with Maite Bell. London. Mimeo.
340 Teaching Translation and Interpreting 2

ZIMMAN, L. 1989c. Teaching materials for community interpreters' training.

London. Mimeo.
INDEX
 INDEX

The present index is not exhaustive. In order to increase the usefulness of the
index, indexing terms may deviate from the wording in the text. Authors quoted
are indexed, whereas references are not. Clarity has overridden consistency on
minor points.
Abbreviations used: 'Also:' refers to related terms; 'cons.' = consecutive
(interpreting); 'def.' = definition/description given in article; 'int.' = interpreting;
'sim.' = simultaneous (interpreting); 'tr.' = translation, translator.
Entry words are normally abbreviated to one letter.

'A'-language, 19, 21-22, 176, 180, 186, 187, Albert, Sándor, 98

193 Allusion, 61; intertextual a., 61-62
A-B-C grading, 21 Alternative solutions, 129-130, 131, 151, tr.
A-B-C languages, 21-22 92-93, 112, 156, 266. Also: 'Acceptable
Abstract knowledge, text as, 296 interpretations'
Academy of Letters, Indian, 27 Analysis, 42, 161, 235; a. of text 200-201,
Acceptability, 157, 158; a. norms, 146, test, 205, 206, 297, a. and target text writ­
109-110; a. in target culture, 103, of inter­ ing, 121-132, 135-141, 143; a., contrastive
pretation, 158, 163; a., cultural, 101, lin­ 56, deficient, 111, cross-cultural, 79-86,
guistic, 110, 234, logical, 187, native speak­ methodological, 111, product-oriented, 238;
er a., 214. types of a., 77-78
Acceptable interpretations, 158, 163. Also: Anderson, R. B., 217
'Alternative solutions' Anecdotes, usefulness of, 169-170
Acceptable vs. unacceptable, 151. Also: Anticipation 193-194, 228; a. cues, types of,
'Alternative solutions' 194
Acquisition, language, 21-22, 26, 35. Also: Appropriateness, 187
'Foreign language acquisition/learning' Aptitude test, 14-15. Also: 'Entrance exami­
Adaptation, 59-67, 101-104, 144-146, 152, nation'
192, 245, 269, 276 Archives, access to, 169
Addition, 192 Articulation, 238
Addressees, (def.) 237. See 'Audience' Arts and sciences, 183, 195
Adequacy, 102-103; a. vs. equivalence, 99 Assessment, 152. Also: 'Evaluation'
Adjustment, 277-278 Assignments in class, 107. Also: 'Realistic
Advance documentation for int., 209, 240 tasks', 'Tasks'
Advertisements, 77-86, as cultural portraits, Assignment, 65, 107, 108, 111, 138, 141,
79-85 223, 236, 312, 315. Also: 'Commission',
Advisor, tr. as, 72, 104-105 'Tasks'
Africa, 19-24, 41-45. Also: 'Igbo', 'Hausa' Assimilation, 53, 54, 56
African culture, 20, 43, 293 Asymmetric int., 12, syntax, 194
'Africanness', 43-44 At-sight paraphrase, 188. Also: 'Prima
Age, 263-264, 274, 291; a. limit, 176; a. for vista', 'Sight'
language acquisition, 21, 22 Attentive reading, 100-101. Also: 'Reading'
AIIC (International Association of Conferen­ Attitudes to computer-assisted tr., 301
ce Interpreters, 167, 233 Audience, shared knowledge with sender,
Albania, 14 102
344 Teaching Translation and Interpreting 2

Audience, 79, 237; a. assessment (of int.), Black-American English, 55

241-242, attention, 82, 279, attitude, 226, Book, characteristics of, 272
comprehension, 230, knowledge, 241, reac­ Book titles, ex. 59-64
tion, 79, 257, reception of subtitling, 266; a. Booklet with tasks, 124
in translation, 101-105; a.'s background Booth, int. 13, 17, 18, 176, 192, 208, 227,
knowledge, 91, knowledge, 241, effort in 228, 229; b. in dining rooms, 13; Pan-Scan­
processing, 246, 250; appeal to a., 194 dinavian b., 17; recording of b., 225; b.
Audiovisual interlingual transfers, 275-283; sentence-by-sentence cons. int. (Japan), 192
a. media as a means for foreign language Bridging course/programme, 34-36, 255
acquisition, 283; a. messages, types of, 275, Brief/briefing, 171-172, 223
texts, in int., 236 British Association of Community Inter­
Audiovisual texts, in int., 236 preters, 219
Audiovisual translation, 261-274, as a new Brittany, 258
genre, 277 Brundage, D.H. and D. MacKeracher, 134-
Austen, Jane, 54-55 135
Austria, 280 Brussels, 13
Authentic conference material, 225-226, 233
Authenticity, 28 'C'-language, 22, 23
Authenticity, as the most important element Cadence, 31
in subtitling, 265-269 Caille, P. F., 273
Author's intention, 63, 239, 297. Also: Cambodian (Khmer), 211
'Sender' Camcorder, ex. 101-105
Awareness. See 'Consciousness' Cameroon, 41-46
Canada, 268
'B'-language, 22, 23, 178, 186, 187, 192, Capacity, cognitive, 21, selectional, 203;
193 professionals' management of c , 230
Back translation, 55-56 Carpentier, A., 64
Background information, cultural, 49-50. Carr, E., 173
Also: 'Culture' Cartellieri, C, 233
Background knowledge for int., general and Case stories (Community int.), ex., 221-222
situational, 207-209, of (economics), 115, Catalan, 258
118, of sender and audience, 102; audien­ Categorial interference, 291-292
ce's b. k., 91, clients' b. k., 72, 74, trans­ Central Europe, 17
lator's, 72, 108-110; deficient b. k., 185 Chafe, W., 196
Background material in class work, 146 Channel, 264, 265; c, types of, 78
Background for subtitling, 255 Chesterman, Andrew, 95
Backgrounds, interpreter students', 14-15, China, Republic of, 14, 183-198
212, 172 Chinese-English int., 185
Baraduc, J., 80 Choice, int.'s, 228, subtitler's, 247, 249;
Basque, 258 translational/tr.'s, 43-44, 107. Also: 'De­
Beaugrande, Robert de, 113 cision'
Belgium, 261, 280 Churchill, Winston, 171
Bengali, 27, 28, 31, 32 Clarity (in language), 95, 108, 186, 192, 234
Benjamin, Walter, 28, 32, 159 Class discussion, 122, session, 125, 139, c.
Bhartrhari, 32 size, 122, 139, 186
Bi-active directionality, 185 Classical language teaching, 121, use of
Bible, 158, 159 (hypothetical), 18
Bicultural competence, 69, 71-72, 74 Classroom assignments, 143-144
Bilingual(ism), 20, 21, 22, 35, 45, 121, 212, Clients, 4, 218, 220, 221, 242; c.'s back­
268-269, 275, 281, 282; b. text files, 306, ground knowledge, 72, 74, c.'s purpose, 72,
tutors, 213; in Finland, 275, in Nigeria, 21- c.'s trust, 104-105, 124
22, in Wales, 255 Closure, tr. as, 52
Binding laws, 12 Cloze, 187, 298
 Index 345

