Translation As Intercultural Communication Selected Papers From The Est Congress Prague 1995 9789027216212 9781556197024 9789027285614 Compress

TRANSLATION AS INTERCULTURAL COMMUNICATION
BENJAMINS TRANSLATION LIBRARY
The Benjamins Translation Library aims to stimulate academic research and training
in translation studies, lexicography and terminology. The Library provides a forum
for a variety of approaches (which may sometimes be conflicting) in a historical,
theoretical, applied and pedagogical context. The Library includes scholarly works,
reference books, post-graduate text books and readers in the English language.
ADVISORY BOARD
Jens Allwood (Linguistics, University of Gothenburg)
Morton Benson (Department of Slavic, University of Pennsylvania)
Marilyn Gaddis Rose (CRIT, Binghamton University)
Yves Gambier (Centre for Translation and Interpreting, Turku University)
Daniel Gile (Université Lumière Lyon 2 and ISIT, Paris)
Ulrich Heid (Computational Linguistics, University of Stuttgart)
Eva Hung (Chinese University of Hong Kong)
W. John Hutchins (Library, University of East Anglia)
Werner Koller (Department of Germanic, Bergen University)
José Lambert (Catholic University of Louvain)
Willy Martin (Lexicography, Free University of Amsterdam)
Alan Melby (Linguistics, Brigham Young University)
Makoto Nagao (Electrical Engineering, Kyoto University)
Roda Roberts (School of Translation and Interpreting, University of Ottawa)
Juan C. Sager (Linguistics, Terminology, UMIST, Manchester)
María Julia Sainz (Law School, Universidad de la República, Montevideo)
Klaus Schubert (Technical Translation, Fachhochschule Flensburg)
Mary Snell-Hornby (School of Translation & Interpreting, University of Vienna)
Sonja Tirkkonen-Condit (Savonlinna School of Translation Studies, Univ. of Joensuu)
Gideon Toury (M. Bernstein Chair of Translation Theory, Tel Aviv University)
Wolfram Wilss (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 20
Mary Snell-Hornby, Zuzana Jettmarová and Klaus Kaindl (eds)
Translation as Intercultural Communication

Selected papers from the EST Congress - Prague 1995
TRANSLATION
AS INTERCULTURAL
COMMUNICATION
SELECTED PAPERS FROM
THE EST CONGRESS - PRAGUE 1995
Edited by
MARY SNELL-HORNBY
University of Vienna
ZUZANA JETTMAROVÁ
Charles University, Prague
KLAUS KAINDL
University of Vienna
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

EST Congress (1995 : Prague, Czech Republic)
Translation as intercultural communication : selected papers from the EST Congress,
Prague 1995 / edited by Mary Snell-Hornby, Zuzana Jettmarová, Klaus Kaindl.
p. cm. -- (Benjamins translation library, ISSN 0929-7316 ; v. 20)
Contributions in English, French, and German.
1. Translating and interpreting--Congresses. 2. Intercultural communication-Con-
gresses. I. Snell-Hornby, Mary. II. Jettmarovâ, Zuzana. III. Kaindl, Klaus. IV. Title. V.
Series.
P306.2.E86 1995
418'.02-dc21 97-21369
ISBN 90 272 1621 5 (Eur.) / 1-55619-702-0 (US) (alk. paper) CIP
© Copyright 1997 - John Benjamins B.V.
No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any
other means, without written permission from the publisher.
John Benjamins Publishing Co. • P.O.Box 75577 • 1070 AN Amsterdam • The Netherlands
John Benjamins North America • P.O.Box 27519 • Philadelphia PA 19118-0519 • USA
Contents
Preface
Part I: Translation — Sociology, Culture and Ethics
Theo Hermans
Translation as institution 3
Rosemary Arrojo
The "death" of the author and the limits of the translator's
visibility 21
Jean-Marc Gouanvic
Pour une sociologie de la traduction: le cas de la
littérature américaine traduite en France après la
Seconde Guerre mondiale (1945-1960) 33
Cay Dollerup
Translation as imposition vs. translation as requisition 45
Leo Tak-hung Chan

The impressionistic approach to translation theorizing, or:
Twentieth- century Chinese ideas of translation through the
Western looking-glass 57
Ubaldo Stecconi/Maria Luisa Torres Reyes
Transgression and circumvention through translation in the
Philippines 67
vi
Saliha Paker/Zehra Toska

A call for descriptive Translation Studies on the Turkish
tradition of rewrites 79
Sirkku Aaltonen
Translating plays or baking apple pies: A functional approach
to the study of drama translation 89
Marta Mateo
Translation strategies and the reception of drama performances:
a mutual influence 99
Rainer Kohlmayer
From saint to sinner: The demonization of Oscar Wilde's
Salomé in Hedwig Lachmann's German translation and in
Richard Strauss' opera Ill
Michaela Wolf
Translation as a process of power: Aspects of cultural
anthroplogy in translation 123
Mira Kadric/Klaus Kaindl
Astérix — Vom Gallier zum Tschetnikjäger: Zur Problematik
von Massenkommunikation und übersetzerischer Ethik 135
Andrew Chesterman
Ethics of translation 147
Part II: Translation and Beyond — Aspects of Communication
Erkka Vuorinen
News translation as gatekeeping 161
Veronica Smith/Christine Klein-Braley
Advertising — A five-stage strategy for translation 173
Zuzana Jettmarova/Maria Piotrowska/Ieva Zauberga
New advertising markets as target areas for translation 185
vii
Ingrid Kurz
Getting the message across — Simultaneous interpreting
for the media 195
Franz Pöchhacker
"Clinton speaks German": A case study of live broadcast
simultaneous interpreting 207
Udo Jörg
Bridging the gap: Verb anticipation in German-English
simultaneous interpreting 217
Irena Kovačič
A thinking-aloud experiment in subtitling 229
Paul Kussmaul
Comprehension processes and translation.
A think-aloud protocol (TAP) study 239
Sigrid Kupsch-Losereit
Übersetzen als transkultureller Verstehens- und
Produktionsprozeß 249
Hanna Risku
Von Scheuklappen, Mikroskopen und Fernrohren: Der Umgang
mit Wissen in der Entwicklung der Übersetzungskompetenz 261
Renate Resch
Ein kohärentes Translat — was ist das? Die Kulturspezifik
der Texterwartungen 271
Michèle Kaiser-Cooke
Murder in the laboratory - Termhood and the culture gap 283
Dorte Madsen
A model for translation of legal texts 291
Irene Rübberdt/Heidemarie Salevsky
New ideas from historical concepts: Schleiermacher and
modern translation theory 301
viii
Anneke de Vries
A matter of life and death: Gender stereotypes in some
modern Dutch Bible translations 313
Part III: Panel Discussions
Christina Schäffner/Beverly Adab

Translation as intercultural communication —
Contact as conflict 325
Daniel Gile/José Lambert/Mary Snell-Hornby
EST Focus: Report on research training issues 339
List of Contributors 351

Preface
The first International Congress of the European Society for Translation

Studies (EST) after its foundation in Vienna in 1992 was held at the Charles
University, Prague from 28-30 September 1995. The setting was particularly
appropriate: the Charles University, founded in 1348 as the oldest university
in Central Europe, has a distinguished tradition of pioneering scholarship and
international cooperation. In the Middle Ages it was a meeting-place for
scholars from all over Europe: in the 1990s it has become a forum for scholars
from all over the world. The EST Congress attracted participants from 35
countries (26 in Europe, 9 from overseas) extending from Canada and Brazil
to the Philippines. For such an audience the basic theme of the Congress
seemed equally apt: Translation/ Interpreting as Intercultural Communication.
The academic programme was presented within a framework of workshops,
round table discussions and themes, some of them extremely topical:
Translation and Advertising, Screen Translation, Media Interpreting,
Translation and Colonialism, Multimedia and Mass Communication.
The selection of 30 contributions presented here reflects the broad range
of topics debated at the Congress. The papers are arranged in two main
sections (Part 1: Translation — Sociology, Culture and Ethics; Part 2:
Translation and Beyond — Aspects of Communication), within which they
centre round various focal themes, from stage translation to simultaneous
interpreting, from the power structures and societal forces behind translation
to the thought processes and know-how within it. The basic theme which runs
through the entire volume, the intercultural and communicative aspect of
translation in the world today, emerges sharply in the workshop summarized
in Part 3; the volume closes with observations on the promotion of research
training, one of the major objectives of EST.
X
It is conspicuous that most of the contributions are written in English,

although only a few of the authors are English native speakers: the reason is
partly the specific wish of the authors concerned and partly the requirement of
the publishers to ensure a maximum readership. The Editors have endeavoured
to respect the variations now accepted for English as an international language,
without imposing any one particular variety.
We would like to thank the contributors for their cooperation in preparing
the manuscript for publication, and our special thanks are due once again to
John Benjamins Publishers, both for accepting this volume for publication and
for their usual efficiency and friendly assistance.
Vienna/Prague, March 1997
Mary Snell-Hornby
Zuzana Jettmarová
Klaus Kaindl
Part I
Translation — Sociology, Culture and Ethics

Translation as institution
Theo Hermans
Let us imagine the following situation as real. We find ourselves in a packed

lecture theatre in Charles University, Prague. In one corner of the room there
is — why not? — a bird's nest. In it, there lives a goldfinch. We all know
about birds, and when this one leaves the nest, as it is about to do now, we
can all see it clearly, and we recognize it: yes, undoubtedly, a goldfinch.
Look, there it is, ready to go. It peeks around, and now it takes off, flying
across the room. What a lovely bird, this little goldfinch. And then, suddenly,
in mid-air, it explodes, right here, before our very eyes.
How now? A perfectly ordinary goldfinch, and it exploded, just like that?
How do we respond? Let's review some of the options at our disposal. We can
doubt our observation. Is there something wrong with our eyesight, or were
we hallucinating? But no, we are quite all right, wide awake and perfectly
sober. Perhaps someone shot it. But no, there was no sound of shooting, no
smoking gun, and we were watching our goldfinch intently: it exploded all by
itself. Perhaps it was not a goldfinch after all. But we know enough about
birds to be pretty sure that this was a real bird, and a goldfinch at that —
carduelis, by its Latin name. Perhaps it was a hitherto unknown subspecies of
goldfinch, carduelis auto-explodens, for instance. Maybe that should remain
a possibility, even if a rather remote one. Or perhaps we should simply forget
the whole incident, put it aside as one of those unclassifiable occurrences, like
seeing a UFO; in which case we reserve judgment, and hope that things can
be left at that.
To recapitulate: we started with a relatively straightforward if not
altogether common situation, and then something wholly unpredictable,
something outrageous happened. Our problem was: how to go on? How to
make the world normal again, how to come to terms with the flagrant
anomaly, how to remove the uncertainty that suddenly challenged and
destabilized our image of the world?
Before we continue I wish to stress that I did not invent this goldfinch. It
lives, and briefly explodes, in the pages of J.L. Austin's essay Other Minds,
4 Translation as Intercultural Communication
in his collection Philosophical Papers (1961). Austin's essay goes on to

explore a series of challenging questions, such as: how do we know what goes
on in someone else's mind, and what is meant when someone says something
is 'real'. But as I am interested here in much more pedestrian issues, there is
no need to follow Austin's philosophical flights. My interest concerns initially
the fact that when faced with something so outrageous as an exploding
goldfinch, and having satisfied ourselves that it is not our observation which
is at fault, we tend to respond in one of basically two ways:
1. we can be flexible, and try to attune our mental picture of the world to the
empirical reality we observe, for example by postulating a new subspecies
of goldfinch; or
2. we can decide to keep our world view intact by dismissing the anomaly,
and discounting the paradox it created.
In the first case we adopt a learning attitude. This means that it is we who
change, in that we seek to incorporate the new experience into our world
picture by adjusting that picture so as to accommodate the new reality. That
is, we adjust our expectations about the world and the range of likely
occurrences in it to the possibility of another occurrence like the apparently
odd one we just observed. In the second case we refuse to learn from
experience. We pretend that the incident did not really take place at all and
leave it at that. At best we accept its reality for just this once. We let it pass,
and hope it will not happen again. This allows us to stick to our beliefs and to
carry on as we were, despite the facts, which, we decide, really should not
have happened in the first place.
Both attitudes are ways of coping with the unexpected, with
unpredictability. More precisely, they are ways of responding to situations in
which the world did not behave as expected, in which our expectations about
the world were disappointed. Now, as the German sociologist Niklas Luhmann
has pointed out, because we know that strictly speaking almost anything can
happen in the world, at more or less any moment, it is good to be prepared for
the possibility of our expectations being disappointed, and to have an
appropriate response ready. This is what the two attitudes are about. In fact,
because in everyday life our expectations are routinely disappointed, we are all
entirely familiar with such response mechanisms, both on a large and on a
small scale.
The first attitude, the adaptive, learning one, is a matter of having
cognitive expectations. This is an attitude which will try to build and revise and
forever rebuild hypotheses about the world. If our mental scheme of things
falls out with the world, we change the scheme. This mode of expectation
Theo Hermans 5
tends towards the world of science. The other attitude, which is unwilling to
learn, corresponds to a normative expectation. It actually provides more peace
of mind because it is more stable. It is 'counterfactually stable' in that
disappointments, anomalous occurrences, flagrant breaches do not really upset
it. It carries on regardless. More than that: following disappointment it may
emphatically and publicly reaffirm its validity. This mode of expectation tends
towards the law, which, as we know, remains intact despite crimes being
committed daily (Luhmann 1984:436ff.).
And now to translation. What if we are happily reading a translation, and
stumble upon a real howler, a glaring, totally unacceptable anomaly, something
wholly incompatible with our expectations of what a translated text should be
like, of what constitutes 'translation'?
Let's be honest. We all know exactly how we respond. We respond with
indignation and condemnation. We say: "Wrong!", "Unacceptable!". We say:
"Do they call this translation?" — by which we mean: I don't, therefore it
isn't, and anyone who knows anything about translation will surely agree with
me (making it doubly hard to disagree); and if the fancy takes us we resolutely
set to work on the text with a red pencil, or write to the publisher, or phone
the translation agency.
In doing so we are emphatically upholding and reaffirming our idea of
'translation', what it is and what it evidently is not, and at the same time we
are appealing to a publicly recognized and acknowledged category, both a
concept and a practice, to which this translation should be made to correspond
if it is to be accepted as a valid translation. It is in this sense, then, that we can
speak of a social entity called 'translation' and a form of behaviour called
'translating' with which, give or take a few nuances, we reckon we are all
familiar in our own language and culture. The meaning of the term
'translation' is codified in dictionaries, there are professional activities called
translation, we have organisations representing translators, institutes for
translator training, etc. It is this 'public face' of translation that I have in mind
when I speak of translation as 'institution'.
There is nothing new in this. What I want to stress however is, firstly,
that translation, as institution, is circumscribed by expectations which have
both cognitive and normative elements in them; secondly, that, beyond this,
these expectations also structure the 'domain', or the 'field', or indeed the
'system' of translation, in the sense in which Niklas Luhmann speaks of
expectations as constituting the structure of social systems (1984:377ff.); and
thirdly, that since we are dealing with translation, the matter is more
complicated than it looks.
Let us begin by returning to Luhmann. Why Luhmann? Niklas Luhmann
is a sociologist. He has applied concepts from modern systems theory to a vast
number of things including education, religion, politics, law, the sphere of
intimate personal relations, art history, science, everything except translation.
His main work to date is called Social Systems (Soziale Systeme, 1984; English
translation 1995), and what a magnum opus it is: vast, abstract, theoretical,
and all the more forbidding for it. The point however is that, if we want to
understand translation as interpersonal communication, as social behaviour, as
institution, we could do worse than to look at someone like Luhmann — or,
for that matter, someone like Pierre Bourdieu. Of course, it all depends on
where one stands and where one wants to go in the Babylonian universe of
discourse that we call translation studies. I can only speak for myself. My
agenda is defined primarily by non-applied, historicizing approaches to
translation and an active interest in how translations and ideas about translation
relate to their socio-cultural environment, why translators act as they do, why
they tend to make certain choices and not others. Speaking from that vantage
point it seems to me that among the reasons for attempting to draw fresh
inspiration and new perspectives from Luhmann's work are the following:
• his use of modern systems theories, i.e. theories of self-reproducing, self-
regulating, self-referential systems;
• his application of these theories to the domain of social and cultural
studies, and to historical issues;
• his use of what appears to me as a very rich concept of communication;
• his view of expectations — i.e. cognitive as well as normative
expectations — as constituting the structure of social systems.
In other words, there is enough there that strikes a chord, and plenty to learn
from. It does not invalidate anything that has been done before in the study of
translation, but it casts new light on old perspectives, and raises other,
intriguing questions.
In what follows I want to draw on three aspects of Luhmann's work, in
the hope of being able to add a few new touches to the way in which we
approach the world of translation. The first has to do with expectations, norms,
and decisions. If there is something new in this by now well-worn perspective,
it is the emphasis on the translator as an active decision-maker within complex
structures of power. The second aspect picks up Luhmann's concept of
communication, and may have methodological consequences which, again,
without being completely new, may serve to shift the focus of some of our
historical research, or at least act as a reminder. The third bears on self-
reference and self-reflexiveness, and has, I think, profound implications for the
Theo Hermans 7
way in which we perceive the nature and status of the discourse about
translation, including, perhaps even especially, our own academic discourse on
the subject.
Conventions and norms
We can now, very briefly, take up the notion of norms and normative
expectations again, bearing in mind that when I speak of a norm I do not mean
some abstract, static, formal or mechanical rule which relates to the practice
of translation as cause relates to effect, of the sort: if this is the feature
displayed in our text, then that must have been the norm that triggered it.
Rather, what I mean by a norm is neither more nor less than a kind of loaded
expectation. The term therefore implies, in the case of translation, structured
interaction between individuals, as clients, patrons, producers, consumers,
teachers or critics of translation. The reason for emphasizing the aspect of
expectation, in contrast with the more traditional stress on the 'rule' character
or the element of 'constraint' in norms, will become clear as we go along.
My basic assumption in all this is that translation, like any other use of
language, is a matter of communication, i.e. a form of social behaviour which
requires a degree of interaction, of cooperation, among those involved. For
communication to take place here and now, as you the reader are reading these
words on the page, both you and I have to coordinate our actions to a certain
extent, even across time and space (e.g. you, at your end, in being prepared
to take in what I have written, in being willing to read on; I, at my end, in
choosing a particular language, in trying to express myself in such a way that
I think you will be able to follow my argument). Norms, like conventions,
arise as answers to problems of this kind of interpersonal coordination.
The classic definition of convention (Lewis 1969) hinges on exactly this
point. Conventions, as defined by Lewis, imply the expectation, which all
those concerned are aware of, that in a given situation one member of the
group is likely to do one thing rather than another. A family may develop the
pattern, for example, that when they start playing a certain board game, the
youngest child always takes the first turn at throwing the dice. After a while
that is indeed what every family member expects to happen whenever this
particular board game is played. If it works well, the expectation may even be
transferred to other games. What the example illustrates is that the convention
has a regulatory function. It restricts the number of practically available
options in recurrent situations of a given type by offering a particular option
as the one known to be preferred by everyone involved. In so doing, in
promoting coordination, the convention makes everyone's behaviour more
predictable by reducing uncertainty and contingency. In this family there will
be no more fighting about who is allowed to have the first go.
The main difference between a norm and a convention lies in the modality
of the expectation. A convention is a purely probabilistic expectation. Norms
tell individual members of a community not just how everyone else reckons
they are probably going to behave in a given situation, but how they ought to
behave. Norms imply that there is, among the range of possible options that
present themselves, a particular course of action which is generally accepted
as 'proper', or 'correct', or 'appropriate'. That course of action, it is agreed,
should therefore be adopted by all who find themselves in that type of
situation. And each time a norm is observed, its validity is confirmed and
reinforced.
In fact, even if it is not observed by all, the norm remains valid. Provided
the breaches do not occur persistently and on a large scale without sanctions
of one kind or another, norms are able to cope with a certain amount of
discrepant, erratic, idiosyncratic behaviour. The conventions and norms that
govern, say, our behaviour when we pick up the telephone, or at a dinner
party, during a funeral service, during an academic lecture, are not invalidated
every time someone fails or refuses to behave in the way everyone else expects
everyone else to behave.
Which norms are observed or broken by whom, where and when, will
depend on such things as the nature and strength of the norm, the kind of
sanction that might apply, the individual's status and motive, and other such
factors. When, in the 1960s, Louis and Celia Zukofsky rendered Latin poetry
into English mimicking only the sound of the words, to the exclusion of just
about everything else, it is relevant to know that this was done in a literary
context, that even in that domain it was generally interpreted as a provocative
— i.e. a deliberately norm-breaking — gesture, that already at this time Louis
Zukofsky was widely recognized as a prominent poet in his own right, etc. The
newly graduated translator who has just been given a job in the European
Union's fisheries department and wants to make a career in the EU is unlikely
to be able to afford to follow the Zukovskys' example.
Norms, then, can be strong or weak, limited or extensive in scope, more
or less enduring over time. They can spell out obligations or prohibitions, and
exert different kinds of pressure on the choices which individuals make. We
can map these modalities (it has been done, e.g. De Geest 1992), and they give
us an insight into differences between well-defined areas governed by clear
Theo Hermans 9
rules understood by all (do this, don't do that) and backed up by explicit
sanctions and rewards, and more fluid areas governed by vaguer, less
pronounced or more permissive or mutually conflicting norms. If the EU's
translation division is an example of the first, the world of modern poetry is
perhaps an example of the second.
We acquire norms by learning them. They are inculcated as part of the
process of socialization. Just as learning to speak is learning to speak
'properly', in accordance with the linguistic norms of the relevant community
(the family, the circle of friends, the school, the workplace), so learning to
translate means learning to operate the norms of translation, i.e. to operate
with them and within them, anticipating, accommodating, calculating,
negotiating the expectations of others concerning the social institution called
translation. In the same way readers learn what they can and cannot expect
when they pick up a book labelled 'translation'. What this means is that, on
both sides of the equation, in fact on all sides since the production and
consumption of translation involves more than two parties, certain bonds,
certain contracts are entered into. They may be clearly set out and understood
by all concerned, or remain rather diffuse. And who controls whom is a
question of power and position. An experienced and well-established translator
of science fiction may feel more confident than the young aspiring novice in
ignoring the wishes and suggestions of a particular editor or publisher. A
contract between a professional translator and a client creates relations of
obligation and claim very different from those between, say, teacher and
student at a translator training institute.
In other words, norms are not independent of local conditions, and of the
social relations within communities, regardless of whether these relations are
material (economic, legal, financial) or what Pierre Bourdieu calls "symbolic",
i.e. relations that have to do with status, with legitimacy, and with who confers
legitimacy. Of course, in large, complex and differentiated societies, a vast
multiplicity of different, overlapping and often conflicting norms coexist. The
translator's work is inevitably entangled in several of these networks at once,
if only because the product of the translator's labour is never a 'translation per
se', but it is a translated computer manual, or a translated novel, or a
translated medical record, etc. In each case the translator enters an existing
network of discourses and social relations. His or her translational discourse
occupies a place in, or at least in relation to, that network. It is part of the
ambivalence of the translated text that it is expected to comply with both the
translational and the textual norms regarded as pertinent by a given community
in a given domain. If the translation does this, because the translator has made
the requisite choices, it will be deemed a 'legitimate' translation.
Learning to translate correctly, then, means precisely the acquisition of
that competence, i.e. of the skills required to select and apply those norms that
will help to produce legitimate translations, that is to say translations socially
recognized as legitimate within a certain community and its concept of
translation. Translating is a socially regulated activity. Communities and social
systems have good reason to regulate translation in this way. Seen from the
consumer or the receptor point of view, or more generally from the point of
view of the translating system, translation typically involves the importation,
and as a rule the linguistic domestication, of material from outside the system.
Since the environment of the system is always more complex and hence less
predictable than the system itself, translation is regulated because this makes
it possible for the system to reduce complexity and to control contingency.
Furthermore, just as one of the main functions of the educational system
as a whole is that of transmitting the requisite social skills, expectations and
'dispositions' (in Bourdieu's sense), continually reproducing and reaffirming
the community's dominant values and models in the process, so, in the field
of translation, one of the roles of the translator training institute consists
precisely in continually reproducing within itself the social institution called
translation, which in turn contributes to the very process of the
institutionalization of translation. Let me add immediately that other discourses
about translation, including so-called descriptive and historicizing discourses,
do very much the same thing, if perhaps less emphatically. They all contribute
to the ongoing self-reproduction of translation, and for that matter to its self-
description.
I will come back to this issue, but let me try to sum up the main point so
far. Which is this: considering the practice of translation in terms of
expectation and contingency, of shared expectation and individual selection in
the context of a multiplicity of conflicting and overlapping norms and pressures
allows us to bring into focus not just the social, institutionalized aspect of
translation, but also the individuals themselves who weave their way through
and around these complex structures, who take up positions and build alliances
so as to be able to achieve their own aims and ambitions — be it making
money, or buttressing or subverting a political system or an ideology, or
acquiring literary fame, or, as has been the case with many women translators
in history, finding a voice when they were prevented from speaking in their
own name. In each of these cases, translators make choices, i.e. they select
one option from among the range of more or less practicable, more or less
Theo Hermans 11
likely options available to them in the circumstances. In other words, the focus
is here firmly on the agents involved in the whole process of the production
and consumption of translation. The operation of translational norms is then
not a matter of texts, or of textual relations, but of acting, thinking, feeling,
calculating, sometimes desperate people, with certain personal or group
interests at heart, with stakes to defend, with power structures to negotiate.
Sense and self-reproduction
There is another angle from which to approach the issue, and this line of
thought will bring us back to the matter of norms and expectations, and to
ways of studying translators and their behaviour. To set out this approach, I
need to refer to Luhmann's concept of communication. Let it suffice for the
purposes of the present exposition to say that for Luhmann social systems are
self-reproducing systems in that they continually produce and reproduce the
elements of which they consist. These elements are communications, i.e.
communicative acts. In other words, social systems consist, not of individuals
or of groups of people, but of communications, and of specific types of
communication. These communications are for the most part produced and
processed by means of signs; they also have to be linked and connected in a
temporal sequence for the system to continue to exist. There are no social
systems without communication, but at the same time communications are
momentary, fleeting phenomena, here one moment and gone the next. This
explains the need for connectivity, for structures that can endure over time.
Luhmann's notion of communication differs from e.g. the Saussurian
notion mainly in that it stresses the aspect of enunciation, the utterance itself,
the actual presentation of the information in a particular context. Luhmann's
term for it is "Mitteilung", which he contrasts with "Information", the
referential data as such (1984:191ff.). Secondly, whenever something is
communicated, in whatever way, the communicative act itself involves an act
of selection: this topic, this theme is picked out and not that one; this means
of transmission is chosen, at this exact moment, and not another one, or this
one at another moment.
Because communication takes place at a certain moment, in a given
context, 'interpreting', 'understanding' it — Luhmann calls this "Verstehen"
— means that making 'sense' (Luhmann: "Sinn") of a communication involves
being alive not only to the 'theme' and the 'code' of the communication, but
also to its selective aspect, its negative foil, the difference between what has
been included (i.e. selected) and what has been excluded, the difference also
between the information that is conveyed and the moment that has been
selected to convey it. This is precisely what constitutes the 'point' of a
communication, its 'intent', and what the receiver rightly or wrongly
constructs as its 'sense' — in a given situation, at a certain moment in time.
One Luhmann commentator (De Berg 1993: 50) speaks of the "temporalization
of semantics", a useful phrase.
A crude example might be the national anthem. The British do not intone
their national anthem because they seriously wish to beseech God to save their
Queen, not even because they fervently hope their Queen will be preserved
long enough to keep Charles from the throne, but as part of a ritualized
ceremony, or as a political gesture, or as a provocation hurled at another
bunch of football hooligans, etc. The communication derives its 'sense' not,
or not only, from the information content of the words by themselves but from
the context which makes it more or less likely that these particular words are
selected at this or that particular moment.
Texts therefore have no fixed meaning in themselves. They acquire
meaning, they are invested with meaning as communications in a selective
environment, a differential context. And, Luhmann maintains, because contexts
are always historically unique, meaning is prevented from sliding into a
Derridean labyrinth of "différence" and deferral. When we look at texts in this
way, through their "temporalized semantics", we may be able to glimpse the
speaker's agenda: how likely was this communication in these particular
circumstances, why was this theme, and this mode of transmission, selected at
this moment, how does it alter the existing state of affairs?
In itself, of course, this way of looking at texts is not entirely new. In his
short essay Man and Language from 1966 Hans-Georg Gadamer already
observed that "[n]othing that is said has its truth simply in itself, but refers
instead backward and forward to what is unsaid... And only when what is not
said is understood along with what is said is an assertion understandable"
(1977:67). We have also come across it, for example, in pragmatics, and in
the literary and cultural semiotics of Yury Lotman, who stressed in the 1970s
that in literature every mode of representation means the selection of one
particular mode against the possibility of other modes — and he too used this
approach to firmly locate texts, conventions and expectations in their historical
context. Luhmann goes perhaps further than this in that his whole concept of
communication is historicized.
If this is true of texts, it is true also of translations. Their 'meaning' or
'sense' or 'point' as communications, their differential agenda, does not reside
Theo Hermans 13
in 'the words on the page', decipherable by means of linguistic and other