Coal and Steel Community, 11 strategies, 228

Code in note-taking, 203-205 Comprehension of technical texts in foreign
Cognition (Cognitive, preparation/approach, vs. mother tongue, 296-297
etc.), 71, 113, 183, 187, 195, 201, 240-241, Computer, 176, 180, 265, 282, and tr. 301-
266, 309, 311-318; c. literacy, 262-263, 311; c.
Cognitive mapping vs. linguistic construct­ spelling and grammar programmes, 130,
ivism, 120, notions, 246, psychology, 183 305, 315
Coherence, 240, 280, 298-299, c. with visu­ Computer-assisted language learning, 258,
al side in subtitling, 282 tr. 301-308, 309-317
Cohesion, 186, 234, 248, 264, 298-299 Computer-based translation, 301
Collective professional lore, 98-105 Conceptual connectivity, 115-120
Collocation, 124, 194, 210, 264, 298 Concept of communication, 70, broad, 61-
Comenius University (Bratislava), 207 66, 72-75
Commentary, 270 Concepts, Western vs. Oriental, 48, 49
Commentary in audiovisual media, 276 Conceptual construction, 113, framework
Commercial market, 256-257, needs, 258, 49, 89, framework (lack of in tr.), 98, links,
translation, 308 113-120, representation, 120, vocabulary,
Commercially available Machine Translation 198
systems, 301, 302-307 Conference, 240, discourse, 195, material,
Commission, European. See 'European 208, 209, 225, 233-234
Commission', 'European Union' Conferences, 93, 189, 278, 302, multiling­
Commission = job, 65, 208, 307. Also: ual, 5, 12-13, 17-18, 185
'Assignment' Confidentiality, 168, 233-234
Common working language for student and Connotation, 73, 155
teacher, 211 Consciousness, (teaching for) heightened,
Communication, 70, situation, texts in, 65, 111, 112, 131, 146, 153, 155, 193, 221, 245,
strategies, 81, strategies in advertisements, 247-248, 251, 262
77-86; c, cross-cultural, 72, naive cross- Cons, int., 186, 188, 189-192, 199-206,
cultural, 70-71, natural vs. symbolic, 272 characteristics of, 272
Community, 69, 83, 90, cultural back­ Consensus, 163
ground, 78, of readers, 163; language c., 79, Consistency, 234, 240, 241, 265, 307, and
Also: 'European Community' machine tr., 305, 307, inappropriate, ex. 268
Community interpreting, 195, 215, 217-224. Constraints, 66, in subtitling, 282
Comparison, 85, 312, between cultures, 70- Contents of courses in int., (EU) 15-16,
71, between subjective and objective reality, (Taiwan) 185-194, (Bulgaria, in note-taking)
234; c. impossible in int., 227 199-206, (Slovakia, in crisis management)
Competence, 98, 241, in comprehension and 208-210, (Minneapolis, for community int.)
production, 183, in target language, 215, 212, (UK, for community int.) 220-223
bicultural, 69, 71-72, 74, cross-cultural, 79, Contents of courses in machine tr. (Carnegie
linguistic, 125, tr.'s, 34-36, 92-93 Mellon) 313-314
Competence, technical, 256. Also: 'Com­ Contents of courses in subtitling, 245, 251,
puter literacy' 255-258, 261-263, 282-283
Completeness, 241 Contents of courses in tr., (bridging course)
Components in course on improvisation, 34-36, (Nigeria) 42-44, (functional
208-210 approach) 65-67, (advertisements) 77-86,
Components of the Translators' Work- (Denmark) 124-125, (Uruguay) 133-134, for
Station, 314 linguists, 288-290, (technical tr.) 295
Composition, 147, 148 Context, of exercises, 187-194
Comprehension, 113-120, 161, 172, in tr., Context, situational in community int., 221-
113; audience c , 230, levels of c , 101. 224
Comprehension (analysis) and production, Context (def.), 246, of excercises, 187-194;
113-120, 143-149, 184-185, 186, 200, 205 c, cultural, 43-44, 66, situational (in com­
Comprehension strategies and production munity int.), 221-224; textual, 315
346 Teaching Translation and Interpreting 2

Contradiction, 123, 162, 241 delivery, 16

Contrastive analysis, 56. Also: 'Comparison' Delivery, clients', 219, delegates', 16, hits',
Control centres, 115-117, 202 179, 183, 185, 189, 199, 229, 234, 237, 240,
Conventionalised texts in teaching, 66 242, parameters, 236-238; in mass media
Conventions, 64, of subtitling, 280-281 (narration), 276
Conversation, rules for, 217-218 Denmark, 11, 17, 121-125, 143, 282
Corpora, representative and limitations, Description, 97, and prescriptiveness 240,
231 266
Correction, 125-127, ex. 126, 138, 138-140, Deviations, 129
143, 144, 228, chart, 138-139, 141, marks, Dhvani, 31-34
124, 126 Diaculture, 69
Corroboration, 91-92 Dialogue, 265, 269
Cost of int., 13 Dictionaries, ex. 114-119, 131, 139, 143,
Council of Ministers, 12-13 144, 208, computerised 305, 306, 313-316
Course programmes. See 'Contents of cour­ Diffent acceptable products, 151. Also:
ses' 'Alternative solutions'
Creation of target text, 147. Also: 'Pro­ Differentness, 51-57, 56
duction' Difficulties in obtaining corpora, 225-226,
Creativity, 31-32, 42, 44, 130, 157, 180, 235-236, in understanding, 108
need for, 151 Directionality of int., 21, 185-186, 192, 227;
Credibility in subtitling, 268-269 in subtitling courses, 262; d. in tr. (Danish-
Crisis management, 208 English), 124-125, (involving Spanish, in
Criteria of relevance, 148 Uruguay) 134
Criticism, tr., 91-92 Directionality, 185-186
Cross-cultural awareness, 51-57, communi­ Discourse analysis, 70, 194, 195, compre­
cation, 69, distinctions, 56, reading, 52-53, hension, 185, familiarisation, 186, handling,
training, 183 185; d. a. and int., 195-197
Cueing in subtitling, 267, 282 Discourse, types of, 188
Cultural adaptation, 62, acceptability, 101, 'Dispositio', 145, 149
aspects, 83, background of community, 78- Discussion in class, 125, 129, 131
79, barriers, 48-50, contexts, 43, 66, dimen­ Distancing effect, 81
sions of tr., 25-36, identifiers, 34-36, ident­ Distortion, 123-124, 129, 210
ity, 20, interpretation, 255, information, 83, Documents, conference, 209, 210, 226, 240
norms, 66, 93, 154, symbiosis, 28-37 Dollerup, C, 149
Culture shock theory, 70 Double constructions, 249
Culture, (def.) 69, tr. norms in, 93 Drama tr., characteristics of, 272
Culture-bound concepts, 53, factors, 48, Dubbing, 261, 263, 265, 272, 275, 276
norms, 70-71, problems, 47 Dumbleton Report, 20, 21
Culture-specific aspects in advertisements, Dummy booths, 16. Also: 'Booth'
81-85, background knowledge, 52-53, pro­ Dutch, 11
jection, 72-76, self-portrayal, ex. 79-81, tr.
conventions, 63 'E'-language, 23
Culture-specificity, 70, 77-86 Eastern Europe, 17
Czeck Republic, 14 Eco, Umberto, 157-163
Economic Community of West African
'D'-language, 23 States, 20
Decision about meaning in int., 205, 228, Economy in note-taking, 202-205
Decision, subtitler's, 252 ECU, 13, 301
Decision, tr.'s, 66, 250, 315-316 Effect, equivalent, 61, in subtitling, 265-266
Decision-making, 200, 206 Effort in processing, hearer's, 246, 250
Deconstruction, 157 'Ego', 159 ex.
Delay, 269-271, 279 Eisenhower, Dwight D., 169, 171
Delegate assessment, 233-235, 236-242, d.s' Elaborate, 146, 250-251
 Index 347