codes, in a social vacuum. Nor can it be reduced to some semantic or other
relation with a source text. In both cases, what is left out is precisely the
aspect of selectivity, of selective difference. It is part of the 'sense' of a
translated text that this and no other foreign-language text was selected for
translation (and not for some other form of transmission or importation), at this
moment, in this context; and that a particular 'translational mode' was
selected, one particular mode of representing the original against the possibility
of other available modes, i.e. in relation to the alternatives not chosen from
among the array of more or less likely candidates, more or less permissible
modes.
Phrases like "the array of more or less likely, more or less permissible
alternatives" are worth noting, however. They take us straight back to
expectations ("more or less likely") and to normative expectations ("more or
less permissible", in a given context of claim and obligation, that is), and
hence to translation as institution. For it is clear that we are talking about
expectations within a limited range of options. The domain of translation, i.e.
of that which is termed 'translation', has limits, a socially acknowledged
boundary differentiating it, sometimes sharply, sometimes only diffusely, from
other modes of representing anterior discourses such as paraphrase, adaptation,
plagiarism, summary, quotation, transliteration, and so on. These expectations,
which police the boundaries of translation as institution, are usually referred
to as the 'constitutive norms' of translation. If you breach them you are
perceived as doing something which is not called 'translation', at least not by
the group that sees itself as being addressed and as having a legitimate claim
to the definition of 'translation'. In that sense we can speak of these expect-
ations as circumscribing the domain of translation. They are comparable to
what the history of science has referred to as the "institutional imperatives" of
scientific practice (the term is Robert Merton's), those social norms —
including e.g. disinterestedness and originality — which serve to ensure the
normal functioning of scientific communities. Within the field of translation it
is customary to speak of 'regulatory norms', meaning normative expectations
concerning what is appropriate in certain cases, regarding certain types or
areas of discourse. These expectations form the structure of the translational
system, in a sense compatible with Luhmann. Luhmann holds that whereas
social systems consist of communications in that communications are the
elements the system is made of, expectations about communications constitute
the structure of social systems. Social structures are structures of expectation
(Luhman 1984:139, 377ff.). Structure here means precisely that some
occurrences and some combinations are more likely than others. If all
occurrences and all combinations were all equally likely, this would produce
entropy in the system.
Both sets of expectations, and their respective normative loads, are
continually negotiated and confirmed by practising translators and by all who
are recogized as having a legitimate claim to discourse about translation. In
that sense we can speak of translation as a self-reproducing system.
Communication generates communication — under the right conditions. We
can translate because there are translations and because, when we translate or
speak about translation, we routinely take account of the conditioning factors
which govern the institution called 'translation' in our culture.This creates the
necessary connectivity and a sufficient 'horizon of expectations' to produce
further translations and statements about translation.
To the extent that the translational system continually reproduces itself,
then, it has its own momentum, its identity and relative stability as a system.
This means that whatever its entanglement in other systems or series, it
interprets its environment in terms of its own interests and priorities. The way
in which translation is constructed as a social institution governed by particular
sets of expectations therefore determines the way in which we translate and
read translations. But this also leads to interesting complications, which may
well be specific to translation, or at least to translation as practised and
theorized in a good part of the Western tradition. Let's pause for a moment on
one such complication.
We can assume that translation as a rule involves the crossing of systemic
as well as semiotic boundaries in that communications which were outside the
system's orbit on account of their being differently coded are transformed, by
means of translation, into communications using a code intelligible in the
receptor system. While the change from an unfamiliar to a familiar code entails
a degree of domestication, the norms of translation in much of our tradition —
by no means the entire tradition, but let that pass — forbid a radical
transformation of the original text. Rather, translation is usually taken to stand
as a full-scale representation of the source text, a substitute for it in the sense
of speaking for it, in its place, as delegated speech. Translation functions as
both replica and proxy. This is the conception which informs what Brian
Harris posited not so long ago as a "fundamental and universal" norm of
professional interpreting in particular, but there is every reason to regard it as
an "institutional imperative" applicable to other forms of translating as well,
perhaps especially to translating culturally prestigious text. For Harris, the
"true interpreter" or "honest spokesperson" norm "requires that people who
Theo Hermans 15
speak on behalf of others, interpreters among them, re-express the original

speakers' ideas and manner of expressing them as accurately as possible and
without significant omissions, and not mix them up with their own ideas and
expressions" (1990:118). In many countries, and at the international level,
copyright laws are indeed there to enforce this very norm. While the discretion
and non-interference expected of the translator is evidently intended to
guarantee the integrity of the source, it restricts — among other things — the
degree of permissible adaptation of the translated text to the communicative
conditions in the host system. Very rarely, for example, will the original text's
orientation towards an implied reader' be radically altered in translation.
Translation has to live with contradictory impulses as a result. While translated
texts are intended to slot into new contexts, this relocation tends to be only
partial because the true interpreter's norm of non-interference sets a limit to
the degree of integration. Hence the well-known observation that translated
texts usually do not fit their new environment, their new space, as snugly and
naturally as fully home-grown, non-translated texts. In his essay Man and
Language Hans-Georg Gadamer touched on exactly this peculiarity of
translation. Having posited, as was indicated above, that all saying operates
against the background of what is not said, so that "only when what is not said
is understood along with what is said is an assertion understandable"
(1977:67), Gadamer went on to observe that in a translation the original
is reflected on one level, so that the word sense and sentence form of the
translation follow the original, but the translation, as it were, has no space. It
lacks that third dimension from which the original (i.e. what is said in the
original) is built up in its range of meaning. [....] No translation is as
understandable as the original. Precisely the most inclusive meaning of what is
said — and meaning is always a direction of meaning — comes to language only
in the original saying and slips away in all subsequent saying and speaking.
(1977:68)
To my mind, Gadamer overstated his case, for at least two reasons. Firstly,
translations are made in response to or in anticipation of demand at the
recipient pole and this provides translated texts with their communicative
space, even though that space is necessarily different from that of the source
text. Secondly, notions like "dynamic" or "communicative" equivalence (Nida
1964; Newmark 1981) signal precisely a degree of deliberate adaptation of the
translated text to the host cultural environment. Nevertheless, while it may be
unjustified to claim that "the translation [...] has no space", it remains true that
more often than not the "institutional imperatives" of translating prevent the
translated text from being wholly reorientated towards the receptor pole's
communicative atmosphere. The result affects the communicative position of

translations, more particularly what Gadamer calls the "third dimension" and
what for Luhmann would be the selective aspect in the enunciation itself and
in the choice of topic to be communicated. Because translated texts as a rule
do not shed their original orientation to the communicative context of the donor
culture, their status as communications in the host environment is inherently
more complex because always double-edged, both domestic and other.
Self-reference and metalanguage
A similar duality, or ambivalence, pertains to discourses about translation. As

I suggested earlier, the expectations pertinent to translation are also continually
being explicitated and transmitted by teachers, critics, publishers, researchers,
etc. as part of the self-description of translation. All of these communications
together, at the primary level and at the meta-level, belong to the 'institution'
of translation, the 'translational system'. Self-reference is thus part of the
system, of the way the system differentiates itself from its environment (i.e.
other systems) and continually reminds itself of this difference.
But just as translations themselves always belong to more than one series,
since they are invariably inserted into or at least perceived in relation to other
discourses, so the scholarly, critical, educational and academic metalanguages
of translation, too, are part of more than one series. These metalanguages of
translation may be part of the self-description of translation, but they do not
belong to the institution of translation only. They have their other foot, so to
speak, in the world of education and/or research, and in various specific
disciplines like linguistics, semiotics, or literary studies. What I mean by this
is not the bland and rather hollow observation that translation studies is an
'interdiscipline', whatever that means, but something altogether more
problematical.
Let me try to illustrate this by means of an example. All of us are familiar
with Roman Jakobson's essay On Linguistic Aspects of Translation of 1959. In
it Jakobson famously distinguished between three kinds of translation. The list
has been the subject of comment by a number of critical luminaries and
theoreticians, from Gideon Toury to Jacques Derrida. Jakobson's three kinds,
as we know, are
• intralingual translation, or rewording
• interlingual translation, or translation proper
• intersemiotic translation, or transmutation.
Theo Hermans 17
Toury (1986) wondered why Jakobson listed two kinds of 'lingual'

translation (intra- and inter-) as against only one semiotic kind (inter-), and
suggested we should add intrasemiotic translation to the list. He was right.
However, I want to focus on another aspect. The paradox in Jakobson's listing,
which Derrida (1985:173ff., 217ff.) astutely identified and exquisitely
commented on, is this: if intralingual translation is a form of translation, then
in Jakobson's own essay the term "rewording" is a translation of the term
"intralingual translation". In this way the first and the third term in the list are
both translated intralingually: "intralingual translation" is rendered as
"rewording", and "intersemiotic translation" is reworded as "transmutation".
But the middle term, "interlingual translation", is not reworded or intra-
lingually translated. It is merely repeated, tautologically restated: this form of
translation is translation, "interlingual translation" is "translation proper". The
addition of the qualifier 'proper' suggests moreover that the other two are not
'properly' translation. This, it will be appreciated, undermines the whole
exercise of ranging them all three together as so many kinds of translation.
Derrida went on from there to question the apparent transparency of
notions like translation, language, etc. I am interested here in the question why
the paradox is there in the first place. The answer, it seems to me, lies in the
fact that Jakobson's essay is anchored in at least two different domains. As a
linguistic or, more properly, a semiotic statement, i.e. a statement uttered by
a professional linguist or semiotician, Jakobson's claim that rewording and
transmutation can also be regarded as forms of translation, is entirely
acceptable, even though in 1959 its 'sense', its 'point' was certainly a good
deal sharper, more loaded, than it is today. Semiotics was in those days a
heady, expanding, ambitious discipline, and the grounds on which Jakobson
extended the familiar 'standard' concept of translation 'proper' to intralingual
and intersemiotic operations were in line with academic, scholarly practice in
linguistics and semiotics. From the point of view of someone professionally
engaged in the study of sign systems there is no reason to restrict the study of
translational phenomena to interlingual translation, to the exclusion of
intralingual, intersemiotic or for that matter intrasemiotic forms.
But Jakobson's own phrasing gave the game away. Seen from the vantage
point of the social institution of translation, from the vantage point, that is, of
what the non-academic community is prepared to call translation, the move is
not permissible because there translation is translation proper, and only that.
Jakobson's formulation itself obviously recognized this. Its unease stems from
ambivalence and transgression in declaring both that translation properly
understood means interlingual translation only and that translation encompasses
other transformational operations not normally covered by the term