Empirical science, 90, studies, 114-120, False friend, ex. 123, 179
225-232, 233-242 Fame in culture, 64
Empirical/icism, 153-155 Familiarisation with technical equipment,
English, 41-46, 211, 221, 222, 225, 227, 256, 262-263
262, 289-293; In Nigeria as 'A', 21, 22, 23, Farsi, 47-50
as 'B'and 'C', 23, status of, 23; E. in India, Fawcett, P., 264
26-36, as L4, 28; as official language, 19-24, Feedback, 92, 94, 95, 122-132, 134, 135-
26 141, 147,213,214-215; f. form (chart/sheet)
English and Farsi, 47-50 124, 125, ex. 128, 130, 121, ex. 136, 138,
English and French, 114-120 ex. 139; f. situation, 143; to developers of
Entrance examination, 14, 18, 133, 175, machine tr., 311; f., individualised, 122,
176, 186-187 138-139
Equivalence, 51, 59-62, 90, 95, 98, 101, Feedback-effect in original in subtitling, 268
102-103, 184, 208, 229, 288-290; equivalen­ Fidelity. See: 'Loyalty'
ce in technical tr. 264-265, of effect, 274, of Figurative Translation, (Uwajeh) 288-293
effect in subtitling, 266, vs. adequacy, 99; Film, 20, 26, 265, 269; tr., characteristics of,
dynamic e., 61, textual e. vs. local corre­ 272; regional f., 20, 26
spondence, 99; concept of e., 59, rejection Final examination, 16, 233, 261, in subtit­
of, 238 ling, 255-256
Equivalent target text, 59-60 Finland, 262, 280, 282
Equivalents, 208, (difficult: Farsi e. of First language acquisition, 21. Also: 'For­
English) 49, (None: English e. of Oriya and eign language acquisition/teaching'
Bengali), 31, potential, 147, (rare) 144; e., FIT (International Federation of Transla­
propositional, 248 tors), 167
Error, 43-33, 56, 65, 67, 125, 110, 111, 114- Flemish, 280
120, 129, 138-140, 143, 148, 151, 179, 196, Fluency, 14, 44, 177, 212, 238, 240
228, 229, 231, 238, 242, 266; e. diagnosis, Footnotes, 32, 34, 43
111, elimination, 89-95, minimisation, 94- Foreign language acquisition and int., 195-
95, in int., 226, in notes, 205; formal e. 124 196
(Also: 'Grammar'), interlanguage e., 122- Foreign language teaching and tr., 121, 144
132; quantity of e., 146; social implications Foreign language teaching and translation,
of e., 124; types of error, 110, types of e. in 144
specific tasks, 140 Foreign words (in Welsh), 257-258
Esperanto, 18 Foreign vs. mother tongue comprehension,
Ethics, 158-159, 167, 176, 208, 212, 213, 296-297
217; e. and int., 171-172, ethics/intervention Forgetfulness, 147. Also: 'Memory'
in community int., 219-224; e. and tr., 94 Formal errors, 124, 129. Also: 'Errors'
EU. See 'European Union' Formal equivalence in subtitling, 267
Eurocentricity, 4 Framework, conceptual, 89, lack of concept­
European Commission, the, 11-18, 301 ual in tr., 98
European Parliament, the, 177 Free Translation, 152, (Uwajeh) 288-293
European Union, 176, 309; and translation Free-lance, 45
studies, 4-5; languages at, 11-12, 17-18 Freedom, degrees of f. in rendition, 197.
European vs. Oriental values, 47-50 Also: 'Alternative solutions'
Europhone African literature, 41-46 French, 11, 13, 20, 21, 60, 115-120, 225,
Evaluation system, 134-135, of int., 240 227, 262, 270, 289-290; F. in Nigeria, 23,
Exoticisation, 64, 73 tr. into English, (in Nigeria) 41-46
Expansion, 146, 250-251 'From-State' (and 'To-State'), 154
Expectations of the target audience, 257 Function, 61, 103, 144; emotive f., ex. 63-
Experience, 75, 245, 295 64
Expert knowledge vs. interpreters' know­ Functional adequacy, 240; (approach/ the­
ledge, 228 ory), 59-67, 60, 69, 145
Functional analysis, 65-67, 74-75, 79-86; in
348 Teaching Translation and Interpreting 2

use of theory 105-112; in translation strat­ Huckleberry Finn, 55-56

egy, 108-112 Human tr. and machine tr., 302, 309-310
Functions of literary translation, 44, of Hungary, 14
source and target texts, 63-65, different f. Hutchins, John, 302
for source and target text, 64 Huxley, A., 62
Hypothesis, 92, 93, 114, 117, 228, 235, and
Galicia, 258 h. verification, 120
General Theory of Tr. and Int., 238 Hypothesis-testing, 91, 108-110
'Gentleman', ex. 54
Genuine communicative purpose, 147. See: Iceland, 261
'Realistic tasks' 'Id', 159 ex.
German, 11, 13, 55, 59, 60, 77, 79-85, 227, Ideas, difusion of, 5, 93-94,132, revision of,
262, 270, 315 153
Germany, 258, 280 Idioculture, 69
Gift, translation as a, 42 Idiolects, 131
Gile, Daniel, 91, 234 Idiomatic, 101. Also: 'fluency'
Global cognitive patterns, 201 Idiosyncrasy, 160-161, 162, 200
Glossaries, task-specific, 5, in machine-tr. Igbo, 19-24, 289, Igbo-English, 291-293, I.
316 world picture, 293
Gnostic, 159 Image cultures claim, (def.) 72, of self, 70
Goal, tr.'s, 152. Also: 'Intention' Image, professional, 173
Grammar, 144, 149, 240, 241, 315; g. Immediacy of Interpretation Hypothesis, 205
checkers/programmes, 130, 305, 315; g. Impact of language technology, 301
drill/excercises/tr., 122, 125, 144 'Implementation', 138-139
Gran, L., 200 Improvisation in int., 207-210
Greece, 11 In-house material/service/training, 14-16,
Greek, 18 253, 274
Greimas, A. J., 162 Inadequacy, 129
Group work, 15, 92, 213, 221, 223, 256, Incomprehensibility, 61, 123
312-313, g. assignments, 313 Increase in length in translation, 245
Guide to Good Practice, 219 India, 25-36
Gutt,E.-A. 103, 247, 251 Indian Journal of Applied Linguistics, 25
Individual and culture, 29
Hall, Donald, 31 Individualised feedback, 122, 125-126, 128-
Hamlet, 47-48 129, 130, 138-141, training, 256, 262
Hard of hearing, 254, 270, 275 Individuality, student, 125, 130, 179-180
Hatim, B. and I. Mason, 263 Individuality in int., 227
Hausa, 19-24 Indo-European, 122, 194
Hearing vs. not hearing, 209 Inferences, 246
Heightened awareness/consciousness. See Information, extratextual, 113, sources of,
'Consciousness' 143, tr.'s background, 49-50
Hermeneutics (def.), 157 Informativeness, 81
Hermetic, 159, poetry, 61 Initiator, 104. Also: 'Clients'
Hermetism, 158, 159, 160 Inspection of int. product, 233
Heynold, Christian, 172 Institutional rules, 175
Hindi, 26, 27, 28. Also: 'Bengali', 'Indian', Instruction/Teaching carried out in non-
'Oriya' native language, 213-215
Hindu philosophy, 31 Instructions, 66. Also: 'Manuals'
History of int., 167-173 'Integration', 140
Hmong, 211 Intensifiers, 35-36
Homework, 148. Also: 'Study outside class' 'Intention', 163
Hong Kong, 14 Intention, 72, 102, clients', 72, modification
Howlers, 129, 132, 307 of clients', 72; speaker's i., 202, teacher's i.,
 Index 349