'translation'. To which we can add that, speaking from today's vantage point,
we can also appreciate Jakobson's statement as being part both of the self-
description and self-reflexiveness of translation, in questioning precisely the
boundaries of the field and thus engaging in the discussion about what is and
what is not translation, what is inside or outside the system, and as being part
of the emerging academic discipline of translation studies.
The Jakobson example goes to underscore the point that we may gain
something if every statement and every claim made in the context of translation
studies is read with this kind of tension in mind. These statements and claims
form part of several domains at once: those of scholarship, education,
research, academia, etc., and that of translation as social institution. The dual
context in which we produce our various meta-discourses of translation is
always there, irrespective of whether we want, self-reflexively, to mark the
distance between object-level and meta-level, as descriptive studies prefer to
do; or whether we actively anticipate the norms, rules and conventions that
govern the practice of professional translating in the market-place, as applied
translation studies tend to do; or whether indeed we are intent on harnessing
the scholarly authority of our critical reflection to work changes in the practice
of translation, as e.g. Laurence Venuti, or Tejaswini Niranjana, or some
gender-oriented researchers often do.
What exactly do we gain by reminding ourselves of the dual context in
which the discourse about translation is located? More perhaps than we
bargained for. It seems to me that the observation has far-reaching
consequences, which I will briefly try to indicate by way of conclusion.
1. The recognition that the academic discourse about translation is itself
rooted in the institution of translation renders the separation between object-
level and meta-level profoundly problematical. This is especially the case if we
accept Jakobson's claim that rewording constitutes, however improperly, a
form of translation. For in discoursing about translation we are constantly
rewording, i.e. translating intralingually, translation itself. The discipline of
translation studies, especially in its descriptive 'pure research' guise, can be
a scholarly discipline only if it can distance itself from its object of study, but
it is always contaminated by it in having to repeat the operation it attempts to
insulate itself from. The problem has been highlighted before, notably by
deconstructionist critics (e.g. Bakker 1995) sceptical of the very possibility of
separating object-level from meta-level. Perhaps the recognition that the
discourse about translation is inherently and necessarily ambivalent in that it
pertains both to the self-reflexiveness of translation and to the discursive
Theo Hermans 19
practices of other disciplines, allows us at least to pinpoint the issue, and to be

less nervous about it. The paradox can then be resolved by tracing the process
of functional differentiation within the complex system of translation as
institution. The more academic metalanguages of translation do translate
translation intralingually, but they differentiate themselves from "translation
proper" by their orientation towards the other scholarly discourses of the
humanities. Translation studies is then a subsystem of the system of translation,
and its guiding difference as a subsystem lies in its discursive orientation.
2. To say that our statements about translation arise out of 'translation as
we know it' is to recognize also that our knowledge about translation is
culture-bound. This, of course, we knew all along. It also follows from the
previous point. The question only becomes acute when we try to speak about
'translation' generally, as a universal given and therefore supposedly present
in all cultures; or when we wish to understand what another culture means by
whatever term they use to denote an activity or a product that appears to us to
translate as 'translation' — whereby we naturally translate that other term
according to our concept of translation, and into our concept oftranslation; and
in domesticating it, we inevitably reduce it. I hasten to add that these large
questions regarding the very possibility of knowing the Other go well beyond
the scope of the present paper and far beyond my competence. Moreover, they
affect several branches of the human sciences and all hermeneutics. But for
translation studies, professionally concerned with Otherness, they are
particularly acute, and we would do well to reflect on them. Perhaps also we
should take more deliberate note of disciplines such as anthropology and ethno-
graphy, which in the face of similar problems of what is there called "cultural
translation" have become very self-conscious, self-critical, and therefore self-
reflexive (see e.g. Clifford & Marcus 1986; Tambiah 1990). That does not
make the problem go away, but it guards against a form of rashness that
ignores its own ethnocentricity and translates all translation into 'our'
translation, instead of patiently, repeatedly, laboriously negotiating the other's
terrain, while trying to conceptualize our own modes of representation and the
commensurability of cultures.
3. Finally, closer to home, if there is insight, and therefore benefit, to be
derived from holding translations up against their negative foils so that their
'sense', their 'import' as communicative acts may be glimpsed through their
differential selectivity, their temporalized semantics, their hidden agenda, then
surely this must also apply to statements about translation, i.e. to those
communications that together constitute this thoroughly problematical discipline
of translation studies. Here too what is said is always more than — or at least
different from — what the words add up to. That this also goes for everything
I have been saying here, goes without saying.
References
Austin, J.L. 1961. "Other Minds". In: Philosophical Papers, Oxford: Clarendon, 44-84.
Bakker, Matthijs. 1995. "Metasprong en wetenschap: een kwestie van discipline". In: D.
Delabastita & T. Hermans (eds.) Vertalen historisch bezien, 's-Gravenhage: Stichting
Bibliographia Neerlandica, 141-162.
Berg, Henk de. 1993. "Die Ereignishaftigkeit des Textes". In: H. De Berg/M. Prangel (eds.)
Kommunikation und Differenz. Opladen: Westdeutscher Verlag, 32-52.
Bourdieu, Pierre. 1994. Raisons pratiques. Sur la théorie de l'action. Paris: Seuil.
Clifford, James/Marcus, George (eds.) 1986. Writing Culture. The Poetics and Politics of
Ethnography. Berkeley: University of California Press.
Derrida, Jacques. 1985. "Des tours de Babel". In: J. Graham (ed.) Difference in Translation.
Ithaca/London: Cornell University Press, 165-248.
Gadamer, Hans-Georg. 1977. Philosophical Hermeneutics. Transi. David Inge. Berkeley:
University of California Press.
Geest. Dirk de. 1992. "The Notion of 'System': Its Theoretical Importance and its
Methodological Implications for a Functionalist Translation Theory". In: H. Kittel (ed.)
Geschichte, System, Literarische Übersetzung /Histories, Systems, Literary Translations.
Berlin: Erich Schmidt, 32-45.
Harris, Brian. 1990. "Norms in Interpretation". Target 2:1, 115-119.
Hermans, Theo. 1996. "Norms and the Determination of Translation: A Theoretical
Framework" In: R. Álvarez/C.-À. Vidal (eds.) Translation, Power, Subversion.
Clevedon: Multilingual Matters, 25-51.
Jakobson, Roman. 1959. "On Linguistic Aspects of Translation". In: R. Brower (ed.) On
Translation. Cambridge, Mass.: Harvard University Press, 232-239.
Krawietz, W. & Welker, M. (eds.) 1992. Kritik der Theorie sozialer Systeme.
Auseinandersetzungen mit Luhmanns Hauptwerk. Frankfurt: Suhrkamp.
Lewis, David. 1969. Convention. A Philosophical Study. Cambridge (Mass.): Harvard
University Press.
Lotman, Yury. 1972. Die Struktur literarischer Texte. Transi. R.-D. Keil. München: Fink.
Luhmann, Niklas. 1984. Soziale Systeme. Grundriß einer allgemeinen Theorie. Frankfurt:
Suhrkamp.
Luhmann, Niklas. 1995. Social Systems. Transi. John Bednarz Jr. Stanford: Stanford
University Press.
Luhmann, Niklas. 1986. "Das Kunstwerk und die Selbstreproduktion der Kunst". In: H.U.
Gumpert/K.L. Pfeiffer (eds.) Stil. Geschichte und Funktionen eines
kulturwissenschafilichen Diskurselements. Frankfurt: Suhrkamp, 620-672.
Newmark, Peter. 1981. Approaches to Translation. Oxford: Pergamon Press.
Nida, Eugene A. 1964. Towards a Science of Translating. Leiden: E.J. Brill.
Niranjana, Tejaswini. 1992. Siting Translation. History, Post-Structuralism and the Colonial
Context. Berkeley: University of California Press.
Tambiah, S.J. 1990. Magic, Science, Religion and the Scope of Rationality. Cambridge:
Cambridge University Press.
Toury, Gideon. 1986. "Translation: A Cultural-Semiotic Perspective". In: T. Sebeok (ed.)
Encyclopedic Dictionary of Semiotics vol. 2. Berlin: de Gruyter.
Toury, Gideon. 1995. Descriptive Translation Studies and Beyond. Amsterdam: Benjamins.
Venuti, Lawrence. 1995. The Translator's Invisibility. A History of Translation. London/New
York: Routledge.
The "death" of the author and the limits of the
translator's visibility
Rosemary Arrojo
In the essentialist opposition which tradition has built between reading and
writing, and between originality and reproduction, translation has not been
merely associated with secondariness and failure. In its long history of
marginality and invisibility, particularly in a culture that often equates
authorship with property and writing with the conscious interference of a
producer, the translator's activity has been related to evil and blasphemy, to
indecency and transgression. In its obvious pretension — even when
understated — to take the place of another and to represent someone else's
voice in a foreign language and culture, in a different time and space, any
translation is bound to raise questions not only of property but, first and
foremost, of propriety. Even the "perennial question whether translation is, in
fact, possible" is basically embedded in "ancient religious and psychological
doubts on whether there ought to be any passage from one tongue to another"
(Steiner 1975:239). If the conscious presence of the author is somehow
expected to be found in her or his writing, and if the original is seen as the
true recipient of its creator's intentions and expression, any translation is, by
definition, devalued since it necessarily represents a form of falsification,
always removed from the original and its author.
While the original is generally associated with stability, with what is
present, primary and authentic, a translation is often related to precariousness
and the absence of what is unconditionally legitimate. If one thinks of the
Tower of Babel as a framework for a reflection about the complex relationship
which tradition has woven between original and translation, one can say that
the original is idealized and related to creation, while translation is associated
with the limits, shortcomings, and inadequacies of what is purely human and
with what is, ultimately, improper. In the well-known myth, the condemnation
to translation is nothing but the expression of God's anger at men's daring
pretension to reach the divine and to have the right to bear their own names.
Thus, God punishes "the sons of Shem" for having wanted "to make a name
for themselves, [...] to construct for and by themselves their own name, to
gather themselves there [...] as in the unity of a place which is at once a
tongue and a tower, the one as well as the other, the one as the other" (Derrida
1985:169). The divine condemnation to translation is, in other words, also the
repression of men's authorial will and of their desire to bear a proper name.
In the plot which tradition has constructed for the relationships which can be
established between translation and original, between translator and author, or
between the translated text and its readers, the translator's name and
interference have been condemned either to oblivion or to disdain by a
conception of originality and of text firmly rooted in a theological basis.
From the perspective of certain trends in contemporary thought and,
particularly, of deconstruction, which Jacques Derrida explicitly associates with
translation (1985:165f.), the typical notions of originality, authorship, and
interpretation are radically revised, as is our reading of the myth of Babel. In
this context, the goals and the incompleteness of men's construction, that is,
the precariousness and the underlying authorial goals of interpretation, begin
to be recognized as that which is essentially human. The acceptance of the
impossibility of reaching any pure origin, or that which could be immortal,
univocal and beyond any perspective, is, thus, also the acceptance of the
inevitability of interference in any act of alleged re-creation. The recognition
of the far-ranging implications of this paradigmatic role of translation, or of
interpretation as translation, which has become a key concept in contemporary
theories of language, culture and the subject, is also one of the inaugural
premises of what has been generally known as postmodernism and which has
been closely linked to Nietzsche's and to Freud's intellectual heritages. The
acceptance of the inevitability of translation as interference is thus intimately
related to the death of God and of the Cartesian subject, and, therefore, also
to the death of the author and of authorship as the definite, controlling origin
of meaning.
The "death" of the author and the birth of the reader
The implications of such "deaths" for a reflection on the translator's task are
obviously far-reaching and potentially revolutionary. Indeed, an important
trend in the contemporary discussion of the translator's visibility raises the
issue of a different relationship between translation and original, translator and
Rosemary Arrojo 23
author, a relationship which we might associate, for example, with the birth
of the reader celebrated by Roland Barthes as a direct consequence of the death
of the author. In the wake of poststructuralist theories of intertextuality,
reading begins to be recognized as "an anti-theological activity, an activity that
is truly revolutionary since to refuse to fix meaning is, in the end, to refuse
God and his hypostases — reason, science, law" (1977:148). For Barthes, the
one place where the "multiplicity" of writing is "focused" is no longer the
author, but the interpreter, or the reader, who becomes "the space on which
all the quotations that make up a writing are inscribed without any of them
being lost": the reader "is simply that someone who holds together in a single
field all the traces by which the written text is constituted." Thus, "we know
that to give writing its future, it is necessary to overthrow the myth: the birth
of the reader must be at the cost of the death of the Author" (1977:161). In the
explicitness of this ambitious Oedipal move, which seems to overcome any
deeply embedded guilt associated with the interpreter's authorial designs and
to end the age-old oppression of an impossible neutrality, "text" is redefined
precisely as that which is read "without the father's signature: it can be
broken, [...] it can be read without the guarantee of its father, the restitution
of the inter-text paradoxically abolishing any legacy" (1977:161). Hence, no
"vital respect" is owed to the author who becomes a mere "limit," or a "guest"
that may, or may not, be invited to the reader's productive reading act. The
author "becomes, as it were, a paper-author: his life is no longer the origin of
his fictions but a fiction contributing to his work" (1977:161).
Relieved of the impossible mission to recover the ultimate origins of a text
and free from the age-old inhibition of flaunting his or her authorial will, the
reader begins to be recognized as an active producer of meaning whose
interference is not merely tolerable but inevitable. However, as the interpreter
becomes "that someone who holds together in a single field all the traces by
which the written text is constituted," that is, as the interpreter becomes a
somewhat stable origin of meaning who can apparently choose what to do with
the author, this privileged position in which Barthes places his reader does not
seem to be very different from the traditional conception of authorship which
has been questioned by poststructuralist theories of language. If
poststructuralism necessarily brings about the death of traditional authorship,
how can the reader born of such theorization be protected from the inevitability
of intertextuality? How could a textual theory promote the birth of a rather
strong, apparently omnipotent reader at the same time that it declares the death
of the omniscient author? In other words, one might say that even though a
new reader was definitely born of non-essentialist theories of language and the
subject, such a birth is inescapably marked by the "deaths" which have made
it possible. That is, in order to be coherent, the postmodern reader has to be
an interpreter whose production is necessarily inscribed within the same
process of différance which deconstructs the possibility of any stable, definite
origin, and which allows us to accept and explore the inevitable transgression
represented by any act of interpretation.
The notion of abusive fidelity
A similar paradox seems to underlie certain notions defended by a few

contemporary translation theories which explicitly declare their compatibility
with poststructuralism or postmodernism in general, and with Derridian
deconstruction in particular, as they justify and celebrate the translator's
visibility or conscious interference in the translated text. Among such theories,
Philip Lewis's notion of "abusive fidelity" has been explicitly recognized as
influential by translators and theoreticians such as Lawrence Venuti (1992;
1995), Lori Chamberlain (1988), Suzanne Jill Levine (1991), Sharon Willis
(1992), and Luise von Flotow (1991).l For Lewis, the "translatability that
emerges in the movement of difference as a fundamental property of
languages" "points to a risk to be assumed": that of a translation which accepts
and exhibits its authorial force, that is, of a translation that "values
experimentation, tampers with usage, seeks to match the polyvalencies or
plurivocities or expressive stresses of the original by producing its own." This
"risk" implies a translation strategy inspired by what Lewis refers to as "the
productive difference consisting in that twist or skewing signaled by the prefix
ab that is attached to the dominant c(h)ord of use" (1985:41). Similarly, for
Venuti, the insights brought about by contemporary theories of language and
text can establish different bases for the relationships that have always
approximated and yet separated originals and their translations. Consequently,
as he argues, the fact that any translation can only be "an interpretive
transformation" "releases [it] from its subordination to the foreign text and
makes possible the development of a hermeneutic that reads the translation as
a text in its own right, as a weave of connotations, allusions, and discourses
specific to the target-language culture" (1992:8). Thus, translation seems to
give up its generally humble and impossible pretension at being transparent or
1
For a commentary on the basic contradictions involved in the notion of "abusive fidelity"
with regard to feminist translations, see Arrojo (1994; 1995).
Rosemary Arrojo 25
invisible and becomes "strong" and "forceful," requiring, as Venuti concludes,

a "more sophisticated translation strategy that acknowledges the complications
poststructuralism has brought to translation, particularly the concept of
meaning as a differential plurality" (1992:12).
Yet, according to Lewis, such a strategy of explicit abuse does not
sacrifice "the faithful transmission of messages to playful tinkering with style
and connotation." Moreover, "fidelity" and "intelligibility" remain intact and
are even reinforced because, in the wake of Derrida's deconstruction, in which
"the clear-cut separability of signifier and signified, of force and meaning, is
dismantled", what is at stake is "a new axiomatics of fidelity, one that requires
attention to the chain of signifiers, to syntactic processes, to discursive
structures, to the incidence of language mechanisms on thought and reality
formation, and so forth," and which implies "fidelity to much more than
semantic substance, fidelity also to the modalities of expression and to
rhetorical strategies" (1985:42). As Lewis recognizes, there is an explicit
contradiction in his intricate logic. Once he proposes a form of abuse which
is, in reality, a form of respect, or merely a "reproduction of the original
abuse," what is in fact the new contribution of a theory of translation
supposedly inspired by poststructuralism? As Lewis himself asks,
if the aggressive translator merely falls into a classic form of complicity,
whereby, for example, deviation serves to ground and sustain the norm, then
why all the fuss about abuse? Maybe this is just the same old trap, well known
to the most conventional theories of translation, that Benjamin derides in "The
Task of the Translator" (1985:43f.).
For Lewis, the impasse is simply resolved by creativity, that is, by "assuming
the contradiction and attempting to make something of it. " Such creativity
implies a "method" that would "focus on a paradoxic imperative: how to say
two things at once, how to enact two interpretations simultaneously?" If, as he
proposes, the project of translation is ultimately impossible since it implies an
impasse of being faithful to the target language while "nonetheless resurrecting
a certain fidelity" to the original, we are certainly back to the "same old trap,"
as he himself concedes (1985:44).
Like the reader imagined by Barthes, Lewis's abusive translator does not
seem to take to their final consequences the very insights which have made
possible the recognition of the inevitable interference of interpretation. While
Barthes's reader seems to be (impossibly) free to do whatever he or she pleases
with the author of the text to be read and becomes a rather stable source of
meaning, Lewis's translator seems to presuppose that his or her interference
may be a fully conscious option, and that he or she knows exactly what the
original is about and thus can decide what should be respected, or abused. In
this sense, Lewis's translator is not very different from the translator idealized
by tradition who is torn between his or her authorial interests and the proper
respect owed to the author and should know exactly what the original is about.
Like the Barthesian reader who seems to take the place of the controlling
author and is allegedly able to produce a reading that can remain faithful to
itself, Lewis's aggressive translator does not seem to consider the fact that his
or her abusive translation will inevitably be subject to other readings and
interpretations and will thus be transformed (and abused) by them. However,
unlike Barthes's reader, who seems to have been set free from any inhibition
to interfere and occupy the position formerly attributed to the powerful author,
Lewis's aggressive translator is still cautious to exercise his or her authorial
thrust and apparently hides his or her intervention under the guise of a
paradoxically abusive fidelity.
It seems that the only way out of this paradox would involve a true
acceptance of the implications of poststructuralism and of Derridean
deconstruction, that is, a true acceptance of difference (and différancé) in
translation which necessarily transforms one text into another, one language
into another, be it a reading, or a translation. If the controlling author is
inevitably dead, that is, if writing can be read without its author's presence or
absent approval, and if no interpretation can ever aspire to truly recover
meaning simply because no meaning is ever present or immanent but is always
already a production, any act of translation is bound to be a transformation, "a
regulated transformation of one language by another, of one text by another"
(Derrida 1987:20). Therefore, what Lewis considers to be "the modalities of
expression" and "the rhetorical strategies" of the original, i.e., what he sees
as belonging to the original and which he proposes to respect in his translation
is inevitably a reflection of his own reading, the inscription of his interference
as a reader and interpreter, which determines what to emphasize and what to
respect, or disrespect, in the so-called original. Just as Barthes's theory of
reading does not seem to take into account that, after the deconstruction of the
notion of traditional authorship as the absolute, controlling origin of meaning,
any act of reading, like any piece of writing, will necessarily be circumscribed
by its own context and history, and will, therefore, also be given to inter-
textuality and to différance, Lewis's notion of abusive fidelity seems to
disregard an implicit, basic consequence of its own argument, that is, that the
fidelity to the original that the aggressive translator intends to maintain cannot
be anything more than a fidelity to his or her own version of what the original
might be about or like, as well as to the goals and the context of his or her
Rosemary Arrojo 27
translation. If we accept the impossibility of perfect repetition, then any

translation is always, and inevitably, also a form of abuse, or transgression,
"insofar as it falls within the violence of its own alterities: untruth in truth,
misrepresentation in representation, resistance in power, desire dividing itself'
(Ross 1990:37). To a certain extent, no matter how good or bad a translation
may be considered, it is always and inevitably unfaithful, since it is sure to
take the place of another, in a different language and culture, and in a different
time. If we accept the translator's inevitably authorial interference and the
transformation that any interpretation always implies, it does not seem to be
theoretically coherent — or even useful — to keep the age-old notion of
fidelity to the original, not even the contradictory definition proposed by
Lewis, which still bears a certain subserviency to the impossible ideal of
perfect reproduction entertained by tradition.
Even a superficial examination of what the proponents of abusive fidelity
describe as their translation practices suggests that the kind of relationship
which they consciously establish with the authors and the texts they translate
could hardly be described in terms of what is generally understood as fidelity,
or infidelity, and truly escapes the middle ground apparently proposed by
Lewis. Lawrence Venuti, for instance, explicitly establishes a relationship with
the authors he translates which is quite different from the traditional model that
expects the translator "to participate vicariously in the author's thoughts and
feelings" and to produce a translated text "which is read as the transparent
expression of authorial psychology or meaning" (1995: 274). In contrast to this
ideal of invisible translation, Venuti proposes his brand of abusive fidelity,
which he calls "resistancy." In his commentary about his translations of Milo
De Angelis's poetry, Venuti explains that his strategy of resistancy "refuses
fluency" and seeks "to reproduce the discontinuity of De Angelis's poem"
(1995: 290). At the same time, however, by adopting such a strategy, he has
been "unfaithful to — and [has] in fact challenged — the dominant aesthetic
in the target-language culture, i.e., Anglo-American culture, becoming a
nomad in [his] own language, a runaway from the mother tongue. " Moreover,
"implementing this strategy must not be viewed as making the translation more
faithful to the source-language text" either, since his version "deviates from the
Italian text in decisive ways that force a radical rethinking of fidelity in
translation" (1995:291). "Resistancy" is thus a "translation strategy" by which
De Angelis's poems become strange to the Italian poet, as well as to the
Anglo-American reader and translator. It is certain that De Angelis will not
recognize his own voice in the translations, not only because his ideas and
texts would seem to make such a way of reading unthinkable for him, but
also because he is unable to negotiate the target language (1995:300).
Even though Venuti does not recognize his own authorial voice in his
translations of De Angelis's poetry, he does concede that his versions are
"transformations" of the originals, being truly faithful neither to the originals,
nor to the target culture and tradition. Yet, such transformations are not
arbitrary, nor willfully predatory; they follow a carefully thought out logic and
have very specific, explicit goals and motivations. They are, so to speak,
performative, as they transcend the mere initiative to make De Angelis's poetry
available in English. As Venuti writes, his translations "resist the hegemony
of transparent discourse in English-language culture [...] by deterritorializing
the target language itself, questioning its major cultural status by using it as the
vehicle for ideas and discursive techniques which remain minor in it, which it
excludes" (1995:305). At the same time that such translations transform their
originals, such a process is regulated not only by the translator's explicit
interests and motivations but also by "information about De Angelis's readings
in literature, literary criticism, and philosophy" (1995: 292). In Venuti's
translation project, De Angelis is definitely not the absolute, controlling
authority to which the translation should be blindly faithful, nor is he simply
a "guest" which the translator decides to bring to his interpretation whenever
and however he pleases. In such a context, De Angelis is not, in any way,
being perversely abused as a poet but is, definitely, one of the organizing
principles that directs and inspires the translator's options. Thus, even Venuti's
explicit refusal of fluency in translation "takes its cue" from De Angelis's own
aesthetic" which "questions whether the translator can be (or should be thought
of as being) in sympathy with the foreign author" (1995:286, 290). The result
of this "collaboration" between Venuti and De Angelis produces a translation
that "resists the transparent aesthetic of Anglo-American culture which would
try to domesticate De Angelis's difficult writing by demanding a fluent
strategy" at the same time that it "creates a resistance in relation to De
Angelis's text, qualifying its meaning with additions and subtractions which
constitute a 'critical thrust' toward it" (1995:291f.).
The limits of the translator's inevitable visiblity
Thus, an appropriate.poststructuralist model for the translator's inescapably

authorial task is not to be found in Lewis's conception of abusive fidelity, nor
in Barthes's apparently liberating notion of reading. Such a model might very
Rosemary Arrojo 29
well be found in the full implications of Derrida's conception of translation as

regulated transformation, which presupposes the inevitable "impropriety" of
interpretation and gives up the attempt to please (or to appease) an author
whose allegedly controlling and potentially punitive powers have already been
deconstructed. At the same time, we may supplement and refine such a model
with Michel Foucault's conception of authorship, according to which the author
is no longer feared as "an indefinite source of signification," and begins to be
recognized as a fundamental regulating element, "a certain functional principle
by which, in our culture, one limits, excludes, and chooses; in short, by which
one impedes the free circulation, the free manipulation, the free composition,
decomposition, and recomposition of fiction," "the principle of thrift" which
represses and somehow controls the "proliferation of meaning" that constitutes
any text (1979:159). In such terms, one might say that Venuti's translations are
ultimately faithful to that which he is interested in associating with his author's
name as well as to the goals and the theoretical and ideological framework that
have inspired and directed his work. Within the conscious universe of his
work, which explicitly "resists" "the dominant Anglo-American literary values
that would domesticate the Italian texts, make them reassuringly familiar, easy
to read" (1995:302), Venuti's translations have not really abused De Angelis's
poetry as they have been neither faithful nor unfaithful to it. They have,
nevertheless, taken possession of such poetry and transformed it in order not
only to make it available in English, but also to say and do something about
what he considers to be the current Anglo-American translation standards. As
he openly resists current translation standards and expectations, and as he
explores and accepts his inevitable visibility and authorial interests, he can
(appropriately) see De Angelis's originals and his own translations as "distinct
entities," "determined by different factors, serving different functions, leading
different discursive lives" (1995:292).
If we follow such conclusions to the ultimate consequences, we will also
have to accept the fact that visibility is not simply a conscious option of the
translator who is theoretically and ideologically motivated to oppose tradition
and essentialist textual theories. Even the traditional translation strategy of
allegedly intentional invisibility or transparency necessarily reveals a certain
conception of what the text is about and a theoretical, ideological perspective
of what should be done in order to make it available in another language and
culture. As Venuti recognizes, transparency is at most "an illusionistic effect"
which "depends on the translator's work with language," at the same time that
it "hides this work, even the very presence of language, by suggesting that the
author can be seen in the translation, that in it the author speaks in his or her
own voice" (1995:287) The transparency idealized by tradition is not exactly

a neutral, ethical stance which any conscientious translator will have to adopt;
it is, rather, a strategy that necessarily serves certain interests. To the extent
that "the effect of transparency masks the mediations between and within copy
and original," it also eclipses "the translator's labor with an illusion of
authorial presence, reproducing the cultural marginality and economic
exploitation which translation suffers today" (Venuti 1995:290).
If the interference of interpretation is not exactly an option but rather the
inevitable consequence of the deconstruction of absolute originality, as well as
the mark of any relationship between subjects, we will also have to accept the
fact that the translator's options and interpretations are not simply present in
the translated text, nor objectively recoverable by its readers and critics. The
relationships which other readers will establish with Venuti's translations, for
example, are not within the scope of the latter's intentions and conscious
control. As he himself recognizes, while commenting on readers' reports he
received after submitting the complete manuscript of his translations of De
Angelis's poetry to a few editors, his "resistant strategy was strange" to all of
his readers who had expected "transparency" in translation and who preferred
"the sort of fidelity [generally related] to univocal meaning and smooth
prosody" (1995:300-303). In brief, what Venuti cannot accomplish as a
translator is precisely that which De Angelis cannot accomplish as a poet, or
as an author. Although the death of the author has brought about the birth of
the interpreter, both are equally condemned to translation, that is, to that
incompleteness which subjects authors, translators, interpreters, and readers
alike to the interference represented by someone else's interpretation, to the
infinite possibilities of différance which are potentially set in motion when a
reader approaches a text, be it an original or a translation.
If the death of traditional authorship implies the birth of the reader and the
acceptance of the interpreter's inevitable visibility, and if the translator that is
thus born cannot be omnipotent and cannot control what will be done with the
goals and options of his or her translation, the most important consequence
poststructuralism could bring to translation studies is precisely a thorough
revision of the relationships that have generally been established between
originals and translations, between authors and translators, and between
translators and their readers, which are no longer adequately described in terms
of the traditional notions of meaning recovery, fidelity or equivalence. If the
author is no longer seen as an "indefinite source of significations," but, rather,
as a "function," or as "the ideological figure by which one marks the manner
in which we fear the proliferation of meaning" (Foucault 1979:159), the
Rosemary Arrojo 31
consciously visible translator should start to build a name, a "proper" name for
him or herself that would make his or her readers aware of the "translator-
function"2 as another key factor in the necessary repression of meaning
proliferation that takes place in any act of interpretation. Furthermore, the
validation of the translator's voice as a legitimate interference in the translated
text will only be truly able to start making a difference when visibility begins
to be marked by the signature of his or her own authorial name. It is the
recognition and the acceptance of this name which can open the space for the
possibility of a "translator-function" as a regulating element that necessarily
and legitimately determines meaning in the relationship which a reader will
establish with a translated text. If, as Foucault writes, "texts, books, and
discourses really began to have authors [...] to the extent that authors became
subject to punishment, that is, to the extent that discourses could be
transgressive" (1979:148), the recognition that translation is in fact a form of
meaning production should be accompanied by the recognition of the
translator's function as that "ideological" element by which, in the reading of
translated texts, one marks certain forms of relationships between original and
translation, as between what is foreign and what is domestic. It is the
recognition of the translator's name as proper and rightful that will free the
translator's visibility from the stigma of impropriety or abuse. In the wake of
poststructuralism and postmodernism, the visible translator's claim to bear his
or her own name may finally begin to change the age-old prejudices that have
always ignored or humiliated the production of meaning that constitutes the
inescapable task of any translation.
References
Arrojo, Rosemary. 1994. "Fidelity and the Gendered Translation". 777? — Traduction,
Terminologie, Rédaction — Etudes sur le texte et ses transformation, VII (2), 147-163.
Arrojo, Rosemary. 1995. "Feminist, 'Orgasmic' Theories of Translation and Their
Contradictions". Tradterm 2, 67- 75.
Barthes, Roland. 1977. Image. Music. Text. Trans. Stephen Heath. New York: Hill and Wang.
Chamberlain, Lori. 1988. "Gender and the Metaphorics of Translation". Signs — Journal of
Women in Culture and Society 13(3), 454-472.
Derrida, Jacques. 1985. "Des Tours de Babel". Trans. Joseph F. Graham. In: J. F. Graham
(ed,), 165-207.
Derrida, Jacques. 1987. Positions. Trans. Alan Bass. London: The Athlone Press.
2
Unlike Myriam Díaz-Diocaretz, whose conception of "translator function" (1985:24-41)
is vaguely reminiscent of Foucault's, I explicitly relate the notion of "translator function"
which I propose here to Foucault's seminal essay "What's an Author?" (1979:141-159).
Díaz-Diocaretz. 1985. Translating Poetic Discourse: Questions on Feminist Strategies in

Adrienne Rich. Amsterdam/Philadelphia: John Benjamins Publishing Company.
Foucault, Michel. 1979. "What's an Author?". In: J. Harari (ed.), 141-159.
Graham, Joseph F. (ed.). 1985. Difference in Translation. Ithaca: Cornell University Press.
Harari, Josué (ed.) 1979. Textual Strategies — Perspectives in Post-Structuralist Criticism.
Ithaca, Cornell University Press.
Levine, Suzanne Jill. 1991. The Subversive Scribe — Translating Latin American Fiction. Saint
Paul: Graywolf Press.
Lewis, Philip. 1985. "The Measure of Translation Effects". In J. F. Graham (ed.), 31-62.
Ross, S. David. 1990. "Translation as Transgression". Translation Perspectives V, 25-42.
Steiner, George. 1975. After Babel — Aspects of Language and Translation. New York:
Oxford University Press.
Venuti, Lawrence (ed.) 1992. Rethinking Translation — Discourse, Subjectivity, Ideology.
London/New York: Routledge.
Venuti, Lawrence. 1995. The Translator's Invisibility — A History of Translation. London and
New York: Routledge.
von Flotow, Luise. 1991. "Feminist Translation: Contexts, Practices and Theories". TTR:
Études sur le texte et ses transformations IV (2), 69-84.
Willis, Sharon. 1992. "Mistranslation, Missed Translation: Hélène Cixous' Vivre L'Orange"
In: Venuti (ed.), 106-119.
Pour une sociologie de la traduction: le cas de la
littérature américaine traduite en France après la
Seconde Guerre mondiale (1945-1960)1
Jean-Marc Gouanvic
Dans l'état actuel des études traductologiques, on est frappé par l'absence de
certaines approches. Je voudrais esquisser ce que pourrait être une sociologie
de la traduction, en m'inspirant des théories de Pierre Bourdieu.2 Pourquoi
Bourdieu? Il est l'un des rares sociologues dont la théorie rende compte des.
productions symboliques (arts et lettres) sans les réduire à de simples biens de
consommation et sans succomber au finalisme ou au fonctionnalisme
mécaniste.3 L'un des traits de la sociologie de Bourdieu est de rétablir celles
et ceux qui sont actifs dans les pratiques culturelles comme les agents de ces
pratiques. Pour la traduction, cela est important, car on a tendance dans les
recherches en traduction soit à surévaluer le rôle du traducteur (en en faisant
l'égal de l'auteur, sans voir que la notion d'auteur est elle aussi
problématique), soit à le sous-estimer (en faisant de lui une simple courroie de
transmission du système). La notion d'agent, telle que Bourdieu l'entend,
permet d'articuler le rôle du traducteur avec le possible social, en proposant
la notion de champ comme modèle heuristique permettant d'analyser ce qui est
au principe des productions symboliques et de rendre compte de la part de
1
Communication réalisée dans le cadre d'une recherche subventionnée en 1995-1996 par
le Conseil de Recherche en Sciences Humaines du Canada.
2
Dans ce premier essai de sociologie bourdieusienne de la traduction, il a paru approprié
de paraphraser les formulations de P. Bourdieu, en les inscrivant dans le contexte de la
traduction.
3
Bourdieu écrit par exemple: Le "moteur" de Taction n'est "ni dans la fin matérielle ou
symbolique de l'action, comme le veut le finalisme naïf, ni dans les contraintes du
champ, comme le veut la vision mécaniste. Il est dans la relation entre l'habitus et le
champ qui fait que l'habitus contribue à déterminer ce qui le détermine. " (1982:48) Nous
reviendrons sur cette relation entre l'habitus et le champ.
créativité sociale qui émerge de la rencontre entre l'habitus de l'agent
traducteur et le possible du champ.
Les champs se présentent à l'appréhension synchronique comme des espaces
structurés de positions (ou de postes) dont les propriétés dépendent de leur
position dans ces espaces et qui peuvent être analysées indépendamment des
caractéristiques de leurs occupants (en partie déterminées par elles). [...] Un
champ [...] se définit entre autres choses en définissant des enjeux et des intérêts
spécifiques, qui sont irréductibles aux enjeux et aux intérêts propres à d'autres
champs (on ne pourra pas faire courir un philosophe avec des enjeux de
géographe) et qui ne sont pas perçus de quelqu'un qui n'a pas été construit pour
entrer dans ce champ [...]. (1984:114)
Qui sont les agents de la traduction? Ce sont bien sûr les traducteurs, mais
aussi les éditeurs, directeurs littéraires et directeurs de collections qui occupent
une position dans un champ de production culturelle (artistique, littéraire,
scientifique...) et il est d'une importance capitale de ne pas établir des cloisons
artificielles entre ces fonctions si l'on veut rendre compte de la traduction dans
un secteur d'activité donné.
Mon propos est d'analyser comment s'articule la théorie sociale bourdieu-
sienne sur la traduction, en étudiant le cas de la littérature américaine traduite
dans la France des années 1945-1960. Il s'agit de dessiner à grands traits deux
états de la recherche que j'ai entreprise: un panorama socio-historique du
champ littéraire français entre 1945 et 1960 du point de vue de la traduction
et une socioanalyse comparative des textes source et cible dans un champ en
émergence à la même période, celui de la science-fiction.
Sociologie des productions symboliques et "effet traduction"
Dans un espace culturel, tout se passe comme si s'opérait une division du

travail de production selon les intérêts des agents dans les positions spécifiques
qu'ils occupent dans leur lutte pour imposer une certaine hiérarchie de
légitimités sociales. Comment s'articulent les intérêts des agents dans l'espace
culturel? Une œuvre culturelle (roman, tableau, écrit scientifique, etc.) n'existe
non pas seulement pour autant qu'une institution (maison d'édition, galerie
d'art, revue scientifique...) l'accepte comme légitime et l'introduit dans le
champ littéraire, artistique ou scientifique; elle existe d'abord dans la mesure
où l'auteur a investi ou intériorisé les règles du jeu en vigueur dans le champ
en cause. Pour Bourdieu,
Jean-Marc Gouanvic 35
[L'] investissement (aux sens de l'économie et de la psychanalyse) c'est

l'inclination à agir qui s'engendre dans la relation entre un espace de jeu
proposant certains enjeux (ce que j'appelle un champ) et un système de dis-
positions ajusté à ce jeu (ce que j'appelle un habitus), sens du jeu et des enjeux
qui implique à la fois l'inclination et l'aptitude à jouer le jeu, à prendre intérêt
au jeu, à se prendre au jeu [ce que Bourdieu nomme l'illusio]. (1984:34-35)
Si donc les produits d'un espace culturel sont déterminés dans leur mode de
production par la logique propre du champ où ils s'insèrent et qui conditionne
leur existence et leur réception, alors la question se pose de savoir comment
se positionnent les produits d'origine étrangère mis sur le marché par les
éditeurs de l'espace culturel cible. Il ne suffit pas de dire que la traduction abat
les barrières nationales, fait circuler les idées par delà les langues, est un
instrument de communication interculturelle, met le propre en présence de
l'étranger... Les textes traduits sont nécessairement destinés à s'insérer et à se
positionner dans un champ spécifique de l'espace culturel cible, à se soumettre
à la logique objective et à jouer le jeu propre au champ où ils s'insèrent. Toute
publication nouvelle, qu'elle soit d'origine indigène ou étrangère, est
susceptible de faire bouger les hiérarchies de légitimités dans le champ en
cause (ici littéraire) en faisant glisser au passé les auteurs immédiatement
contemporains publiés dans le champ. Mais il y a un effet traduction tout à
fait particulier du fait que les auteurs traduits ont d'abord existé dans un champ
littéraire étranger dont les particularités et les enjeux peuvent être et sont
effectivement souvent sans grand rapport avec les traits et les enjeux du champ-
cible (de traduction). La décision de publier est aussi une décision de traduire
prise en fonction de la légitimité qui est celle de l'auteur et/ou de l'œuvre dans
le champ littéraire source et qui est évaluable de diverses façons, notamment
en termes d'écho critique, de prix littéraire, de consécration et en termes de
chiffres de vente (best-sellers). Les œuvres traduites jouissent d'un capital de
légitimité supérieur du fait qu'elles sont jugées dignes d'être diffusées dans un
espace culturel étranger. Mais ce capital attaché à une œuvre ou un auteur
repose aussi largement sur le capital de légitimité dont l'espace culturel cible
investit l'espace culturel étranger: ce qui est traduit, c'est telle œuvre de tel
auteur, mais surtout (par exemple dans le cas du champ littéraire français de
la fin des années 1940) telle œuvre de tel auteur américain, et c'est l'origine
américaine qui est ici le trait déterminant. Cette forte légitimité des œuvres
américaines traduites est-elle de quelque effet sur le champ littéraire français?
On verra que le champ de la littérature réaliste en France repousse la
traduction des textes dans ses marges - quel que soit le succès rencontré par
les œuvres traduites -, évitant ainsi des chambardements du champ cible. La
littérature traduite, même la plus légitimée, n'est pas traitée comme la

littérature indigène du simple fait qu'elle est publiée dans des collections
spécialisées réservées aux littéraires étrangères. De même, la traduction de la
science-fiction — genre littéraire non canonique — n'est prise en charge par
Gallimard et Hachette qu'à la condition que ce genre soit maintenu à une
distance respectueuse des œuvres de la littérature dominante traduite et non
traduite. On assiste alors à l'émergence ex nihilo d'un champ de SF spécifique
par la translation des structures éditoriales américaines pour accueillir les
œuvres traduites: c'est la création de revues et de collections spécialisées sur
le modèle de la science-fiction américaine. Cette translation est possible sous
la forme d'un compromis entre les structures sources américaines et une partie
de la tradition française dans le genre. C'est ainsi que s'est constitué le champ
français de la science-fiction à partir de 1951. Nul ne sera dès lors légitimé à
se réclamer de la science-fiction s'il ne joue pas le jeu en vigueur dans le
champ nouvellement constitué; tout agent potentiel doit donc lutter pour
prendre position dans le champ, ce qui établit par le fait même le champ
comme structure seule légitimée à dire ce qui est science-fiction et ce qui n'en
est pas, ce qui peut être traduit et ce qui ne peut pas l'être.
La littérature réaliste traduite assume une position différente dans l'espace
littéraire. Globalement doté de la légitimité institutionnelle optimale, le
discours fictionnel réaliste joue sur d'autres distinctions que celles qui
fonctionnent pour la science-fiction. Dans ce dernier genre, l'ostracisme est
principalement lié au mixte science\littérature traité sur le mode spéculatif,
inadmissible aux yeux des tenants de la littérature du circuit lettré. Dans le
champ de la science-fiction, le fait qu'un texte soit traduit de l'anglo-américain
est une plus-value en termes de légitimité. Dans le champ de la littérature
canonique, par contre, qu'il s'agisse d'une traduction constituera plutôt une
marque d'infériorité dans l'échelle des légitimités.
Les œuvres traduites jouissent d'un statut un peu particulier dans cette
dynamique. Il est entendu que, si la consécration dans l'espace social source
a fait exister l'œuvre source comme œuvre digne d'être traduite, les agents
cibles cherchent à reproduire cette logique des "mêmes causes\mêmes effets",
c'est-à-dire "mêmes succès" dans l'espace social cible. Auquel cas, les agents
joueront sur la proclamation ostentatoire de l'origine du texte, surtout si la
culture source bénéficie dans la culture cible d'un capital symbolique élevé (cas
dans la littérature américaine dans la culture française).
Dès la décision de traduire, bien en amont du texte, des modèles sont donc
en place dans le champ littéraire et vont déterminer les manières de traduire.
En apparence, pourtant, le traducteur en tant qu'agent semble jouir d'une
certaine marge de manœuvre en fonction de sa légitimité sociale.