121, 125, 131, 138 Japan, 192

Interactant, 70, 217, 236. Also: 'Student- Jerome, 158
teacher relation'
Interaction of texts, 184, social, 71-75, King Report, the, 12
student, 180, student and machine, 311 Knowledge, culture-specific background, 52-
Interference, 229, 231, (English and Chi- 53, 54-57
nese) 188, categorial, 291-292 Knowledge, background k. of tr., 42
Interlanguage, 121, 124, 130, i. errors, 122- Knowledge of African literature, 42, of
132 audience, 92, of cultures, 36; audience k.,
Interlingual translation, types of, 272 241; real-world k. 302, nature of k., 90,
International organisations, 4-5, 11, 20, 41, scientific k., 89, 93, collective translatorial
188-189 knowledge, 98,99; background knowledge of
Interplay between teacher and students. See clients, 72, for int., 207-208, of tr., 42, 72,
'Student-teacher relation' 108-110, tr.'s economic k., 115, 118, tr.'s
Interpretation, 157, 185, 267, indefinitive- technical k., 295-299
ness of i., 160, limits to, 161; culture-bound Knowledge storage, 196,
i., 52-53 koller, W., 60-61
Interpretation and Overinterpretation, 161 Korostovetz, I., 168
Interpreter training, and history, 169-170, Krishnamoorthy, K., 32
172-173; int. training in the Republic of Kyle, K., 254
China, 183-198; int. training, use of research
in, 184-196, 229-232, 233, 242; and note- L1, L2, L3, L4, 19, 27-28, 217. Also: 'A',
taking, 199-206; in working conditions/im- 'B', ' C , 'D'-language
provisation, 208-211; with non-native in- 'Langage', 159
structor, 213-215 'Langue', 159
Interpreter as advocate, 217-224 Language professionals, 4-5, 141
Interpreters on int., 167-169 Language of published literature, 41
Interpreter, types of students, 178, 179-180 Language acquisition, 19, first vs. second,
Interpreters, candidates for, 14-15, 185, lack 21, 1. a. by means of audiovisual mass
of qualified candidates, 14, 186-187 media, 283
Interpreters' working conditions, 170-172 Language combinations, 17, 258, at course
Interpreters' information on, 167-169 in Minneapolis, 213, for interpreters, 176
Interpreting, directionality of, 21, nature of, Language learning and int., 187, learning
184-185, price of 13, strategies, 225, 228; and int. training, 212-213, 215, 1. teaching
int., asymmetric, 12, partial, 210, poor, 13; vs. int., 175, teaching and tr. teaching, 97,
modes of i., 276 121
Interpreting services at the European Com- Language norms, 157,1. professionals, status
mission, 11-18 of, 4-5, 20, 24 (Also: 'Professionals'), 1.
Intertextuality, 61-62 rights, 24
Intervention in community int., 217-224 Language as culture, 25
Intonation, 230, 249, 265 Language use in conversation, 217, used in
Intra-African English-French bilingualism, note-taking, 203-205
20 Language technology, 308. Also: 'Machine
Intuition/Intuitive, 93, 183, 195, 198, 245, tr.'
248 Language, functions of, 95
'Invention', 145, 149 Language International, 305
Ireland, 11,258,283 Languages at the EC, 11-12, 17-18, 1. of
Irish, 258 colonisation, 41
Isotopy, 162 Languages, regional and English, 26-28
Italian, 11,55,270 'Langue', 287
Italy, 16 Lao, 211
Larson, M., 152
Jacoby, L., 201 Latin, 18
350 Teaching Translation and Interpreting 2

Laye, Camara, 43 Manzoufas, M., 269

Layout, 77, of subtitles, 267 Marshall Plan, 172
League of Nations, 168 Materialism, 31, 49
Learning styles, 179 McDavid, J., 78
Length of tr. works, 245 Meaning, 158, 161-162, and context, 162;
Level, textual, 313 uncertainty of m., 157
Levels, 65-66, in communication, 70, in tr. Media, 254, 255, m. discourse, 195, use,
as communication, 122-123, of solutions, 238. Also: 'Audiovisual', 'Subtitling'
90-91, of tr., 288-290, of understanding, Member states of the EU, 11, 12, of the
101; 1., linguistic and semantic, 113 UN, 12
Lexical, 289, errors, 117-120 Memorisation, 188
Lexical Translation, (Uwajeh) 288-293 Memory, 15, 147, 179, 185, 187-188, 189,
Lexicon, 192, 312, 314, 315. Also: 'Vocab­ 192, 199, 200, 205; in machine tr., 306
ulary' Methodological problems in int. research,
Limitations in audiovisual media, 264. Also: 227, 235
'Constraints' Methods, teaching, for technical translation,
Limitlessness of interpretation, 158 297-298
Limits to acceptability in subtitling, 266, 1. Micro-evaluation, 134
to interpretation, 161, 163, to tr., 157 Minimal segment vs. global meaning, 162
Linguistic acceptability, 110, 234 Minimisation of effort of processing, 246,
Linguistics Association of Nigeria, 20 250, of errors, 94-95
Links, conceptual, 113-120 Minor(ity) languages, 258, 261
Lip-synchrony, 271, 276 Mishearing, in int., 171 ex.
Listenability, 176, 229 Misperception. See 'Errors'
Listening ability, 15 Mistakes. See 'Errors'
Literal rendition in int., 220 Mistranslation. See 'Errors'
Literal translations, 47-48, 62, 63, 103, 122, Misunderstanding, 95, 171
125, 152, 154, (Uwajeh) 288-293 Mock conference, 188, (def.) 226, 227, 229
Literary texts, 124 Model tr., ex. 127, 129
Literary translation, 41-46, 47-50, 51-57, 59- Modules, 296
66, ex. 73, 98, 280, 308 Monitoring, 139-140, in int., types of, 230-
Localisation of concept, 22 231
Lodge, David, 55 Monolingual(ism), 21, 22-23, 35, 143, 275
Logic, 108, 241 Monosemiotic channels, 265
Longitudinal study, ex. 199-200 Moscow Linguistic University, 170
Loss in subtitling, 278, in tr., 53-54, 56; 1. Mossop, Brian, 268
of teacher control, 148; quality of 1., 208 Mother tongue, 26, m. t. education, 19,
Lotman, Juri, 25 incomplete command of, 21
Loyalty, 61, 63, 66, 122-123, 124, 152, 155, Motivation, 66, 112, 121, 146, 147, 148,
198, 234 170, 215, 232
Luther, Martin, ex., 81-82 Multi-level context, 236
Luxembourg, 13 Multiculturalism, 283
Multilingual culture and translation studies,
Machine interactive form of learning, 258 20
Machine translation, 42, 90, 130, 155, 198, Multilingual(ism), 1, 19, 21, 22-23, 35, 185-
287, 301-308, 309-317; number of m. t. 186, 283
documents, 305 Multilingual media world, 261
Machine translation systems in curriculum, Multilingual transfer, types of, 277-279
309-317, types of m. tr. systems, 309 Music and sound effects, 265
Macro-evaluation, 134
Man-woman relationship in India, 29-30 Naive intercultural communication, 70-71
Manual, 97, ex. 101-105, 143, 265, 315. Narration (in audiovisual media), 271, 276
Also: 'Instruction' National language, (Hindi) 26
 Index 351

National Book Trust of India, 27 taking, 199-206

Natural communication, 272 Oral vs. written discourse. 277-279
Naturalness, 272, 273 Oral vs. written norms, 279-281
Nedergaard-Larsen, Birgit, 269 Organisations, international, 1, 11, 20, 188.
Need for authentic texts in teaching, 65-67, Also under names.
145 (Also: 'Realistic tasks'), n. for creativ­ Originator, 104
ity, 151, for refinement, 229, for language Orissa, 26, 29, 31
professionals, 4-5, 20, for successful tr. Oriya, 26, 27, 28, 29, 31. Also: 'Indian'
strategies, 97, n. of int., 11, 12, 13; n., Otherness, 51-57
social, for translation, 98; n., commercial, Overinterpretation, 160, 161, 163
258
Neotext, 192 Paraculture, 69
Netherlands, 261, 280 Paradigm, use of wrong, 118
Newmark, P., 78, 152, 153 Paralinguistic factors, 236-237, 240
Nida, E., 61 Parallel texts, 66, 91, 143, 146, 297-298, as
Nigerian, 19-24, N. Pidgin English, ex., 290 aim of translation, 103
Nir, Rafael, 273 Parameters, 235, 237, in int. and tr., 184, in
Nomenclature, neutral, 152-155 sim. int., 233-242; societal p., 124. Also:
Non-translations, 66 'Spectrum'
Non-synchrony, 269-271 Paraphrase, 186, 187, 188, 192-193. Also:
Nonsensical texts, 209-210 'Shadowing', 'Summary'
Nordic Council, the, 17 'Parole', 287
Nordic countries, 261 Part-time study, 134, 256
Norms, tr. in culture, 93 Partial reduction, ex. (English-Slovene),
Norms, student, 112 247-248. Also: 'Omission'/Reduction'
Norms, translational, 93 Partners, teachers as, 148
Norms for acceptability, 146, 158, for lan­ Passive languages in int., 15, 176
guage usage, 217, in teaching, 151, quality Pattanayak, D.P., 26
n., 240, target language n., 157; cultural n., Pedersen, Viggo Hjørnager, 266
66, 70-71, 93, 154, 214, 222; student n., Peer, 138, 140, 178, 312, correction, 132,
112, tr. n., 93; oral vs. written n., 279-281. 138-140, group, 147, work, 138-140
Also: 'Acceptability', 'Errors', 'Prescriptive- People-orientation, 267-275
ness' Perception of the world, different, 49
Norway, 17, 280 Perfection, 123, 144, 155, 157, not required,
Notes, ex., 191, 202, 203, 204 307, unattainable, 90, 91
Note-taking in int., 15, 185, 189-192, 199- Perspectives on int., 172, 173. Also: 'Cons­
206, 213, 226; n., teacher with students, 214 ciousness, heightened'
Notional difficulties, 115-117 Philosophy of Sanskrit grammar, 32
Nunan, D., 140 Philospohy of process-orientation, 108
Picture composition, 265, pictures ex. 79-85
Objective Knowledge. An Evolutionary Pidgin, 43
Approach, 89-95 Pivot, 18. Also: 'relay'
Objective reality, 234 Plastic control, 93-94
Objectives of translator training, 42. Also: Plenary, 237
'Task of teaching' Poland, 14
Official languages, limitations on, 12, 17-18, Polysemiotic channels/texts, 265, 266, p.
in Nigeria, 19, Official languages at EC, 4, text types, 269
11-13, 17-18, 176, at UN, 12 Poor int., 13, originals, 264
Ojo, Ade, 41, 43 Popper, Karl, 89-95
Omission, 104, 140, 192, 210, 281. Also: Portugal, 11, 16,258,261
'Loss', 'Reduction' Portuguese, 41
'On their own'. See: 'Study outside class' Post-editing/editor, 303-304, 310, 312, 313,
Optimisation, 107, 144, 175, 245, of note- 314
352 Teaching Translation and Interpreting 2