Concrètement, il peut avoir la confiance de l'éditeur ou du directeur littéraire
de sorte qu'il pourra être le seul décideur. En tant que spécialistes des langues
et cultures sources, les traducteurs ont souvent en effet des pouvoirs de
décision assez importants. Je pense par exemple à Raymond Queneau qui
traduit la littérature américaine dès 1934, fonde la revue Volontés avec Henry
Miller en 1937, entre au comité de lecture de Gallimard en 1938, traduit neuf
textes dans le numéro spécial de Mesures sur la littérature américaine, implante
la science-fiction en France à partir de 1950 avec Boris Vian en y consacrant
les premiers articles critiques. Le rôle de Queneau (et aussi de Vian traduisant
dans les années 50 la science-fiction de Bradbury, Leinster, Robinson, Padgett,
Van Vogt, etc.) dans l'espace culturel français des années 1935 à 1950 ne peut
être justement saisi que si l'on ouvre la réflexion à l'ensemble des activités
auxquelles le traducteur s'adonne, seule condition d'une évaluation juste de la
position de l'agent et de sa légitimité.4
Cela dit, cette analyse du rôle d'un traducteur ne peut se contenter de faire
l'apologie d'une personne, aussi influente soit-elle. Nous n'avons encore fait
qu'effleurer le sujet en évoquant le rôle créatif, novateur, de Queneau. La
question suivante est plus importante: dans quelle dynamique sociale s'inscrit-il
dans le champ, autrement dit comment négocie-t-il sa position par rapport aux
autres agents avec lesquels il est en concurrence pour la légitimité? Lorsque
Queneau traduit Edgar Wallace en 1934 ou lorsque Vian traduit les auteurs de
SF en 1950, l'un des faits clé de l'histoire littéraire et de l'histoire des idées,
c'est la non-existence de champs spécifiques dans les configurations de textes
traduits. Ce fait est fondamental, car les traductions de ces textes seront fort
différentes selon qu'elles prennent place ou non dans un champ subculturel
spécifique. Voyons donc maintenant comment s'organisent le champ de la
science-fiction et celui de la littérature réaliste.
4
Insistons encore une fois sur la nécessité d'une historiographie culturelle des traducteurs
et de la traduction, une historiographie qui tienne compte de toutes les déterminations,
y compris métatraductives (et par là il faut entendre les activités dans lesquelles n'entre
pas directement la traduction, comme la participation à un débat critique ou à une
polémique, par exemple).
La traduction de la science-fiction américaine et l'émergence d'un champ

français de SF dans les années 1950
Avant 1951 en France les romans scientifiques indigènes et a fortiori traduits

n'ont d'autres points de chute éditoriaux que les magazines de récréation et de
vulgarisation comme Je sais tout ou Sciences et voyages, les collections de
littérature populaire de Ferenczi, Tallandier, etc. Ces romans scientifiques sont
disséminés dans des lieux éditoriaux si divers qu'ils ne semblent pas appartenir
à la même configuration générique, en dépit ou à cause de Jules Verne qui
n'existe que repoussé dans les marges de la littérature pour jeunes. Or, et ici
nous entrons dans un paradoxe fréquent en traduction, si la "science-fiction"
s'impose à la fin des années 1920 dans le magazine américain Amazing Stories
(fondé en 1926 par un Américain d'origine luxembourgeoise, Hugo
Gernsback), c'est en tant qu'héritière de Jules Verne traduit et publié dans
Amazing. Verne est le modèle de la SF américaine, celle-là même qui
s'imposera en France dans les années 50 comme un "genre nouveau". A ce
premier paradoxe s'en ajoute un second: l'une des "sources" de Jules Verne
est l'Américain Edgar Poe, mais bien sûr traduit par Baudelaire! On voit que
dans et par la traduction s'opèrent des transformations, des manipulations dont
les enjeux sont très différenciés, pouvant faire advenir du nouveau dans les
cultures traversées.
A partir d'Amazing Stories, un champ culturel spécifique se constitue: un
public se forme (un jandom), des auteurs se spécialisent en SF, publient dans
des collections et des revues spécialisées comme Astounding Stories (apparue
en réaction à Amazing Stories), des prix, des congrès et des clubs de mordus
sont créés. C'est Astounding que Queneau, Vian et Pilotin liront à la fin des
années 40 et au début des année 50, y puisant les modèles de la SF qu'ils
apprécient et qui les autorisera à présenter la SF comme un "genre nouveau".
Voyons comment se constitue le champ de la science-fiction dans l'espace
culturel français et comment la structure du champ conditionne le traduit. Les
éditions Hachette, auxquelles se joignent les éditions Gallimard, sont les
premières à prendre position dans le champ en formation en 1951 en fondant
une collection spécialisée, "Le Rayon Fantastique". Cette collection ne publie
que des traductions pendant des années. C'est la collection où l'on retrouve le
plus la marque de Vian (il a traduit deux romans de Van Vogt) et de
Queneau.5 Les éditions Fleuve Noir créent aussi une série, "Anticipation", à
5
A côté de Georges-Hilaire Gallet qui a été, avant-guerre, Tun des tout premiers
connaisseurs de la SF américaine, avec Régis Messac. Sur les modèles socio-esthétiques
cheval sur l'héritage des romans populaires d'avant-guerre et le modèle de la

Série noire (Gallimard). Elle publiera finalement peu de traductions et encore
sous forme très "adaptée" en 180 pages aérées pour deux heures d'évasion.
Puis ce sont les éditions Denoël avec "Présence du futur" qui, à l'époque,
publie des romans plus "littéraires", moins populaires ou juvéniles, à côté de
récits fantastiques classiques d'auteurs de la littérature réaliste (Hougron,
Curtis, etc.) et de la SF américaine plus fantaisiste ou poétique (paradigme:
Bradbury). Et puis il y a les revues Fiction, l'antenne de The Magazine of
Fantasy and Science Fiction américain et Galaxie où pour ainsi dire tout est
traduit du Galaxy américain.
Le champ est structuré autour d'une part la revue Fiction qui rapidement
détient une position hégémonique et, d'autre part, les éditeurs Hachette-
Gallimard et Denoël. Une lutte de pouvoir s'instaure entre ces deux derniers,
qui tendent à offrir de la science-fiction une image fort différente; les récits
publiés chez Denoël sont propres à plaire à un lectorat sensible aux thèmes
fantastiques hérités de Nodier, Mérimée, Maupassant, où la science est peu
présente; les récits publiés chez Hachette-Gallimard sont exclusivement
scientifiques. Le modèle de Hachette-Gallimard est nettement celui qui s'est
imposé dans le champ de la science-fiction française, même si la collection "Le
Rayon Fantastique" devait cesser de paraître au début des années 60 et ce pour
des raisons encore obscures. Le relais a été pris par Denoël et par les éd. J'ai
lu, qui d'ailleurs ont réédité les titres d'Hachette-Gallimard, les faisant ainsi
accéder au statut de classiques. Comment ces traits socio-esthétiques ont-ils été
négociés dans la traduction des textes? Comment se sémiotisent la position des
agents dans les textes? Une analyse comparative des traductions du Rayon
Fantastique (Hachette-Gallimard) avec les textes sources montre que les
éditeurs et les traducteurs ne manipulent pas sciemment les textes (Gouanvic
1995). Je n'ai constaté aucun cas de censure, malgré l'image peu flatteuse que
les Français ont parfois dans certains récits. Le traducteur ne procède que très
rarement à des adaptations; manifestement son mandat est de traduire, non
d'adapter. Ce qui ne va pas sans quelques difficultés, en particulier dans les
récits catastrophiques et de critique sociale, où le lecteur français se trouve
extérieur au drame vécu par les personnages américains et les institutions
américaines sur le territoire des États-Unis. En fait, cela est l'un des traits
majeurs de 1'"effet traduction" et que l'on pourrait nommer "décentrement",
en empruntant à Henri Meschonnic. La particularité de ces traductions est que
de la science-fiction française et sur la mutation générique subie dans les années 1950,
voir Gouanvic (1994b).
le décentrement traductif est pleinement assumé comme positif du fait que la
traduction de la SF est dans la logique de la reconnaissance d'un genre
spécifiquement américain. Il n'est pas question d'adapter le texte américain,
de l'assimiler à la culture française, mais bien d'en proposer une traduction
dissimilatrice, qui correspond à la demande du public et jouit d'une légitimité
sociale dans le champ en formation. On est loin de l'image très largement
répandue, selon laquelle les genres "paralittéraires" seraient systématiquement
traduits "n'importe comment" ou sur le mode de l'infidélité assimilatrice.
La traduction de la SF est prise au sérieux par ses traducteurs. Mais qui
sont-ils? Comme le champ de la SF n'existe pas encore au début des années
50, ce sont des traducteurs non spécialisés en SF : Jean Rosenthal, Jacques
Papy, Marie-France Watkins, Amélie Audiberti, Boris Vian... Les seuls qui
tendront à prendre place exclusivement dans le champ de la science-fiction à
titre d'agents de traduction, mais pas seulement de traduction, sont peu
nombreux: Georges-Hilaire Gallet, Alain Dorémieux. Il demeure que, à
mesure que le champ de la science-fiction s'autonomise et se spécifie, il
devient de plus en plus évident que le traducteur doit connaître de l'intérieur
le discours subculturel de la science-fiction, avec ses sociolectes, ses mots-
fiction qui entrent en résonnance de textes en textes et que les lecteurs
reconnaissent immédiatement (par exemple une allusion à l'une des Lois de la
robotique d'Asimov qui apparaissent dans ƒ, Robot). Tel est le "jeu" en vigueur
dans le champ nouvellement constitué de la science-fiction. Qu'en est-il des
œuvres réalistes américaines traduites dans le champ littéraire canonique?
La littérature canonique américaine dans l'espace culturel français après

1945
On pourrait penser que la dominance américaine en science-fiction serait dans

la logique de l'hégémonie des États-Unis en matière de sciences et de
technologies et que l'espace culturel français des années 1945-1960 ne
connaîtrait d'autre vague de fond de traduction de la littérature américaine. On
va voir qu'il n'en est rien, en esquissant un panorama de la traduction de la
littérature américaine canonique à partir de 1945. L'analyse fait apparaître
d'emblée un phénomène peu banal. Les œuvres américaines traduites après la
Seconde Guerre mondiale (1945-1960) sont de deux types: ce sont celles qui,
datant du 19e s. et du début du 20e, font l'objet d'un rattrapage de traduction
et celles qui sont d'auteurs contemporains.
Hawthorne (1804-1864): L'auteur de The House of the Seven Gables et
de The Scarlet Letter est très sélectivement traduit dans la culture française
entre 1850 et 1950. Certes The Scarlet Letter connaît de nombreuses versions
françaises dès sa publication en 1850. Mais il faut attendre 1952 pour que The
Blithedale Romance (paru cent ans plus tôt) soit mis à la disposition des
lecteurs français par Gallimard accompagné d'une préface d'André Maurois.
Quant à The Marble Faun qui date de 1860, il n'est publié dans une version
française complète qu'en 1949 et toujours par le même Gallimard. La
publication de The Marble Faun est accompagnée d'une préface de René Lalou
intitulée — de façon très significative — "Nathaniel Hawthorne, créateur et
précurseur". Historiquement, c'est Hachette qui se positionne comme l'éditeur
de Hawthorne en français dans la foulée des premières éditions américaines.
Cependant, Hachette ne publie qu'un nombre restreint de ses œuvres. Sauf The
Scarlet Letter, qui sera rééditée régulièrement, seuls les contes pour enfants de
Hawthorne (comme A Wonder-book for girls and boys de 1851 et The
Tanglewood Tales de 1853) sont traduits et publiés avec succès dans le champ
français de la littérature pour les enfants. En France, Hawthorne est ainsi
presque exclusivement un écrivain pour enfants jusqu'après la Seconde Guerre
mondiale. En 1945, il sort de la marginalisation, grâce au rattrapage intense
dont il est l'objet grâce à Gallimard, qui entreprend de faire traduire les
romans "oubliés" (The Blithedale Romance et The Marble Faun, en
particulier). Cette découverte de Hawthorne dans la France de l'immédiat
après-guerre est-elle propre à Hawthorne, ou faut-il rattacher les efforts de
rattrapage à un mouvement plus général? Les cas de Henry James et de
Fitzgerald sont instructifs à cet égard.
Henry James (1843-1916): Avec James, on constate une situation encore
plus criante. L'effet Hawthorne, si l'on peut dire, ne lui est pas particulier:
Les éditeurs français font l'impasse sur certaines œuvres comme The Turn of
the Screw pendant 30 ans et sur The Europeans, Washington Square, The
Bostonians, The Awkward Age et What Maisie knew, que les lecteurs français
ne découvriront qu'après la Seconde Guerre mondiale. Fitzgerald (1896-
1940): Pour ce qui est de Scott Fitzgerald, il y a de quoi être étonné de
l'intérêt très mitigé des éditeurs français pour son œuvre. Seul The Great
Gatsby (de 1925) est publié en traduction française immédiatement, en 1926.
Tender is the Night (1934), The Last Tycoon (1941) et la célèbre nouvelle "The
Diamond as big as the Ritz" (1922) ne connaissent des versions françaises que
dans les années 50. This Side of Paradise (1920), The Beautiful and the
Damned (1922), Tales of the Jazz Age (1922), All the Sad Young Men (1926)
et Taps at Reveille (1935) ne susciteront l'intérêt des éditeurs français qu'après
les années 60.
Parmi les auteurs américains régulièrement traduits en français, Faulkner,
Hemingway, Dos Passos (pour ne citer que ces trois cas) sont l'objet d'un
intérêt tout particulier. Faulkner a été traduit par Maurice-Edgar Coindreau,
l'une des figures dominantes de la traduction avant et après la guerre, mais
aussi par R.-N. Raimbault (avec C. P. Vorce), Hilleret, André du Bouchet et
Maxim Gauchet. A peu près tout l'œuvre romanesque est ainsi traduit et publié
chez Gallimard à 90 %, sauf des textes plus anciens comme Mosquitoes (de
1927), traduit en 1948 aux éditions de Minuit, ou encore Soldiers' Pay (de
1926), traduit par l'éditeur Arthaud en 1946.
Hemingway bénéficie aussi de la faveur de l'édition française, puisque dès
1928 la NRF publie 50 000 dollars (titre d'un recueil de trois récits
comprenant le titre éponyme), suivi en 1931 par l'Adieu aux armes, traduit par
Coindreau (encore lui) chez Gallimard, le Soleil se lève aussi en 1933,
également traduit par Coindreau pour Gallimard. Après guerre, Marcel
Duhamel sera le principal traducteur d'Hemingway, avec l'exception notable
du Vieil homme et la mer traduit par Jean Dutourd en 1952 pour Gallimard.
Troisième, exemple remarquable, John Dos Passos: l'omniprésent
Coindreau s'est chargé de la traduction de Manhattan Transfer (publié par
Gallimard en 1928), laquelle traduction sera rééditée de nombreuses fois avant
et après la guerre. Cependant, c'est là le seul Dos Passos traduit par
Coindreau. Dos Passos (13 romans entre 1920 et 1954) a été traduit par
d'assez nombreux traducteurs, N. Guterman, Charles de Richter, Yves
Malartic, Jean Rosenthal, Hélène Claireau, Maurice Rémon, Jean Collignon,
R.-N. Raimbault, Jean Castet, et j'en passe, la plupart pour Gallimard.
Ce tableau général permet de tracer les grandes lignes des impasses que les
éditeurs français ont fait jusqu'à la Seconde Guerre mondiale sur certains
auteurs ou sur certains textes de la culture américaine source; à l'inverse, il
permet de mettre en lumière l'intérêt considérable que suscite la littérature
américaine dans la France de 1945-1950. L'éditeur dominant est clairement
Gallimard qui, avec sa collection "Du monde entier", exerce un quasi-
monopole. Les autres éditeurs qui créent des collections spécialisées en
traduction sont: Hachette avec la coll. "Grands Romans étrangers"; Calmann-
Lévy avec la coll. "Traduit de..."; Ferenczi avec "Les Romans américains".
Et Gallimard ne se cantonne pas à la littérature du circuit lettré: il prend pied
très tôt en Série noire (en 1944) et en science-fiction (en 1951) comme on l'a
vu.
L'effet transformateur de la traduction apparaît donc on ne peut plus
évident du fait qu'il s'inscrit non pas seulement dans la "distance
géopolitique", mais dans la "distance historique". Cette "distance géo-