Post-structuralism, 157 Pseudo-text, 146

Practical guide for professional interpreters, Public translators (in Uruguay), 133
233 Publisher, 41, 44
Practical difficulties in programme develop­ Purpose, genuine, 147, of target text, 144.
ment, 254 Also: 'Intention'
Practice and theory, 97-105, 196 Pym, Anthony, 92-93, 97
Pragmatic aspects, 59, equivalence, 184, Pochhacker, F., 236
perspective, 160, situation, 114, p. situation
factors, types of, 85, texts/tr., 98-105 Qualified trainee interpreters, 14-15, 186-
Pragmatics, 59-60, 149, 195, of translation, 187
145, of discourse, 186 Quality, 111, 112, 123, 238; in audiovisual
Predication, (def.) 200-201, 200-204, 206, media, 264, in int. 233-242, in machine tr.,
types of p., 201 306, in tr., 97, 123, 151, 307; and salary, 5,
Predictability, 209. Also: 'Anticipation' 90; q., subjective and objective perception
Prejudices, 73 of, 234, 242
Prescriptiveness, 97-105, 151 Questionnaire, ex. 220, 234
Pressure of time, 255
Previously translated documents, computer­ Radice, William, 31
ised, 305 Ramayan, 34
Price of int., 13 Re-translations, 158
Pride and Prejudice, 54-55 Readability, 280
Prima vista, 208, 261, in subtitling, 262. Reader and text, 51-53
Also: 'Sight' Reader response, 52-57, 91
Primary languages, 41. Also: 'Mother tong­ Reader's expectations, 63
ue', 'Official 1.' Readership, 92
Printed translation, 272 Reading, 51, 57, 101; r. time, 267; attentive
Problem solving, 89-95, model for, 89 r., 52, 54-55, 100-101, cross-cultural, 41-57
Process orientation, 107-113, 146-147, 148- Real communicative situation, need for, 65
149, 178-179, 195, 234-235; p. of cross-cul­ Real-world knowledge, 302
tural reading, 52-53, of technical tr. 296, p. 'Realistic' translation tasks, 65-67, 145
of translation, 107-113; student-process o., Recall of source text, 143, 204
137, 178-179 Receivers, television (in Wales), 253-254
Processing capacity, 199 Receptor's culture-specific background, 73-
Processing effort, hearer's, 250 74
Product inspection, 241 Recording, 266, 235-236
Product-orientation, 107-113, 238 Recordings, 226
Product-oriented research, methodological Recruitment test, 16, 177
problems in, 235 Reduction, 273-274, 276, 281, in subtitling,
Production of new text in target culture, 84 245-251. Also: 'Loss'
Profession, translation as a, 4-5, 98 Redundancy, 249-251, 273, 280
Professional lore, 93, self-control, 229, Reference, (def.) 113
standards, 231 (Also: 'Standard'); p. tr, 75, Referendum in class, 131
104, 105, 114, 121-122, 141 Reformulation, 108-110
Professional vs. student strategies, 229-230 Rege, M.P., 27
Profile, ideal student, 176 Regional languages in India, 26, regional
Programme. See 'Contents of courses' languages and translation, 26-28
Progression in int. training, 186 Register-switching, 188
Projection, culture-specific, 72-76 Regulation No.l (at the EC), 11
Projection of image of own culture, ex. 70, Reid, H., 273
70-75 Reinterpretation, 158
Propaganda, 78 Relationships, semantic, in text, 113-117,
Prose fiction, 41 200. Also: 'Student-teacher relationships'
Protocols, think-aloud, 113-120 Relativation, 228
 Index 353

Relay, 12, 17-18, (def.) 186, 188 Semantic interpretation, 162

Relevance, 100; r. theory, 99, 103, 245-251 Semiotic(s), 158, 160, 162, channels in
Religion, 25, 29, 36, 49, 160 collision in subtitling, 268, composition of
Remuneration, 4, 105, 185 tr. types, 269-272, s. systems, 277-278
Renaissance, impact of, 47-49 Sensitise. See 'Consciousness, heightened'.
Renditions, 266, muddled, 123 Sentence splitting, 230
Reorganisation of order in target texts, 145 Sequential process, tr. as, 144
Repetition and machine tr., 306 Sets of terms, 298-299
Reports, by King (on UN) 12, by Dum- Shackman, J., 219
bleton (on languages in West Africa), 20-21, Shadowing, 186, 188, 189, 192-193
student r., 108 Shakespeare, William, 47-50, 63
Representation, conceptual, 120 Shifts, 53, 56
Representativity of corpora, 231 Shlesinger, Miriam, 171
Reproduction, 144 Shochat, E. and R. Stam, 264, 269
Republic of China, 14, 183-198 Sianel Pedwar Cymru (= S4C), 253, 254,
Research goal affecting method, 227-228 256
Research questions, 231 Sight tr., 277. Also: 'At sight', 'Prima vista'
Responsibility, 73, 74, 148, 175, 219 Simultaneity, 269-271
Retrieval, 201, 208 Sim. int., 16, 100, 186; sim. int. and subti-
Revision, 4, 91-92 tling, 267; s. subtitling, 276, s. translation,
Revoicing, 270, 276 277
Rewriting, 99, no r., 138 Situation, pragmatic, 114
Roberts, R., 62 Situational context, 208, in quality assess-
Roditi, Edouard, 172 ment, 242, s. context in community int.,
Role play, 223, relationships, 222 221-224; s. c. of conversation, 218;
Roy, Debi, 32-34 situational facts (linguistic), 65
Rules, 107, 111, 175 Skopos, 62-63, 69, 99, 100, 101, 102-103,
Russian, 18, 270 104, 144. Also: 'Scopos'
Slovene, 54, 55-56, 245, 247-250
S4C (= Sianel Pedwar Cymru), 253, 254, Small World, ex. 55
256 Snell-Hornby, Mary, 25
Salabega, 29-31 Social conventions, 79
Sanskrit, 31, 32 Social implications of translation errors, 124
Sapir, Edward, 25 Socialisation, 71, 74
Scarcity of good candidates, 186-187 Society and translation, 28-29
Schema, (Popper) 89, (Rose) 151-152 Socio-cultural background of target culture,
Science, translation as incipient, 98-99 84
Scientific knowledge, 89, 93 Sociolinguistic factors, 91
Scientific method, 89-95; model for s.m. 89 Socrates, 134-135
Scopos, 73, S. Theory, 69, clients' s., 72. Software, 155. Also: 'Machine tr.'
Also: 'Skopos' Solutions, levels of, 90-91
Scotland, 258 Solution of translation problems, 65
Screen translation, types of, 269-272 Solution, alternative/different, 90-91, 112,
Scriptures, 158, 159 129-130, 131, 151, 266, 267; one s. only,
Segmentation, 193, of subtitles, 267 265, 295; (good) student s., 129, 179, ex.
Segments in tr. teaching, 144; linkage of s., 208; teacher s., 107
in machine tr., 306 Something or nothing at all in int., 210
Selection of information in message, 202 Song, subtitling of, 257
Selectional capacity, 203 'Soul', 304, 307
Self correction in int., 231 Source language-orientation, 290
Self-criticism, 95 Sources of information, 143, on history of
Self-image, 70, 71-72 int., 167-169
Self-monitoring, 140-141 Spain, 11, 16
354 Teaching Translation and Interpreting 2

Spanish, 41, 77, 79-85, 221, 222, 262, 270 Subtitler training and relevance theory, 246,
Spanish-American War, 170 247-251; subtitler, organisation of pro­
Specimen texts, 147 gramme, 254-259; 261-264; contents of
Spectra in translation theory, 152, spectral programme, 255-258, 261-269, 281-283
model, 151-155 Subtitlers, shortage of, 258
Speech analysis, 188, 190, s. construction, Subtitling, 272, 275-283, and tr. 263-264; s.
188, 189, from notes, ex. 191, ex. 203 machines, 258; s. in India, 26; advanced
Speech, spontaneous, 273 course in s. 261; types of (cinema vs. video
Speeches, videotaped, in int. 189 vs. TV), 275
Speed in translation, 307 Summary, 138, 146, 186, 187, 222-223,
Speed and machine tr., 307 263, 281
Spelling, 124, 144, 267 Supervision, 15
Spelling programmes, 130, 305 Supratitles, 276-277
Sperber, D. and D. Wilson, 247 Surtitles, 276-277
Spirituality, 30, 31 Sweden, 17, 258, 280
Spontaneous speech, 273, 276 Swedish, 60, 262
Srivastava, R.N., 27 Symbiosis, 25-36, in content, 30-31, at
Status of language professionals, 4-5, 20, linguistic level, 31
24; s. of tr. (in India), 26; s. of languages, Symbolic communication, 272
22; s. of participants in conversation, 218 Synchrony, 270-271, 276
Stay in foreign country, 178 Synonyms, 187, 209
Stenzl, C., 235 Syntagma, 296
Stereotype, 73, 82, 83
Strategies, translation, 98-105 Taboos, 258, 280
Strategies in subtitling, 282, for text produc­ Tag protection in machine tr., 306
tion, 84, of learning, 135; tr. s., 98-105, Taiwan, 183, 185
need for successful tr. s., 97, teaching dif­ Target audience needs, 152; t.a. and transla­
ferent s., 155; student/trainee vs. pro­ tion, 44
fessional s. in int., 226, 229-230 Target language norms, 91
Student attitudes to translation, 121-122, Target text function, 92; orientation, 238
135; s. expectations, 177; student individual­ Task of teaching, 151
ity, 125, 130, 179-180 (Also: 'Individualised Tasks in teaching, 107, 111, 146, 315;
feedback'); student time (outside class), 132 booklet with t., 124, 'authentic' t., 114
(Also: 'Study outside class'); student rights, (Also: 'Realistic tasks')
138-139; s. and machine interaction, 311 Taxonomy of transfer types, 270-271; t. of
Student-focus approach/process, 135-137 tr., 122-123
Student-teacher relationship, 131, 138, 148, TCIP, (def.) 211, 214
148-149, 178, 179-180, 213-214 Teacher personalities, 177-181; t. solution,
Students and industry, 257, 316 107, 144, 157, t. survival, 122; t. evaluation,
Study outside class. 132, 179, 189, 207, 209 180. Also: 'Student-teacher relationship'
Study programme (in Spain on advertise­ Teaching styles, 178-179
ments) 77-86, (in Uruguay), 133-134, (in Team work, 145-146, 278, 305, 307; t.w. in
Taiwan for int.) 185-198, (in US for mental drama, 280; t.w. in tr., 257
health int.) 212, (in Wales for subtitlers) Technical documents and machine tr., 304;
255-256, (in Denmark for subtitlers) 261- t. fields, 316; t. terminology, 296-299; in
263, (in Germany for technical tr.) 295, (in machine translation, 307-308; t. terms, 209;
US in machine tr.) 313-314 t. texts, 265, 267; t. texts for lay readers,
Style checkers, 305; s in learning int., 175- 153 (Also: 'Manuals'); t. tr. , 295-299, t. tr.
181; s. in teaching of int., 175-181 work, 304; technical, commercial tr. is, 308
Stylistics, 153 Technology, 256, 301-308, 309-316
Sub-language, 296-297 Technology in tr. training, 310-317
Subject orientation, 298 Technology-oriented translation programme,
Subjective perception, 234 310-317
 Index 355

Teletext, 253-254 ' Trans-manipulation', 154

Television programmes, 26, 261, 265 'Trans-metamorphosis', 154
Tempo, 238 Transcoding, 205
TEMPUS (= European student and teacher Transcription of int., 227, 231, 235-236
exchange programme), 14 . Transfer of meaning, 217; t. in subtitling,
Tentative Theory, 89-95 281; t. problems in subtitling, ex. 257-258
Terminology, 24, 110, 212-213, 234, 240, Transformation 159, 278, in subtitling, 278-
265, 307, 313; t. documentation, 240; t 279
management, 305; t. computerised, 305 Translatability, 34
Testing, 89-91 Translating literally. See: 'Literal trans­
Text in teaching without purpose, 65, 146 lation'
Text analysis, 77, t. building, 189-192, 203, Translation, (def.) 90, (def.) 100, (def.) 154;
296, 298; t. fragments, 298-299; t. produc­ tr. decisions, 43-44; tr. decisions, 315-316;
tion, 61, 77, 100-101, 146, 147, tr. as t. tr. drafts, 91-92; tr. process minimising
production, 145-146; t. types, 147, 297 error, 94-95; tr. process (model) 107-113; tr.
Text-world representation, 113-120 spectrum, 151-155; tr. strategies, 85, 144, tr.
Textbooks, 143, 155 strategies and 'top-down' procedure, 65-66
The Waste Land, 48 Translation studies and multilingual cultures,
Theory of Translational Action, 69 20, 22-24, tr. s. and the EU, 4-5; tr. tasks,
Theory in training, 107 144 (Also: 'Tasks').
Theory in translation, 94 Translation theory, 97-105, 261; tr. and
Think-aloud, 113-120,226 foreign language teaching, 121; tr. and int.,
Three-tier system in language acquisition (in components in study of, 184-185; tr. and
India), 26 quality, 98; Translation and Text Transfer,
Threshold age (for language acquisition), 21, 97
22 Translation as closure, 52; tr. as text produc­
Time factor in interlingual transfers, 269- tion, 145-146; tr. as a technical profession,
271; t. lag, 193, 194 98; tr. as theory, 90-91; tr. as a tool in
Timing, 267, 282 linguistics, 287-293
Title, ex. 62-65 Translation into standard language, 55-56
Tools (technological) for tr., 304-305 Translation, birth of, 160; tr., special fea­
'Top-down' procedure, 65-66 tures in, 145; tr., status of, 73, in India, 26-
Total reductions, 248-251 27, in Nigeria, 22-24; tr., types of, 270-271
Tourist brochures, ex. 60, 66 Translation, literary, 47-50;
Toury, Gideon, 25, 97, 121 Translation, literal vs. free, 152
Trani, E., 168 Translational competence (def.), 92, 92-93,
'Trans-', (as a general prefix) 153-155 conventions, 63, practice, 93, principles in a
Translation theory and subtitling, 261; tr. community, 93, shifts, 56
teaching and language teaching, 97; tr. as Translatology, 97, 98, 100, 101, 198, 287,
sequential process, 144 293
Tr. for communicating source language Translator as advisor to client, 72, 104-105,
grammar, 290-295; tr., levels in, 288-290; as reader, 52
tr.'s background knowledge, 72 Translator training, in multilinguistic and
Tr.s' reservations about computers, 301-302, multicultural settings 22-24, 34-37, 42-46;
306 for overcoming cultural barriers, 47-50; for
Trani, Eugen, 168 tolerance 51-57; in textual and functional
'Trans-adaption', 153 analysis, 59-68; for cultural mediation, 69-
'Trans-creation', 154 76; in textual, functional and cultural under­
'Trans-elucidation', 153 standing, 77-86; for heightened conscious­
' Trans-explication ',153 ness, 107-112,; for error elimination, 121-
'Trans-forming', 154 132; for error elimination and conscious­
'Trans-fusion', 153 ness, 133-141; for awareness of source
'Trans-imitation', 153 language peculiarities, 287-293; in technical
356 Teaching Translation and Interpreting 2