historique" est-elle d'un effet déterminant sur la structure du champ littéraire?
En paraphrasant Bourdieu, on pourrait dire que, quels que soient les dates de
publication des œuvres dans l'espace littéraire source et donc les délais de
traduction, la lutte pour le pouvoir symbolique dans le champ de la littérature
ne reconnaît pas dans le champ culturel cible des auteurs d'apparition récente
ou ancienne, des auteurs traduits (classiques ou modernes dans la culture
source): tous les auteurs sont contemporains dans le champ littéraire cible du
fait que leur existence dans le champ les reconnaît comme un enjeu de
pouvoir.6
Au vu de ce panorama, aussi schématique soit-il, de la traduction de la
littérature américaine dans la culture française après la Seconde Guerre
mondiale, il apparaît que les auteurs américains — toutes catégories de textes
confondues — font l'objet d'un immense intérêt dans le champ littéraire
français. On peut tenir Hawthorne pour représentatif de l'effet traduction de
la littérature américaine canonique en France après 1945. C'est d'une seconde
vie que jouit Hawthorne avec les traductions et retraductions auxquelles André
Maurois, François Mauriac et Julien Green vont consacrer préfaces,
introduction et appendice. Au point où une préface, celle de René Lalou à la
traduction de The Marble Faun où Hawthorne est investi de la qualité du
précurseur, fait figure de modèle discursif typique de la traduction de la
littérature américaine dans la culture française. Elle n'est pas tant justifiée par
la logique de la découverte que par celle du raccrochage du présent de la
traduction au passé de l'original. Ce raccrochage est bien évidemment un coup
de force par lequel un éditeur tente d'accréditer la thèse de la modernité de la
culture américaine, y compris celle que représente Hawthorne.
L'enseignement le plus significatif de cette analyse sociologique de la
traduction est qu'il convient de rétablir la source dans les études traducto-
logiques. Qu'il y ait transformation-manipulation du texte traduit selon les
intérêts de la culture-cible (ou mieux du champ-cible), cela est indéniable. Il
reste que le texte manipulé l'est en tant que texte américain. La "revisitation
de la littérature américaine" par les agents de la traduction s'opère avec l'idée
fortement ancrée, et intériorisée, que la société américaine est le modèle à
suivre, que l'avenir de la société française est inscrit dans la société américaine
— et cette remarque vaut pour des groupes socio-culturels fortement
différenciés: les éditeurs catholiques traduisent le puritain Hawthorne à leur
profit alors que Sinclair Lewis, l'écrivain de gauche et prix Nobel, est traduit
6
Voir Bourdieu (1992:224) sur la "synchronisation des temps discordants".
par Gallimard. On peut sans exagérer avancer que les traducteurs se font les
agents d'une certaine américanisation de la littérature française (Gouanvic
1994a); et je dis "certaine" américanisation, pour deux raisons: (1) parce que,
bien évidemment, (mais on a tendance à l'oublier) ce ne sont pas les États-Unis
dans leur totalité culturelle qui prennent la parole à travers les œuvres, mais
— vaille que vaille — la classe moyenne instruite (educated middle class),
même en science-fiction; (2) parce que l'américanisation dont il s'agit est une
américanisation à la française de la société française: en d'autres termes, c'est
la façon française de "voir" les États-Unis selon certains intérêts français. Tel
est l'enseignement général que, à ce stade, je tirerais d'une sociologie de la
traduction informée par les idées de Pierre Bourdieu.7
Références
Bourdieu, Pierre. 1971. "Le marché des biens symboliques". L'Année sociologique 22, 49-126.
Bourdieu, Pierre. 1979. La Distinction — Critique sociale du jugement (Le sens commun).
Paris: Minuit.
Bourdieu, Pierre. 1982. Leçon sur la leçon. Paris: Minuit.
Bourdieu, Pierre. 1984. Questions de sociologie. Paris: Minuit.
Bourdieu, Pierre. 1987. Choses dites (Le sens commun). Paris: Minuit.
Bourdieu, Pierre. 1992. Les Règles de Vart. Genèse et structure du champ littéraire (Libre
examen). Paris: Seuil.
Gouanvic, Jean-Marc. 1994a. "La traduction et le devenir social: le cas de l'irruption de la
science-fiction américaine en France après la Seconde Guerre mondiale". TTR — Etudes
sur le texte et ses transformations 7(1), 117-152.
Gouanvic, Jean-Marc. 1994b. La Science-fiction française au XXe siècle (1900-1968): essai de
socio-poétique d'un genre en émergence. Amsterdam-Atlanta: Éditions Rodopi.
Gouanvic, Jean-Marc. 1995. "Sémiotique et traduction: les enjeux sociaux de la traduction de
la science-fiction américaine au Rayon Fantastique". Francophonie plurielle. Montréal-
Casablanca: Hurtubise HMH-Eddif, 199-214.
7
Le propos de cette étude n'a pas été de distinguer la sociologie bourdieusienne d'autres
théories présentant certains points d'intersection avec elle, comme la théorie du
Polysystème, mais de la développer selon ses exigences propres. S'il fallait cependant
esquisser une différenciation, il me semble que l'apport fondamental de la théorie de
Bourdieu est de construire une théorie sociale générale des productions culturelles, en co-
pensant le travail de production des œuvres culturelles selon leur logique propre dans
leur champ spécifique (littéraire, etc.) et l'usage social de ces œuvres dans l'espace social
général et les champs du pouvoir. La théorie du polysystème interprète les systèmes
culturels comme des systèmes sémiotiques et soumet leurs productions à une analyse
descriptive interne, alors que la socio-analyse de Pierre Bourdieu ne traite jamais le
sémiotique en dehors du social, le sémiotique n'étant pas — pour Bourdieu —
interprétable en dehors du social.
Translation as imposition vs. translation as requisition
Cay Dollerup
In this article I shall discuss societal forces which propel texts in source
languages to become translated. My examples will be Danish, but they are
discussed for their paradigmatic and international, rather than their specific and
national value.
I hope these observations will contribute to make endeavours in translation
work for cultural transfers more successful in today's world where numerous
new nations are establishing their identity, in terms of past and present,
industry, trade and culture. In the last field, the most significant element, in
the eyes of intellectuals, is that their national art and literature should be
recognised, enabling them to aspire to the coveted international symbol that
this goal has been reached — the Nobel Prize.1
One persistent feature in discussions of translation practice and theory is
the acceptance of the movement from left to right, that is from a sender who
utters the source message which moves through various stages, to the end as
a target text with an audience; it is a view abundantly reflected in most models
of the translation process.2
I intend to question the automatic applicability of this left-to-right model
in actual translational activity in societal contexts from a historical, diachronic
perspective with particular reference to the present-day scene.
1
My attention was drawn to these features in connection with some research (e.g. Dollerup
1995) as well as to the frequent patriotic query which I have met with: "Is it only because
of poor translations that our most prominent poet/author has not received the Nobel
Prize ".
2
See e.g. Wilss (1982:57, 81), Levy (1969:33). Nida implies this sequentially in his
diagram of the transfer or translation (1964:147). That the models are not exclusively
European is demonstrated in Mohanty (1994:194); and in Uwajeh (1994:247).
Senders and translation
In today's world there may be texts which are produced with a structure, style
and vocabulary which will facilitate translation.3 I have never met with such
texts, but they might conceivably be found among the scripts of television
serials intended for international consumption, among addresses to international
audiences, and, perhaps the most likely case, among delegates' speeches at
international meetings with conference interpreting. At all events such texts
will make up only a fraction of the texts eventually transferred to other
cultures. In the vast majority of cases, today as well as in previous ages,
authors and original senders have not troubled to facilitate interlingual
mediation of their product in the moment of conception; in most cases they
have not taken subsequent translation into account at all.
In other words, no matter whether their product is an advertisement or a
literary masterpiece, authors normally produce with only an audience speaking
their own language in mind. Texts are formed for source language receivers,
with their implied background, ideas, notions, and frames of references. In
discussing translation in a societal and national context, the point of departure
should therefore be the simplest model of communication:
Sender (and sending culture) message recipient (and receptor culture)
This model implies that translation is not an integral part of your ordinary
source text. Translation is not part of the creation, the existence and the
primary reception. Translation is an outside force in relation to the message
incorporated in the source text. This approach allows for the legitimacy of the
question: "How does translation come about?"
'Imposition' vs. 'requisition' by means of 'cultural bridgeheads'
When translation is forced upon source texts, their realisations in target

cultures will vary from being 'imposed' by the source culture (in the broadest
sense of the term) to being 'requisitioned', that is wanted, desired, by target
cultures. Throughout history and depending on purpose and genre, there have
been fluctuations in these respects. 'Imposition' is normally deliberate; it is
always driven by the source culture, often with little regard for the receptor
3
Even if this should be the case, such facilitation can only be undertaken with language-
specific translation in mind: i. e. no source text can possibly take into account translation
problems and language pecularities which apply to all other languages in the world.
Cay Dollerup 47
culture, and therefore pays much attention to the intention or intentionalities

behind the original text manifestation; 'requisition' springs from the target
culture and therefore implies a more relaxed attitude (perhaps out of ignorance)
towards the sender's intentionality. The most obvious historical (and present)
examples of imposition are found in religious writings.
Similarly most political and technological texts are normally also translated
at the instigation of the sending cultures in order to be imposed on target
cultures. Nowadays the main areas of 'imposition' would seem to be
international relations, formerly often domination and imperialism; and
international trade, specifically sales of products. In these cases sending
languages have dominated, and generally speaking, 'initiators' and translators
tend to agree that there should be loyalty to the sender. Previously this fidelity
was taken to be realised in a literal translation. Even today, and especially until
some 20 years ago, one would meet with abysmal translations in advertise-
ments, recipes, manuals and the like which went along with foreign products.
Today, there is a much higher awareness among firms that they must bow to
the language and culture in foreign markets if they want to sell their product.
Permit me to exemplify: in 1975, I collected international sales material from
Danish firms, including the world's leading manufacturer of diesel engines for
ships. I confidentially received what the firm clearly took to be a fantastic
endeavour in advertisement, namely 'exactly the same text' in Spanish,
German, French, English and Danish. Since then the shipyard has gone
bankrupt — which was, I hasten to add, not the translators' fault. In 1994 I
repeated the operation with numerous other firms, and found that by now
nearly all target language brochures and manuals were adapted to national
purchasers to an extent which made it hard to discuss most target texts in the
traditional terms of translation studies. This was striking and it also went for
highly technical texts: even when illustrations were identical, the brochures and
specifications would foreground and expand features which were particularly
pertinent in the target language nations and suppress information which was
irrelevant. In a brochure describing thermostats and the like, the European
versions would thus refer to European Union standards, whereas the Russian
version referred to performance. The overall lesson is that both 'initiators' and
'translators' (who may well be (independent) teams of e.g. translators and
engineers) are aware that fidelity and loyalty to the sender are best served if
the target-language version deviates from the actual phrasing of the source text.
Scientific and educational material is translated mostly as requisition but
there is a difference in the fidelity towards the source text: scientific texts will
tend to be loyal and literal, whereas educational ones will allow for more
latitude and adaptation.
Adaptation will apply, in particular, to literary translation where successful
translation is characterised by an overall requisitioning attitude. With the
possible exception of educational material, literature differs from the other
types of texts in that, at least previously, translations did not come out of the
blue. All through the Middle Ages and the Renaissance, it was people with
some knowledge of the classical languages who would mediate the classics to
their contemporaries: prior to translation, these translators were 'cultural
bridgeheads' for classical lore in their own cultures who then requisitioned the
classics to dress them in new garbs in their own vernaculars.
As vernaculars began to establish their own literatures, the process became
more obvious: the sonnet made it into English because Thomas Wyatt was in
Italy (1527) and could translate Petrarch's work into English. Once back in
England, he (and Henry Surrey) paved the way for the genre's English form,
rather than for specific translations from Italian.4
Throughout subsequent centuries we meet with similar concrete ties
between historical facts and translation. Translation between the contemporary
languages is undertaken by amateurs with anything ranging from the most
superficial to the most thorough command of the source language. Molière's
comedies and the French neoclassical theatre, Corneille and Racine, were
introduced into England thanks to the 'cultural bridgehead' established among
theatre-goers in the British aristocracy and gentry during their exile in France
during Cromwell's Republic (1643-1660). In the next century, Shakespeare and
the English novel travelled in the opposite direction — to the Continent,
notably Germany thanks to the personal union (established in 1714) which tied
part of Germany to Great Britain.
4
In this context I shall leave out a detailed discussion of the implication that this mechan-
ism makes it ontologically impossible to talk about equivalence, no matter whether of
form, content, or effect, between source and target language reception: 'equivalence' is,
at best, a comparable entity between the response of a target language audience which
can read the original source text, and the response of the target language audience which
does not know the source language and therefore needs translation. This last incisive
observation is due to Jens Nørmark Lind and Peter Sestoft (essay, Spanish, University
of Copenhagen 1992).
Cay Dollerup 49
Large-scale translation: the 19th century
I have suggested that large-scale professional translation, in the sense that

many translators could actually turn it into a living, came into existence in the
last century as the outcome of the Napoleonic wars, subsequent nationalism,
improvements in national infrastructures and the increase in international trade
which brought home to the rising bourgeoisie that communication with other
nationalities would benefit trade and culture (Dollerup 1996). So they learnt
foreign languages, and, more often, consumed translations of foreign literature:
there was an ever-increasing need for education and for entertainment.
Moreover, literary translators were still normally paid only token sums for
their efforts, so they continued the time-honoured practice of translating as a
labour of love. To sending national literatures, however, 'cultural bridgeheads'
were a prerequisite for speedy translation. To take one example: the Tales of
the brothers Grimm were published in Germany in 1812. They were translated
into Danish as the first foreign language as early as in 1816. This was the
work of Adam Oehlenschläger, the leading Danish romantic poet who himself
published in German, as did other Danes at the time (e.g. Jens Baggesen). In
Denmark, there were German 'cultural bridgeheads' galore, since one third of
the realm (Slesvig-Holsten) was German-speaking. Conversely, many
contemporary Germans knew Danish and constituted Danish 'cultural
bridgeheads'. So it was not surprising that Hans Christian Andersen's first
serious writing from 1827 was translated as early as 1831, and his novel The
Improviser was published in German the same year it came out in Danish. By
1838 it was claimed that "Sein Name ist in Deutschland so bekannt wie in
Dänemark" (Quoted from Möller-Christensen. 1992:101).
In other cases there were no 'cultural bridgeheads' and no translators who
could lift literature directly out of the source language, even if there might be
pockets of 'cultural interest' generated by such stimuli as indirect news in
potential target cultures. This brings to the fore the function of some languages
as 'gateways' to other cultures. There is no doubt that the Grimm Tales were
destined for an international career when they were translated into English in
1823, in the same fashion that Hans Christian Andersen's fame in this genre
was assured with the 1839 German translation of his first fairytales. 'Gateway
languages' in Europe, especially German, English and to some extent French,
have always been central to translation, notably so for the propagation of the
literature and other messages to and from minor language communities. The
increase in the translational activity in the 19th century draws attention to
Germany as the 'cultural bridgehead' for British literature, Scott, Byron,
Shelley. As gateway language, German provided source texts for minor

European languages both in Scandinavia as well as in Central Europe.5
Conversely, then as now, it is English which is the 'gateway language' for
overseas success for much European literature. Although there tends to be a
correlation between being a 'gateway' and a 'dominant' language, even minor
language communities may sometimes function as gateway languages for
literature which was sufficient to draw the attention of 'cultural bridgeheads'
or of 'pockets' to their existence: Ibsen and Strindberg both used Danish as
their gateway language for getting known in the world at large, first of all in
German, which then functioned as the second gateway language for them.
This analytic overview permits us to draw the conclusion that until the
19th century, the requisitioning attitude to literature has mainly been motivated
by wishes to present the target language readerships with (a) the classics, (b)
literary innovations, and (c) entertainment (or educational material). Much
translation work was prompted by idealism in some form or other, be it
enthusiasm or religious zeal.
New developments
I shall cut short the chronological overview at this stage because in the late
19th century the mechanisms of translation by requisition were changing. It
was not only translators who were becoming professional. So were target
language publishers with commercial interests and distribution networks. They
began to pay attention to 'pockets of cultural interest' in national cultures that
would make translation from a gateway language profitable. Financially, they
only had to worry about the translator's fee, for there was no international
copyright law protecting authors. With 19th-century improvement in education
and the consequent creation of mass readerships, translation of literature —
and educational material — became a money-spinning industry, a money-
driven activity subject to market forces in capitalist societies. The scene was
set for extending the publishers' financial interest in promoting authors when
the international Berne convention (1886) protected the original authors'
copyright in translation.
5
For Scandinavia and Denmark in particular, I refer to the list in Nielsen (1966: 11-12;
and 1976, various places). For Central Europe to Hans Vermeer (personal
communication).
Cay Dollerup 51
The Communist translation policy in the Soviet Union, which began in full
in the 1920s, introduced yet another new feature in literary translation:
translation was not only requisitioned but also selective and ideologically
driven. The Danish literature that made it into Russian was social criticism,
exemplified by such writers as Martin Andersen Nexø (1869-1954). Since
Russian functioned as the gateway language for other Communist receptor
cultures, Nexø's name dominates in the perception of Danish literature in
China and, I would assume, other previously Communist countries as well.
The present scene
Jumping to the present Danish scene, there is still imposition. Mainly in terms
of export and import in trade; and toleration of poor quality translation in these
contexts is high. The reason is not hard to find, for as a consumer one is more
motivated to make sense of the opaque instructions than to throw out the newly
acquired dishwasher. Imposition is found in religious contexts within
denominations and sects. Educational texts seem to be just as much of a
mixture as previously. In medicine and the natural sciences, however, English
now functions as the lingua franca.
In the field of literature, there are still idealistic translators and publishers
who try to boost contemporary authors and thus function as bridgeheads for
foreign literature in Denmark, and, to a lesser extent, for Danish or
Scandinavian literature abroad, the latter especially with some minor specialist
publishers in the US. We also find a wish to present the public with Danish
and Nordic 'classics', best exemplified by one or two North American
academic presses which operate without subsidies from the source cultures.
Press-runs are clearly small and production is mostly for libraries, since the
prices are prohibitive for most individual purchasers. It is also interesting that
in both cases translations are, more than ever before, direct translations that
do not pass over gateway languages. This change indicates either that more
emphasis is paid to 'fidelity' or that the translators/publishers assume their
audience knows more of the source cultures than others (perhaps thanks to
previous (and less "loyal") translation by way of gateway languages).
The money-driven market forces are gaining ground, but even so there is
ambiguity in the attitude to translation.
Karen Blixen, better known by her pen-name Isaak Dinesen, bypassed
translation since she preferred to retell her linguistically complicated and
fascinating tales herself in a less convoluted English. In these retellings, she
not only omitted lengthy descriptions localising her stories in Denmark, but she
also changed numerous details (age and name of characters, dates etc.).6 The
procedure worked well, for she was an international success. Others who have
done their own translations or supervised them have fared less well.
Several Danish writers, such as Anders Bodelsen, have made it into
Dutch, German, and English in direct translation, and thanks to success in the
latter gateway language, into other languages such as Spanish and Italian.
Henrik Stangerup has been translated into French. Peter Høeg has been
successful in English and a host of other languages with his Miss Smilla 's
Feeling for Snow (1994; UK title). Without a major research effort (which is
frequently bound to be thwarted because of publishers' trade secrets), it is hard
to find out how these authors got to be known to the 'cultural bridgeheads'.
In some cases, they have clearly not been requisitioned in the same way as
Andersen or Grimm, but sold to (that is 'imposed' on) other countries by the
Danish publishers at international conventions, such as the Frankfurt Book
Fair. The large-scale institutionalised promotion by publishers of in-house
authors is fairly recent but is, as mentioned above, ultimately due to the Berne
Convention.
There is also a Danish national state-operated, money-driven translation
policy for literature which promotes Danish literature abroad by subsidies. The
sum is insignificant (c. 200,000 dollars worldwide per year) and it may be
used only for publications translated directly from Danish. Given these
limitations, its importance is small.
Just for the record: the last twenty-five years or so have also seen the
introduction of translation prizes. They are given out internationally by such
bodies as Unesco (no Danish winners), the European Commission (one Danish
prize-winner), and also by various institutions at the national levels. They are
tokens of appreciation, rewards for good work, and occasionally they are
mentioned briefly in the newspapers. But I have yet to see the day when such
a prize boosts sales substantially.
6
Received opinion has it that Karen Blixen first did her work in English and then retold
it in Danish. Given the types of linguistic and content divergencies, I find this is
extremely unlikely. It must be the other way round in the vast majority of cases. These
linguistic divergencies have been touched upon in Dollerup et al (1990: 273-274). Dr.
Kristine Anderson, Purdue University, is at present (1997) studying 'Karen Blixen as a
bilingual writer'. Her preliminary findings seem to contradict my views. Many nations
actually boast of writers who do their own translations, e.g. Samuel Becket (French into
English). A study of their procedures would provide us with supplementary information
on the issue at hand.
Cay Dollerup 53
Discussion
What I have described so far bears some similarity with the situation in the
Netherlands as discussed by Ria Vanderauwera (1985), but also considerable
dissimilarity. The differences may lie in my having taken a historical bird's
eye view and therefore found more success than she. I also think that Danish
literature is better off for the simple reason that — despite the official policy
— it is not subsidised to any appreciable extent. This means that Danish books
are translated, if at all, on their own merits.
Vanderauwera studied English response to Dutch literature. I have not
studied foreign response to Danish literature systematically: I noted that
Bodelsen was translated into Spanish, because I studied an excerpt and found
it distorted. I vividly remember how the criminal's simple action of hiding
money in a box became an incomprehensible operation in Spanish, and felt
assured that this probably met Spanish expectations about how complicated life
is in Denmark. The inaccuracy corresponds with the results I have met in other
works, so I believe that most real-life literary translation will always show less
fidelity to the original than we accept in translation classes. I even believe that
translations from small language cultures will be more inexact than translations
from major languages into minor language cultures. But to return to the
question of success: I have noticed cursorily in my newspaper that Stangerup
met with French critical acclaim, and that Peter Høeg was on the bestseller list
in the US for a couple of months. On the other hand, I have studied Danish
critical response to foreign literature systematically, and my findings are that,
as a result of some incisive debate, critics have finally come round to taking
translation sufficiently seriously to assess the 'quality', which is normally one
or two sentences on the felicity of the style. They are clearly not bothered with
undertaking a detailed collation with the original.
But otherwise, what are the paradigmatic lessons of this discussion?
First and foremost it is obvious that the models of translation discussed at
the beginning do serve to illustrate the process of translation as communication
at some level or other. But they must not blind us to the fact that the vast
majority of texts are not propelled into translation the moment they are created
and produced. They are translated because of forces which are external to the
text, and they are translated in an interplay between the target culture and the
source message which I have here termed imposition' and 'requisition'. To
illustrate this, the communication model must be modfied as follows:
Sender (sending culture) — message > < translator > < recipient (receptor culture)
The arrows indicate the degree of intense interplay in the process of
cultural transfer in translation.
In a larger perspective, I believe that imposition will continue to exist as
long as there is (a) power and superiority, and (b) tangible objects for
discussion. Democracy and the desire for profit may make for adaptation to
target languages, but not total severance from source texts.
As far as literature is concerned, ideologically-driven translation is
unlikely to survive into the next century, for the simple reason that computer
networks will bypass censorship. From a narrow-minded perspective, one can
make a case that the European Union subsidised translation — whose
importance is negligible anyway — is ideologically-driven, but since the
motive is to further the minor languages in the name of linguistic and cultural
equality, the motive is far different from that of the Soviet state.
Conversely, I do believe that, although there will always be room for
idealism, translation will become an increasingly money-driven activity due to
market forces where publishers will have a much greater say than authors and
translators. The publishers' professional insight into the advertisement
channels, command of the distribution networks, and their intimate knowledge
of the potential audiences, will enable them to monopolise the market.
Compared to the world at large, Denmark (and for that matter the rest of
Scandinavia) may be a special case since we have been around for more than
a thousand years and our polar bears in the streets make us sufficiently exotic
for others occasionally to requisition our literatures.
This is not the case with newly emerging nations which, I pointed out,
have a legitimate wish to have their national cultures recognised. Is there a
lesson in this study? I think so: it is impossible to beat the market forces with
subsidies and any number of prizes for translators. A few authors make their
own translations, and are thus their own 'bridgeheads'. Mention has already
been made of Adam Oehlenschläger and Karen Blixen in Danish letters. For
others, national publishers may establish professional money-driven contacts
with foreign publishers at book fairs, and, as we have (presumably) seen with
Bodelsen and Høeg, the procedure may be successful. In principle there is a
third option: in so far as the cost and effort are accepted by the political
powers that be, it should hypothetically be possible to avoid the inbuilt target-
language directionality we find in normal 'cultural bridgeheads' by having
subsidised translators and teams of professionals to translate and revise
Cay Dollerup 55
translations of national classics.7 Even so, this procedure is fraught with