tr. 295; in machine tr., 309-316 Video films (as aids), 238
Translator training, objectives of, 42 Videotape(d performance), 180
Translator's choice, 247 (Also: 'Translation Vietnam, 14, 170
decisions'); tr.'s competence, 34-36; tr.'s Vietnamese, 211, 213
goal, 152 (Also: 'Intention'), tr.'s status, 22- Visual information, 236
24, 26-27, 73 Vocabulary, 207, 208, 209, 212-213, 215,
Translator's Workstation, 311-317 314-315
Trust in tr., 124. Also: 'Clients' Voice quality, 238, 280; v. q. in subtitling,
Truth, 90, 159, 160 265
Tuition, 255, individualised, 256 Voice-over, 270, 276
Turkey, 14 Waiting for Godot, 48, 49
TV film, 269, 270. Also: 'Subtitling' Wales, 253-259, 261, 282
TV programmes, 265 Walters, V., 171,172
Types of advertisement, 77, 79-85; t. of 'Warm' texts, 147-148
analysis, 77-78, of channel, 78; of errors in Welsh, 253-258
task, 130; of interpreter students, 178, 179- 'What'-'Why'-'How', 295
180; t. of monitoring in int., 230-231; of 'When' and 'where' factors, 154
screen translation, 269-272; texts for trans- Wills, W., 60, 194
fers according to semiotic types, 269-272; Word processing/ processors, 305, 306, 308,
transfer types, taxonomies of, 270-272 309, 310
Types of machine tr. systems, 309 Word-for-word translation, 154, 156
Work outside class, 132. Also: 'Study out-
UN. See 'United Nations' side class'
Undergraduate background (Danish), 121- Work stations, tr.s', 307-308
122 Working conditions of interpreters, 170-171,
Underlining of key terms, 209 225-227
Understanding source language peculiarities, Workshop, 255
288-293 Workstation, translator's, 176, in subtitling,
Understanding, 113-120, 267, limited, 113 262
Unit, 208, 238-240, 249, 296; u. in int. 193, World 1, World 2, World 3 (Popper), 93-95
in rendition, 230, in tr., 108, 110, 120; of World knowledge, 186, 187
discourse, 184; of sign, 187; textual u., 298 World-view, 49
United Kingdom, the, 11, 256, 259 Writing and translation, 146
United Nations, 12, 100 Written language in int., 242
Untranslatability, 25, 49-50, 55, 61, 62
USA, 170, 175-181,211-215 Yoruba, 19-24
Use, descriptive, 103, interpretive, 103
User expectations, 234 Aquivalenz, 59
User-orientation, 229
Users, 5

Vermeer, H., 69
Vernacular languages (as opposed to official
languages), 20, 21, 22
Versions, alternative, 152. Also 'Alternative
solutions'
 This is the book from

The Second Language International Conference

"Teaching Translation and Interpreting: Insights, Aims, Visions"

Together with the previous volume

Teaching Translation and Interpreting: Training, Talent and Experience

(Edited by Cay Dollerup and Anne Loddegaard.
Published by John Benjamins B.V. Amsterdam, 1992)

it prepares the common ground for:

The Third Language International Conference:

"Teaching Translation and Interpreting: New Horizons"
The conference takes place 12:00 noon, Friday 9 June to 14:00, Sunday 11 June, 1995
at the LO-Conference Centre, Elsinore, Denmark

Attn. Cay Dollerup

Center for Translation Studies and Lexicography
University of Copenhagen
DK-2300 Copenhagen S
Denmark
In the series Benjamins Translation Library the following titles have been published thus far or are
scheduled for publication:

68 DUARTE, Joao Ferreira, Alexandra ASSIS ROSA and Teresa SERUYA (eds.): Translation Studies at the
Interface of Disciplines. vi, 201 pp. + index. Expected September 2006
67 PYM, Anthony, Miriam SHLESINGER and Zuzana JETTMAROVÁ (eds.): Sociocultural Aspects of
Translating and Interpreting. viii, 251 pp. + index. Expected September 2006
66 SNELL-HORNBY, Mary: The Turns of Translation Studies. New paradigms or shifting viewpoints? 2006.
xi, 205 pp.
65 DOHERTY, Monika: Structural Propensities. Translating nominal word groups from English into German.
2006. xxii, 196 pp.
64 ENGLUND DIMITROVA, Birgitta: Expertise and Explicitation in the Translation Process. 2005. xx, 295 pp.
63 JANZEN, Terry (ed.): Topics in Signed Language Interpreting. Theory and practice. 2005. xii, 362 pp.
62 POKORN, Nike K.: Challenging the Traditional Axioms. Translation into a non-mother tongue. 2005.
xii, 166 pp. [EST Subseries 3]
61 HUNG, Eva (ed.): Translation and Cultural Change. Studies in history, norms and image-projection. 2005.
xvi, 195 pp.
60 TENNENT, Martha (ed.): Training for the New Millennium. Pedagogies for translation and interpreting. 2005.
xxvi, 276 pp.
59 MALMKJÆR, Kirsten (ed.): Translation in Undergraduate Degree Programmes. 2004. vi, 202 pp.
58 BRANCHADELL, Albert and Lovell Margaret WEST (eds.): Less Translated Languages. 2005. viii, 416 pp.
57 CHERNOV, Ghelly V.: Inference and Anticipation in Simultaneous Interpreting. A probability-prediction
model. Edited with a critical foreword by Robin Setton and Adelina Hild. 2004. xxx, 268 pp. [EST Subseries 2]
56 ORERO, Pilar (ed.): Topics in Audiovisual Translation. 2004. xiv, 227 pp.
55 ANGELELLI, Claudia V.: Revisiting the Interpreter’s Role. A study of conference, court, and medical
interpreters in Canada, Mexico, and the United States. 2004. xvi, 127 pp.
54 GONZÁLEZ DAVIES, Maria: Multiple Voices in the Translation Classroom. Activities, tasks and projects. 2004.
x, 262 pp.
53 DIRIKER, Ebru: De-/Re-Contextualizing Conference Interpreting. Interpreters in the Ivory Tower? 2004.
x, 223 pp.
52 HALE, Sandra: The Discourse of Court Interpreting. Discourse practices of the law, the witness and the
interpreter. 2004. xviii, 267 pp.
51 CHAN, Leo Tak-hung: Twentieth-Century Chinese Translation Theory. Modes, issues and debates. 2004.
xvi, 277 pp.
50 HANSEN, Gyde, Kirsten MALMKJÆR and Daniel GILE (eds.): Claims, Changes and Challenges in
Translation Studies. Selected contributions from the EST Congress, Copenhagen 2001. 2004. xiv, 320 pp. [EST
Subseries 1]
49 PYM, Anthony: The Moving Text. Localization, translation, and distribution. 2004. xviii, 223 pp.
48 MAURANEN, Anna and Pekka KUJAMÄKI (eds.): Translation Universals. Do they exist? 2004. vi, 224 pp.
47 SAWYER, David B.: Fundamental Aspects of Interpreter Education. Curriculum and Assessment. 2004.
xviii, 312 pp.
46 BRUNETTE, Louise, Georges BASTIN, Isabelle HEMLIN and Heather CLARKE (eds.): The Critical Link
3. Interpreters in the Community. Selected papers from the Third International Conference on Interpreting in
Legal, Health and Social Service Settings, Montréal, Quebec, Canada 22–26 May 2001. 2003. xii, 359 pp.
45 ALVES, Fabio (ed.): Triangulating Translation. Perspectives in process oriented research. 2003. x, 165 pp.
44 SINGERMAN, Robert: Jewish Translation History. A bibliography of bibliographies and studies. With an
introductory essay by Gideon Toury. 2002. xxxvi, 420 pp.
43 GARZONE, Giuliana and Maurizio VIEZZI (eds.): Interpreting in the 21st Century. Challenges and
opportunities. 2002. x, 337 pp.
42 HUNG, Eva (ed.): Teaching Translation and Interpreting 4. Building bridges. 2002. xii, 243 pp.
41 NIDA, Eugene A.: Contexts in Translating. 2002. x, 127 pp.
40 ENGLUND DIMITROVA, Birgitta and Kenneth HYLTENSTAM (eds.): Language Processing and
Simultaneous Interpreting. Interdisciplinary perspectives. 2000. xvi, 164 pp.
39 CHESTERMAN, Andrew, Natividad GALLARDO SAN SALVADOR and Yves GAMBIER (eds.): Translation
in Context. Selected papers from the EST Congress, Granada 1998. 2000. x, 393 pp.
38 SCHÄFFNER, Christina and Beverly ADAB (eds.): Developing Translation Competence. 2000. xvi, 244 pp.
37 TIRKKONEN-CONDIT, Sonja and Riitta JÄÄSKELÄINEN (eds.): Tapping and Mapping the Processes of
Translation and Interpreting. Outlooks on empirical research. 2000. x, 176 pp.
36 SCHMID, Monika S.: Translating the Elusive. Marked word order and subjectivity in English-German
translation. 1999. xii, 174 pp.
35 SOMERS, Harold (ed.): Computers and Translation. A translator's guide. 2003. xvi, 351 pp.
34 GAMBIER, Yves and Henrik GOTTLIEB (eds.): (Multi) Media Translation. Concepts, practices, and research.
2001. xx, 300 pp.
33 GILE, Daniel, Helle V. DAM, Friedel DUBSLAFF, Bodil MARTINSEN and Anne SCHJOLDAGER (eds.):
Getting Started in Interpreting Research. Methodological reflections, personal accounts and advice for beginners.
2001. xiv, 255 pp.
32 BEEBY, Allison, Doris ENSINGER and Marisa PRESAS (eds.): Investigating Translation. Selected papers from
the 4th International Congress on Translation, Barcelona, 1998. 2000. xiv, 296 pp.
31 ROBERTS, Roda P., Silvana E. CARR, Diana ABRAHAM and Aideen DUFOUR (eds.): The Critical Link 2:
Interpreters in the Community. Selected papers from the Second International Conference on Interpreting in
legal, health and social service settings, Vancouver, BC, Canada, 19–23 May 1998. 2000. vii, 316 pp.
30 DOLLERUP, Cay: Tales and Translation. The Grimm Tales from Pan-Germanic narratives to shared
international fairytales. 1999. xiv, 384 pp.
29 WILSS, Wolfram: Translation and Interpreting in the 20th Century. Focus on German. 1999. xiii, 256 pp.
28 SETTON, Robin: Simultaneous Interpretation. A cognitive-pragmatic analysis. 1999. xvi, 397 pp.
27 BEYLARD-OZEROFF, Ann, Jana KRÁLOVÁ and Barbara MOSER-MERCER (eds.): Translators' Strategies
and Creativity. Selected Papers from the 9th International Conference on Translation and Interpreting, Prague,
September 1995. In honor of Jiří Levý and Anton Popovič. 1998. xiv, 230 pp.
26 TROSBORG, Anna (ed.): Text Typology and Translation. 1997. xvi, 342 pp.
25 POLLARD, David E. (ed.): Translation and Creation. Readings of Western Literature in Early Modern China,
1840–1918. 1998. vi, 336 pp.
24 ORERO, Pilar and Juan C. SAGER (eds.): The Translator's Dialogue. Giovanni Pontiero. 1997. xiv, 252 pp.
23 GAMBIER, Yves, Daniel GILE and Christopher TAYLOR (eds.): Conference Interpreting: Current Trends
in Research. Proceedings of the International Conference on Interpreting: What do we know and how? 1997.
iv, 246 pp.
22 CHESTERMAN, Andrew: Memes of Translation. The spread of ideas in translation theory. 1997. vii, 219 pp.
21 BUSH, Peter and Kirsten MALMKJÆR (eds.): Rimbaud's Rainbow. Literary translation in higher education.
1998. x, 200 pp.
20 SNELL-HORNBY, Mary, Zuzana JETTMAROVÁ and Klaus KAINDL (eds.): Translation as Intercultural
Communication. Selected papers from the EST Congress, Prague 1995. 1997. x, 354 pp.
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 conference on interpreting in legal, health and
social service settings, Geneva Park, Canada, 1–4 June 1995. 1997. viii, 322 pp.
18 SOMERS, Harold (ed.): Terminology, LSP and Translation. Studies in language engineering in honour of Juan
C. Sager. 1996. xii, 250 pp.
17 POYATOS, Fernando (ed.): Nonverbal Communication and Translation. New perspectives and challenges in
literature, interpretation and the media. 1997. xii, 361 pp.
16 DOLLERUP, Cay and Vibeke APPEL (eds.): Teaching Translation and Interpreting 3. New Horizons. Papers
from the Third Language International Conference, Elsinore, Denmark, 1995. 1996. viii, 338 pp.
15 WILSS, Wolfram: Knowledge and Skills in Translator Behavior. 1996. xiii, 259 pp.
14 MELBY, Alan K. and Terry WARNER: The Possibility of Language. A discussion of the nature of language, with
implications for human and machine translation. 1995. xxvi, 276 pp.
13 DELISLE, Jean and Judith WOODSWORTH (eds.): Translators through History. 1995. xvi, 346 pp.
12 BERGENHOLTZ, Henning and Sven TARP (eds.): Manual of Specialised Lexicography. The preparation of
specialised dictionaries. 1995. 256 pp.
11 VINAY, Jean-Paul and Jean DARBELNET: Comparative Stylistics of French and English. A methodology for
translation. Translated and edited by Juan C. Sager, M.-J. Hamel. 1995. xx, 359 pp.
10 KUSSMAUL, Paul: Training the Translator. 1995. x, 178 pp.
9 REY, Alain: Essays on Terminology. Translated by Juan C. Sager. With an introduction by Bruno de Bessé. 1995.
xiv, 223 pp.
8 GILE, Daniel: Basic Concepts and Models for Interpreter and Translator Training. 1995. xvi, 278 pp.
7 BEAUGRANDE, Robert de, Abdullah SHUNNAQ and Mohamed Helmy HELIEL (eds.): Language, Discourse
and Translation in the West and Middle East. 1994. xii, 256 pp.
6 EDWARDS, Alicia B.: The Practice of Court Interpreting. 1995. xiii, 192 pp.
5 DOLLERUP, Cay and Annette LINDEGAARD (eds.): Teaching Translation and Interpreting 2. Insights, aims
and visions. Papers from the Second Language International Conference Elsinore, 1993. 1994. viii, 358 pp.
4 TOURY, Gideon: Descriptive Translation Studies – and beyond. 1995. viii, 312 pp.
3 LAMBERT, Sylvie and Barbara MOSER-MERCER (eds.): Bridging the Gap. Empirical research in
simultaneous interpretation. 1994. 362 pp.
2 SNELL-HORNBY, Mary, Franz PÖCHHACKER and Klaus KAINDL (eds.): Translation Studies: An
Interdiscipline. Selected papers from the Translation Studies Congress, Vienna, 1992. 1994. xii, 438 pp.
1 SAGER, Juan C.: Language Engineering and Translation. Consequences of automation. 1994. xx, 345 pp.