dangers ranging from infelicitous phrasings immediately seized upon by
reviewers as "bad translationese", to distribution problems in the target
nations. In other words, whether we like it or not, the 'cultural bridgeheads'
always constitute by far the most efficient avenue for the exportation of
national literatures.
Perhaps it is possible to further the process by introducing national
masterpieces — in the form of subsidised classics in gateway language
versions, notably in English — in translation by non-native speakers and hope
for the best, that is, for 'cultural bridgeheads' in other cultures to take note of
them, but I know of no successful example.
In my view, if national literatures are to become international, this must
take place according to a natural process: nations have to cultivate their
'cultural bridgeheads', especially those in gateway languages and bide their
time until they voluntarily start requisitioning literature. I am not saying the
result is perfect in terms of fidelity, once a work of literature has been over a
couple of gateway languages, but it is the price which minor language societies
have to pay for getting themselves heard. Hopefully, some traces of the
original work will still be there — witness the brothers Grimm and Hans
Christian Andersen.
References
Dollerup, Cay & Iven Reventlow & Carsten Rosenberg Hansen. 1990. "Reader, text,
translation and interpretative potentials." Multilingua 9, 271-284.
Dollerup, Cay. "Translation as a creative force in literature: the birth of the European
Bourgeois Fairy-Tale". The Modern Language Review 90 (1995), 94-102.
Dollerup, Cay. 1996. "The emergence of the teaching of translation. " In: C. Dollerup & V.
Appel (ed) Teaching Translation and Interpreting 3. New Horizons. Amsterdam: John
Benjamins, 19-30.
Levy, Jin. 1969. Die literarische Übersetzung. Theorie einer Kunstgattung. Frankfurt:
Athenäum.
Möller-Christensen, Ivy York. 1992. Den gyldne trekant: HC Andersens gennembrud i
Tyskland 1831-1850 med tilhørende bibliografi. Odense: Odense University Press.
Monhanty, Nirinjan. 1994. "Translation: an integration of cultures". Perspectives: Studies in
Translatology 2, 194.
7
In principle there are various ways of doing it. The main point is that most countries,
including large ones, do subsidise translations of literature into foreign languages. This
goes for the People's Republic of China (its subsidised journal is available from
embassies [personal communication from Eva Hung]), Slovenia (personal communication
from Meta Grosman), the Netherlands (Vanderauwera). I am sure readers can supplement
this list.
Nida, Eugene A. 1964. Toward a Science of Translation. Leiden: Brill.

Nielsen, Jørgen Erik. 1966. Den engelske litteraturs dyrkere i Holsten omkring 1820.
København: Gad.
Nielsen, Jørgen Erik. 1976. Den samtidige engelske litteratur og Danmark 1800-1840.
Copenhagen: Nova.
Uwajeh, M.K.C. "The case for a performative translatology". Perspectives: Studies in
Translatology 2, 247.
Vanderauwera, Ria. 1985. "The response to translated literature: a sad example." In: T.
Hermans (ed.) The Manipulation of Literature. London: Croom Helm.
Wilss, Wofram. 1982. The Science of Translation. Tübingen: Narr.
The impressionistic approach to translation
theorizing; or: Twentieth-century Chinese ideas of
translation through the Western looking-glass
Leo Tak-hung Chan
Much of the current evaluation of Chinese translation theory has tended

towards one of two extremes: either it has been valorized as belonging to a
distinctive, separate tradition, so that any attempt to seek Western equivalents
can only be futile, or it has been denigrated as lacking in analytical depth and
philosophical insight as compared with Western translation theory. There is
some truth in both of these views, though difference does not need to be
equated with inferiority or, for that matter, superiority. Speaking of the
distinctiveness of Chinese views of translation, it is a well-known fact that in
China, translation has for centuries been regarded as a marginal, if not trivial,
activity. St. Jerome's (3467-420) belief that translations can be used to
expropriate ideas from another culture to enrich one's own would have found
little favor with the Chinese. Chinese thinking on translation remained for
some time strongly influenced by an attitude which saw the target culture as
infinitely superior, and hence not quite the "recipient" — until the tables were
turned at the beginning of the present century.
As for the criticism that Chinese translation is deficient in analytical rigor,
it must be admitted that many Chinese translation theorists are prone to vague,
impressionistic assertions concerning translations. That is the case with the
early Buddhist translator-theorists working in the second to the tenth centuries,
with the seventeenth and eighteenth-century Christian converts who translated
religious and scientific writings from the West, and even with the turn-of-the-
century theorist Yan Fu (1853-1921), whose "three principles of translation"
practically set the perimeters for present-day discussions on translation in
China. This impressionistic bent is evidenced in the direct borrowing of
terminology from the discourse of traditional literary criticism, presumably in
the absence of existing terms for the description of translated works. It is not
until the extensive importation of Western linguistic parlance since the sixties
that a more systematic, less subjective, analysis of the translational process was
made possible.
What this scenario reveals in effect is that, up until recently, intuitive
judgments concerning a translation often formed the basis for theory. This
showed itself in a proclivity to theorize with reference to "good" translations
as opposed to "bad" ones. Before the linguistic approaches of theorists like
J.C. Catford and Eugene Nida came to China, there was in Chinese translation
theory less emphasis on the translation process — on what happens in
interlingual transfer — than on the quality of the product itself, and on what
constituted a good translation. For James Holmes, translation theory is distinct
from criticism in that theory is concerned with evolving principles and models,
not "in describing existing translations, observed translation functions"
(1988:73), whereas criticism always focuses on translated texts and inevitably
entails an element of subjectivity. If that is the case, was much of the
discussion that passed for translation theory in China actually translation
criticism? Or was this a theory that focussed more on description and
evaluation of the product than on analysis of processes? These are issues that
this paper seeks to address, through a study of the key ideas propounded by
noted translation theorists of the first half of the century, among them Yan Fu,
Fu Lei (1908-1966), and Qian Zhongshu (1910- ).
Yan Fu's three principles
Yan Fu's three principles — fidelity (xin), fluency (da) and elegance (ya)1 —
were widely accepted as essential criteria used to discuss translations ever since
their appearance almost a century ago in Yan's preface to his own translation
of T.H. Huxley's Evolution and Ethics (1898). They have also become the
fundamental tenets of twentieth-century Chinese translation theory, and though
there have been attempts to remove "elegance" from the list or replace it with
other principles, the importance of fidelity and fluency has gone pretty much
unchallenged. Perhaps these three principles are best defined by Yan himself,
1
Yan's three principles have been variously translated: xin also as "faithfulness", da also
as "comprehensibility" or "readability," and ya also as "polish" or "embellishment." The
three translations in this essay have been chosen because they would be readily
understood by those familiar with the current Western discourse on translation theory.
"Fluency" is used in the sense that Lawrence Venuti intends it to mean (Venuti 1995).
Leo Tak-hung Chan 59
rather than by the multitude of translation theorists in his wake who sought to
extract other meanings from them:
Translation involves three requirements difficult to fulfil: fidelity, fluency and
elegance. Fidelity is difficult enough to attain but a translation that observes the
rule of fidelity but is not fluent is no translation at all. Fluency is therefore of
prime importance. Since China's opening to foreign trade by sea, there has been
no lack of interpreters and translators. But if you assign them any book to
translate and tell them to meet these two requirements, few can succeed.2
It is easy to see the degree to which fidelity, elegance, and especially fluency
are terms of an evaluative nature, and indeed, Yan Fu proceeded in his treatise
with a critique of his own translation of Huxley's philosophical work. He noted
how much he had tampered with the original text in the interest of fluency: he
freely added to or deleted from it, since to him the translation should not be
unnecessarily constrained by the linguistic structures of the source text. For a
brief while it appears that he was privileging fluency over and above the other
two terms of reference, though a little later on he observed that, while there
should be room for the translator to re-create, this was nevertheless "not the
right way of doing a translation" (1973:4). Hence, to cut short the ongoing
debate on whether Yan Fu regarded fidelity or fluency as the more central
criterion, we need to note that, in principle (as against even his own actual
practice), he stood on the side of fidelity to the original. In so doing, Yan Fu
falls squarely within the tradition of the majority of Bible translator-theorists
in the West, for whom faithfulness, or respect for the source text, was to be
defended as a virtue.
For some years there have been rather harsh criticisms of Yan Fu's theory
of translation, most of them directed against his principle of elegance, and
some against that of fidelity. Several scholars underlined the uselessness of
"elegance" as an analytical term, and asserted that Yan Fu had included it in
his tripartite model simply because he wanted to suggest that the ornate
classical prose style of the Tongcheng school, in which his Evolution and
Ethics was translated, was the best language for translations.3 Now that such
period tastes have become outmoded (and plainer styles preferred), so should
the criterion of elegance. Others, eager to elevate the criterion of fluency,
2
The translation is adapted from C.Y. Hsu (1973:4).
2
Among those who suggested doing without "elegance" is Qu Qiubai, for whom this
criterion is counter-productive and undermines the effectiveness of the other two criteria.
For Frederick Tsai, another prominent twentieth-century translation theorist, it can be
replaced with "adequacy" (tie) (1972:18).
argued that the pursuit of embellishment in translations can be subsumed under

"fluency", since whatever style is chosen, the goal is simply to attract readers
to the translation. A fluent style could serve the purpose even better than an
elegant one. For the present writer, the problem with both terms, elegance as
much as fidelity — or even fluency — is their lack of specificity, which
weakens considerably their use as analytical tools, so there are as many
interpretations of them as there are theorists who choose to talk about them.
As we shall see, such vagueness of reference is to be seen in several other
recurrent terms in Chinese translation theory.
While Yan Fu's ideas have by and large provided the framework for
Chinese thinking about translation in the present century, a little observed fact
is that there was yet an alternative approach to translation theory at the end of
the nineteenth century, expounded by the leading philologist of the time and
a contemporary of Yan's, Ma Jianzhong (1845-1900). While spending the
greater part of his time writing a voluminous grammar of the Chinese language
based on borrowed Western grammatical categories,4 Ma presented "A
Proposal for the Establishment of a Translation Bureau" in 1894 (cf. Chen
1992:99-103). In this treatise he outlined an approach to translation drawing
on the insights of what must be termed (with some hindsight) contrastive
linguistics. For Ma, in order to succeed at his task, the translator needs to
analyze with the minutest care the source and target languages. By placing
together for comparison individual words and sentences from the two
languages, he seeks to identify the causes for similarities and differences in
expression, and only after thoroughly understanding the original would he
proceed to translate.
Ma Jianzhong differs markedly from Yan Fu in his emphasis on close
textual analysis and his valorization of the literal method in translation. In
contrast to Ma, Yan Fu would appear to be a proponent of "paraphrase" —
which John Dryden (1631-1700) defined as "translation with latitude"
(Schulte/Biguenet 1992:17) — although he did concede, as we noted above,
that fidelity is something not to be disregarded. Of course Ma's attention to the
language of the original (and that of the translation) did at times go to
exaggerated lengths; with philological zeal he exhorted the translator to pay
special heed to the etymologies of words, as well as semantic changes over
time. Nevertheless one will not have been amiss in viewing Ma Jianzhong as
the first of a line of Chinese linguists who actively enlisted the aid of Western
linguistics to explicate Chinese grammar and syntax; indeed, he is the pioneer
4
For an extended discussion of Ma's Grammar, see Shen Xiaolong (1992:180-218).
of Chinese translator-theorists who adopted a language-oriented approach to

translation, focussing on equivalence in translation. Yet the rise to prominence
of Yan Fu's three principles was paralleled by the virtual neglect paid to Ma's
ideas in the present century. The linguistic turn was one that Chinese
translation theory was never to take, at least not until after mid-century, when
theorists like Liu Miqing and Jin Di appeared on the scene.5
Fu Lei's "likeness-in-spirit"
Meanwhile, the stock of impressionistic terminology with an evaluative

coloring continued to expand. Perhaps the fourth most widely used term in
twentieth-century Chinese translation theory is Fu Lei's "likeness-in-spirit"
(shenxi). To many, Fu Lei had released the discussion of translation from the
constraints imposed by Yan Fu's three principles with his introduction of this
translational paradigm in 1951, in his preface to his second rendition of
Balzac's Le Père Goriot (cf. Jin 1994:208). Fu left no doubt that his was a
term appropriated from traditional Chinese aesthetics, a term associated in
particular with painting criticism. According to him, "As far as its effects are
concerned, translation should be like copying a painting. What is aimed for is
not affinity in shape but likeness in spirit."6 Though "affinity in shape"
eventually came to be redefined by translation scholars brought up on Western
linguistics as "formal equivalence," it is clear that Yan Fu's use of the term
was more vague: he merely intended it to refer to whatever is not "likeness-in-
spirit," the paired but opposed term. The two terms only set up a continuum
of sorts with an evaluative prejudice, since the rendering of the spirit is
adjudged to be infinitely superior to that of, if we may, the "body."
Other than the evaluative bent, Fu Lei's terms suffer also from a looseness
of reference; in fact "likeness-in-spirit" has remained perennially enigmatic.
5
Among the most influential books on translation theory written for a Chinese audience
is Liu Miqing's Present-day Translation Studies (1993). In 11 chapters it deals with
"translation as a discipline", "a model for Chinese translation theory", "translatability and
untranslatability", "the aesthetics of translation", "the translation of style", and so forth.
The contrastive linguistics background that informs Liu's discussion throughout is made
evident in his detailed references to the ideas of Western linguists like Saussure,
Humboldt and Martinet, among others. Ji Di collaborates with Eugene Nida in writing
their book On Translation, again a widely popular text in translation theory currently
cónsulted by both Chinese teachers and students of translation.
6
For Fu Lei's ideas on translation, see his Essays on Fu Lei's Translations (1981). For
a recent study of the various aspects of his life and work, see Serena Jin (1994).
Like Yan's three terms, it has kept theorists busy hunting for exact
connotations for decades, without coming any closer even today (than forty
years ago) to a grasp of its precise implications. Innovative as it may seem at
first sight, when understood in context, this concept has an ancestry traceable
back to discussions of "spiritual assonance" (shenyun or fengyun) in the
twenties and thirties. At the time these terms were most often bandied about
by poetry translators like Guo Moruo (1892-1978), translator of Shelley and
Goethe, and Zhu Shenghao (1912-1944), translator of Shakespeare. Guo
Moruo's discussion of "the achievement of spiritual assonance in translation"
in an article he published in 1922 is especially pertinent to the present
discussion. For him:
The translator of poetry does not exercise his skill through checking up the
dictionary for others, nor does he act as if he is deciphering telegrams at the
telegraph office. The life of poetry resides in an inherent musical spirit. . .If we
simply translate poems literally, then we turn out translations not of an artist, but
of a linguist (1992:268).7
Two telling points are conveyed by this passage. First, in spite of the fact that
Guo Moruo shows a keen concern for translating the essential spirit of a work
of art, he still offers little help in clarifying the meaning of the term "spirit"
— which for him seems largely a matter of rhyme and metre. Second, once
again there is an assault on the linguistic approach, this time through a
disparagement of the linguist's concern for capturing the literal meaning, or
"equivalence", in contemporary translatological terminology.
The painter/translator comparison, as well as the dichotomy stipulated
between the outward "shape" and the inward "soul" of a literary work, reminds
us how closely this school of Chinese translation theorists resembles
seventeenth and eighteenth century Western translation theorists like Dryden
and Alexander Fraser Tytler (1747-1814). For example, Tytler — whose
theories were introduced to the Chinese through Zheng Zhenduo (1898-1958)
in an article, "Three Problems in Translating Literature" (1921) — has said
that, even without using the same colors, the translator has to give his picture
the same force and effect of the source text, to re-capture the "soul" of the
author. Yet this is not to suggest any direct Western influence on Chinese
translation theory; quite on the contrary, a term like "conveying the spirit" has
occurred in as ancient a Chinese text as The Book of Changes, and terms like
"Spiritual assonance" have for centuries figured prominently in the poetry-talk
7
For a discussion of Guo Moruo's views, see Chen Fukang A Draft History of Chinese
Translation Theory (1992:262-272).
(or poetry criticism) tradition.8 Hence one would be missing the mark if one
attempted to re-cast Fu Lei's ideas in modern Western linguistic discourse. To
re-interpret "likeness-in-spirit" as equivalent to Eugene Nida's theory of
"dynamic equivalence," for instance, serves little more than to delimit the field
of reference of this term. As is typical of critical terminology used in
twentieth-century Chinese translation theory, their vagueness is also partly the
cause of their continued relevance.9
Qian Zhongshu's "realm of transformation"
In common with Yan Fu's three principles and Fu Lei's all important aesthetic
criterion, Qian Zhongshu's "realm of transformation" (huajing) describes what
an ideal translation is like, differentiates the good translation from the bad, and
contains hidden echoes of similar terminology from traditional Chinese poetics
and art criticism. Qian's critical term is marked by even greater imprecision
in that it simply posits a state that the successful translation is supposed to have
reached, and which is out of bounds to poorer translations. Unlike his
predecessors, however, Qian does not define the "realm of transformation"
through a critical discussion of his own work. In his seminal article on Lin Shu
(1852-1924), renowned translator of Charles Dickens, Walter Scott and Rider
Haggard, Qian began by talking briefly about the etymological and semantic
associations of the Chinese character yi ("to translate"), to which we shall
return below. Then he explained what he meant by "transformation":
The supreme principle for literary translations is "transformation." One can be
said to have attained this state when, in converting the words of one language
into those of another, no traces are left of one's having been constrained to
accommodate linguistic differences to which one is habituated, though the "feel"
of the original is fully conveyed. (1984:696)
Lest the sources of Qian Zhongshu's theory be thought of as completely
Chinese, especially given the Buddhist and Daoist overtones carried by the
term "transformation," one needs to be reminded that Qian's sources were in
fact Western. In a footnote, he said that a similar translational criterion had
8
Wong Wai-leung has traced the use of impressionistic critical terms in the discussion of
traditional Chinese poetry. For him, more of the terms are used descriptively and
evaluatively, and the analytical terms are scarce. Wang Wai-leung (1976), esp. Chap. 3.
9
Among those who have registered their dissatisfaction with "likeness-in-spirit" is Huang
Yushi. See Huang Yushi (1995:285).
been mentioned in the seventeenth century by the French scholar, George

Savile, First Marquess of Halifax, and then in the present century by the
German scholar Wilamowitz-Mollendorff, as well as the French poet Valéry.
In this way Qian's ideas become clothed in a cross-cultural guise. The
metaphor that Qian proposed for this kind of perfect translation is the
transmigration of souls (again a phrase with Buddhist associations), wherein
the body undergoes a transformation, but the "soul" is retained. This may
sound oddly similar to Fu Lei's "likeness-in-spirit", yet Qian's theory of
transformation is hardly a variant version of the latter. As Qian's detailed
analysis of Lin Shu's translations later on in the essay shows, this
transformation can take sundry forms, producing translations that are
immensely successful while differing on the surface from the original.
Hence, to say that Qian Zhongshu's "realm of transformation" remains
very much an impressionistic jargon and not of much analytic utility is not to
belittle Qian's contribution as a translation theorist. At its very least the idea
of transformation implies that the translator can have great laxity as well as
latitude as he carries out his task. Qian, too, defines the function of a "good"
translation differently from theorists before him, in a way that renders his
theory of transformation justifiable. For him, "a good translation annihilates
itself" (1984:689); by enhancing readers' interest in the original, it encourages
them to seek out the source text, leaving the translation behind. By contrast,
a bad translation annihilates the original; the reader will not want to read either
(contrast Arrojo, in this volume). In his role as mediator between the original
and the translation, the translator uses all the energies and skills at his disposal
to effect a successful transformation. By thus re-orienting the perspective of the
translator, Qian in fact opens the door to a consideration of the possibility that
the translated text can be an improvement on the original, and the translator
can exercise judgments as to how his source can best be translated.
With Qian Zhongshu's notion of total transformation, of the original text
being "reborn" as a translation, we also come very close to a contemporary
Western conception of the autonomy of the translated text which lives a life of
its own, and which may even bring the original work to completion. Jacques
Derrida, for one, has remarked that "transformation" is a term that he believes
should replace "translation":
In the limits to which it is possible or at least appears possible, translation
practices the difference between signified and signifier. But if this difference is
never pure, no more so is translation, and for the notion of translation we would
have to substitute a notion of transformation: a regulated transformation of one
language by another, of one text by another. (1981:26)
Qian Zhongshu sought to rationalize the connection between "translating" and

"transforming" by recourse to some verbal antics. German readers are already
familiar with the semantic links between Übersetzung on the one hand and
transfer and transport on the other, while Italians can ponder with bemusement
the maxim, "Traduttore, traditore." For Qian, the Chinese character for
"translation" (yi) has etymological and associative connections with the
characters for "seduction" (you), "error" (e), "mediator" (mei), and
"transformation" (hua). These express precisely for him the manifold aspects
of translation: the translator tries his best as a mediator between texts to seduce
the reader, to lure him to the original; the translator is always liable to errors
in crossing from one language to another, from one culture to another; and of
course the translator "transforms". And so, like his Western counterparts,10
Qian forges linkages between terms, which he then uses to build his
theory.11
The purpose of the foregoing discussion, however, has not been to argue
for convergences between Chinese and Western thinking about translation as
a process of cultural and linguistic transfer, but to define twentieth-century
Chinese translation theory with greater precision. This we have done by
looking at five central concepts, and on the whole it appears that, while
comparisons at every point can be made with Western theories, Chinese
theorists have very much gone their own way in that they have manipulated
terms derived from traditional Chinese poetics in general and painting criticism
in particular, to describe a realm of activity that suffered initially through its
marginal status. The choice of terminology, however, reflects a special
Chinese emphasis on evaluating (rather than describing or analyzing) the
translated product impressionistieally; discussions of translation almost
invariably begin by proposing ways of "telling the good translations from the
bad ones". The preference for evaluation, together with the overall de-
emphasis of the linguistic approach, and the blurring of the lines of
demarcation between theory and criticism, are perhaps the distinguishing
10
For example, Eugenio Donato (1985:127) has taken advantage of the fact that
Übersetzung has as one of its senses "leaping over an abyss" to make his point on
" specular translation ".
11
There are other semantic links mentioned by Qian that may be of some interest: yi has
been defined by traditional Chinese philologists as referring to the "transmission of the
language of the barbarians, of birds and beasts"; fan refers to "the turaing-around of a
piece of embroidered silk", so that everything faces the opposite direction. One may add
that one of the homophones for yi also means "to change".
hallmarks of Chinese translation theory for the better part of the present
century.
References
Chan, Sin-wai and David Pollard (eds.) 1995. An Encyclopaedia of Translation: Chinese-
English, English-Chinese. Hong Kong: Chinese University of Hong Kong Press.
Chen, Fukang. 1992. Zhongguo yixue lilun shigao (A Draft History of Chinese Translation
Theory). Shanghai: Waiyu jiaoyu chubanshe.
Derrida, Jacques. 1981. Positions. Translated by Alan Bass. Chicago: University of Chicago
Press.
Fu, Lei. 1981. Fu Lei lunwenji (Essays on Fu Lei's Translations). Hefei: Anhui remin
chubanshe.
Holmes, James. 1988. Translated! Papers on Literary Translation and Translation Studies.
Amsterdam: Rodopi.
Huang, Yushi. 1995. "Form and Spirit." In: C. Sin-wai/D. Pollard (eds.), 277-287.
Jin, Sheng-hua (Serena Jin) (ed.) 1994. Fu Lei yu tade shijie (Fu Lei and His World). Hong
Kong: Joint Publishing Co.
Liu, Miqing. 1993. Present-Day Translation Studies. Taibei: Shulin chubanshe.
Luo, Xinzhang (ed.) 1984. Fanyi lunji (Essays on Translation). Beijing: Shangwu yinshuguan.
McDonald, Christie V. (ed.) 1985. The Ear of the Other: Otobiography, Transference,
Translation. New York: Schocken Books.
Qian, Zhongshu. 1984. "Lin Shu de fanyi" (The Translations of Lin Shu). In: Luo 1984, 696-
725.
Schulte, Rainer and John Biguenet (eds.) 1992. Theories of Translation: An Anthology of
Essays from Dryden to Derrida. Chicago; University of Chicago Press.
Shen, Xiaolong. 1992. Yuwen de chanshi: Zhongguo yuwen chuantong de xiandai yiyi
(Explicating Meaning: The Modern Meaning of the Chinése Linguistic Tradition).
Shenyang: Liaoning jiaoyu chubanshe.
Tsai, Frederick. 1972. Fanyi yanjiu (Studies on Translation). Taibei: Dadi chubanshe.
Venuti, Lawrence. 1995. The Translator's Invisibility: A History of Translation. London:
Routledge.
Wong, Wai-leung. 1976. Chinese Impressionistic Criticism: A Study of the Poetry-Talk
Tradition. Ohio State University: unpubl. PhD thesis.
Yan, Fu. 1973. "General Remarks on Translation." Translated by C.Y. Hsu. Renditions 1,
4-6.
Transgression and circumvention through translation
in the Philippines
Ubaldo Stecconi/Maria Luisa Torres Reyes
History
The history of the Philippines is one of colonial rule: three and a half centuries
under Spain, about fifty years under the U.S. and five years of Japanese
occupation result today in a feudal-like dominance of local elites sustained by
their collaboration with foreign interests. The 25 years after the 1946
independence were marked by an increasing political awareness of organized
sectors of the population resulting in widespread popular discontent with the
ruling class. Thus, the late 1960s saw the emergence of nationalist movements
and mass organizations which sought to reform or overthrow the country's
neo-colonial structures and institutions. In January 1970, a student protest
dubbed 'The First Quarter Storm' broke out; its repression initiated a history
of outright state violence against the opposition which eventually resulted in the
declaration of Martial Law on 21 September 1972.
In the years leading to the Marcos dictatorship, the major colleges and
universities saw a surge of militancy: students, teachers and younger members
of religious institutions rallied behind the issues of nationalism, socialism and
democracy. In the Ateneo de Manila University — an institution owned by the
Society of Jesus and noted for its elite and westernized orientation — the
students were agitating for the Filipinization of administration and the
curricula. On 4 August 1970, Trend, a student paper at the Ateneo, spelled out
the meaning of Filipinization as follows:
Involvement in the cause of social transformation must be a characteristic of the
University faculty. It should be fearless and unflinching even to the point of
supporting a radical restructuring of our society, even as it is true the main
contribution of a faculty member to the nation will be along the lines of his
trained competence both within the University and outside. (Trend 1970:3)
Ateneo and Kamao
In this environment, the Philippine Studies Department of Ateneo decided to

devote two issues of its quarterly journal Katipunan to translation (1 (3 and 4),
issued in July and October 1971, respectively) that were soon re-printed in
Kamao (Fist), an off-print edition. Katipunan's editor recently described the
publication as follows: "We printed out so many copies of Kamao because we
wanted as many people to read it. We felt so strongly about this so that later
we were giving them away for free everywhere to whoever was interested and
wanted to read it but couldn't afford to buy it." (Tiongson 1994).
Kamao is a collection of translated poems from around the world with a
prevalence of third-world texts (table of contents in Appendix A). Among its
aims were: (1) showing the viability of Filipino as a language of intellectual
discourse; (2) encouraging professors and students alike to use Filipino in their
discussions; and (3) making literary texts available to the common people.
Kamao's texts also intended to give an image of revolutionary struggles around
the world to young writers who were steeped in the Anglo-American literary
canon adopted by their universities. According to a leading figure in the
project:
Kamao was principally meant for these young writers; as artists and activists, we
felt that they ought to be concerned with aesthetic norms as well to be effective
... to avoid slogans, for example, which already seemed endemic at that point of
activist writing. (Lumbera 1994)
Most of the translations included in Kamao were signed by the Translation
Committee of PAKS A (PAnulat para sa Kaunlaran ng SAmbayanan, or
'Writing for the Progress of the People'), or individually by its members.
PAKSA was an organization founded by nationalist writers politically active
during the First Quarter Storm. With PAKSA's appearance "for the first time,
nationalist writers and artists comprehensively and incisively analyzed the
history and condition of literature and art in the country, laid down new norms
in their creation and propagation and set a new direction and service, based on
the principles of Marxism-Leninism-Mao Zedong Thought." (Rebolusyonar-
yong... 1992:backcover).
After September 1972, PAKSA went underground together with all the
organizations suspected to be linked to the banned Communist Party of the
Philippines. Indiscriminate arrests and detention of known political dissenters
began even before martial law was announced. Political dissent moved into
detention camps, in the so-called political underground or 'elsewhere.' This
latter place was a volatile and ambulatory space which needed to be imagined
Ubaldo Stecconi/Maria Luisa Torres Reyes 69
and mapped-out day by day. 'There', the Left re-organised its resistance in
a precarious legitimacy status where a new coalition had to seek a range of
ways and byways in the often tortuous and tortured struggle against 'the
U.S.-Marcos dictatorship'.
The University of the Philippines and the Philippine Collegian
With the possible exception of Manila's largest institution, the University of

the Philippines, campus life was deeply affected by the forcible disbandment
of all organizations that were alleged to have overt or covert links with the
Left. Slowly, however, the students regrouped; at U.P. they initiated
campus-wide activities that on the one hand called for the right to organization
and freedom of expression, and on the other wanted to reclaim 'academic
freedom', insisting on their identity as intellectuals and artists. These efforts
became more systematic and braver until the Philippine Collegian — the
official student newspaper of the University of the Philippines — was
re-opened in 1973. On 4 February and 14 August 1975 two special issues of
the Collegian appeared. The February special issue was titled Tinig ng lkatlong
Daigdig, 'Voices of the Third World,' and was devoted to translations of
literary texts from Asia, Africa and Latin America (table of contents in
Appendix B). An editorial read:
The literary works in this slim anthology are characterized by vigor and passion
that only writing committed to a just cause can manifest. ... In translation, their
works are collective voices and hearts that eloquently speak against oppression
and exploitation anywhere in the Third World. ... This anthology is being offered
as a tribute to the brave people it represents, a reminder to writers that their
inexhaustible subject matter is their people's struggles, and a call to all readers
to support the vigorous and passionate art of the committed. ("A Tribute, A
Reminder, A Call," Philippine Collegian 1975a:2)
The August special issue was titled Tinig ng Pilipino, 'Philippine Voices'
(table of contents in Appendix C). The intent of this second issue was specified
in a note. "Many students and faculty members were surprised by [the
February] issue: ... if we have given attention to the literature of other
countries, why not to our own literature written in the different languages?"
("Pagsasalin Tungo sa Pambansang Kamalayan," Philippine Collegian 1975b: 16)1
The note admits that the knowledge of Filipino literature in the vernaculars
1
All quotations originally in Filipino have been translated into English by M.L. Torres
Reyes.
was scarce and claims that translation in the national language was vital for
building a national awareness and making these texts available to a larger
readership. Interestingly, it added: "... [our] literature is a weapon and a shield
against western literature which has long dominated our country and for so
long tied us to an orientation that blurs our view of the problems of the
country" (Philippine Collegian, 1975b: 16). Far from being in contradiction
with the previous issue, the August publication completes the overall strategy
of the Collegian editors: translations from foreign cultures and from the
various Philippine vernaculars should converge in the creation of the national
language.
To locate more accurately the political space occupied by the Collegian,
one should not forget that Marcos was particularly stern with the press and the
other media. On the day of the declaration of martial law, the national police
shut down all of Manila's media: press, TV and radio stations. In 1976 the
Collegian was ordered to suspend publication again and several editors were
imprisoned. However, the paper continued to circulate through unofficial
channels as it became one of the very few alternative and opposition news
sources available in Manila.
Translation
The three anthologies can be said to have followed a logic of their own as they
appeared in sequence. The overall principle of selection was clearly based on
an anti-canonical commitment since not a single Anglo-American author was
included; they made an explicit political statement in their expression of unity
with writers from a given geo-political space and a quest for international
solidarity on the basis of the right to self-determination and freedom from
domestic and foreign domination. In this way the site of oppression is
transformed into a site of resistance. The three collections were openly and
consciously at odds with the texts and views that passed for news and public
opinion under martial rule. As indicated earlier, apart from a total
militarization of social and political life, the oligarchic and colonial sectors that
supported the dictator tried to wage a blatant psywar through total control of
information. But this highly transgressive strategy was presented in the
ostensibly innocent guise of translations of poems and short fiction into
Filipino. It could be argued that these militant writers were reduced to using
translation in a critical move to circulate transgressive views. Not quite;
deploying translation's power to dodge repression while using repression's own
Ubaldo Stecconi/MariaLuisa Torres Reyes 71
mechanisms of power is itself an act of civil disobedience. Circumvention is

constitutive of transgression. Had the authors been allowed to write their own
protest literature — which they did, too, elsewhere — they would have lost
semiotic traits that perhaps only translation can provide.
Filipino as target language
Using Filipino as target language is a characteristic feature of the three

publications and should be regarded as a political objective in itself:
"translation into Filipino was an effort to 'Serve the people' ... every
self-respecting people should have a national language" (Lumbera 1994). The
lack of an established national language is seen as a "divisive tool that has
worked against the interest of the majority of the Filipino people. ... A
national language empowers the people because it can express their genuine
feelings, it does not alienate the Filipinos from themselves, it comes from the
heart. In this respect, its development is vital to the development of democracy
and an egalitarian society" (Lumbera 1994).
The second special issue of the Collegian, however, was potentially
problematic in this respect because it translated from other languages of the
Philippines, which are supposed to have an equal footing with the 'national'
language in a truly egalitarian society. What goes under the name of Filipino
is based on Tagalog, the language dominant in the region of Manila; however
natural it may seem that the language spoken in the capital becomes the
national language, this fact reflects a tendency towards domestic hegemony in
a multi-cultural setting like the Philippines, which is symptomatic of the
country's colonial history. This situation is not new in the history of translation
but it takes fresh specifications in the Third World:
Steiner suggests that the faithful translator "creates a condition of significant
exchange. The arrows of meaning, of cultural, psychological benefaction, move
both ways. There is, ideally, exchange without loss" (George Steiner, After
Babel:302). I need not reiterate the idea of the futility of such remarks in the
colonial context, where the "exchange" is far from equal and the "benefaction"
highly dubious, where the asymmetry between languages is perpetuated by
imperial rule. (Niranjana 1992:59)
The collections' translators were aware of such asymmetry; in his preface to
Kamao, for instance, Tiongson attacks the educational system set up by the
American colonizers in the Philippines as a "trap" to "capture the mind of the
Filipino" using the illusion of universalism as bait. He states that the Filipinos
were estranged from themselves and their values: "We wallowed in American
cigarettes, in American cars, in American clothes (the sunlight is hottest!), and
in nearly all things American from Colgate to Coke, from Dial to Dole and
Del Monte." Tiongson wondered whether what was presented in English
departments as "the experience of man all over the world" was actually the
experience of just North America and Europe. The western canon was
"swallowed whole" without considering whether it was good or bad for the
Filipinos. Such criticism would normally lead to an outright rejection of
universalism, but the blend of western scholarship, Filipino values and political
commitment produced a different position. "Universalism is not bad if it is not
used to blindfold the eyes of a country," it is actually good if scholarly
knowledge is used to understand the Filipino condition and pagka-pilipino,
Philippineness. "The foreign should be studied [on one condition]: make the
foreign serve the Philippines. Only in this 'narrow' sense will a real
universalism be achieved" (all quotes are from Tiongson 1971). Thus,
Tiongson proposes two reversals. Firstly he reverses the master-servant
colonial relation and turns it into the main point of his translation project.
Secondly, he proposes the paradox — a reversal of common-sense — that
'narrowness' is a condition of universalism. These views seem to depict a
world upside down, but if you are standing in the South of the world this may
be the best way to have a clear view.
Faithfulness and katapatan
How were these positions elaborated into translation strategies? The scholars
who organized the translation work were operating between a tension: on the
one hand, the western metaphysical tradition of essence, meaning and
faithfulness learned in school; on the other, their own Filipino material
conditions. As to the position towards western values, Tiongson's words were
eloquent; but how was the notion of faithfulness 'translated' into Filipino
practice as kcatapatan? We will try to show, with the help of a story, how
katapatan/faithfulness was caught in the middle of a semiotic tension.
The story is as follows: six hours' drive from Manila is Banawe,
Mountain Province; six hours' walk from Banawe is a place called Batad: a
steep mountain cut into rice terraces with three or four small scattered villages
said to be over two thousand years old. Due to the far-flung location,
everything is in its pristine form: the Ifugao language, huts and g-strings.
Batad is as close as you can get to the Filipino 'original' in the north of the
country. But then, since the huts are made entirely of hard-wood and straw,
at the end of each rainy season the villagers have to partly re-build their
homes. In the course of — say — a century or two, every bit of them will be
replaced with new material. But this is immaterial: it is still the same house.
Now, let us make a comparison. Western heritage is affected by adverse
environmental conditions too; the marble of Michelangelo's David in Florence
is attacked by acid rain and pigeons' droppings. If we pasted new marble
powder over every corroded limb and — in time — completely replaced the
material the statue is made of, would we think of it as 'the same' David? We
would not; in fact, the original David is kept safely indoors and what stands
in Piazza della Signoria is a mere copy, no matter how faithful to the original.
This story has some interesting implications. Two parallels can be drawn
between Batad and katapatan. Firstly, both respond to a criterion of nimble
adaptation to reality: the natural environment is pretty hostile in Batad, yet
homes are needed, so they have to be constantly rebuilt; likewise, political
struggle was fierce in the early 1970s in Manila, yet those messages of
resistance and struggle were necessary, so they had to be translated. For the
second parallel, we assume that the clearest observable fact in translation is the
replacement of old words with new ones. Now, a message's words can be
regarded as its material: while essentialistic westerners are always disturbed by
its replacement, a non-western-educated Filipino would be rather indifferent;
it would still be the same message. However, Kamao and the Collegian were
not translated by Ifugao mountain dwellers but by western-educated writers,
this is why faithfulness occupied a place of great tension in our translators'
theory:
Afraid of moving very far from the original, the translations tried to be as
faithful as possible to the English versions, even if often they seemed too literal.
When the meaning became vague in the literal translation, the nearest adaptation
into the Filipino idiom became the solution. Whenever possible, katapatan, clarity
and artistry became our standards. ("Talang Pampatnugutan," Philippine
Collegian 1975a:2)
To conclude, the excerpt maintains that faithfulness — even literality — is a
hallmark of translation, but when we read katapatan, we suspect that it spells
out a notion that contradicts the western metaphysics of the original.
Living in translation
An illustration from the second Collegian issue may provide in detail the
complex and problematic relationship at work in these translation projects. We
will analyze Kasaysayan ng Isang Liham the translation of a short story by the
Filipino writer Carlos Bulosan originally written in English in 1946 with the
title The Story of a Letter, The tale is about a poor peasant boy — the narrator
— who goes to America in the 1930s to escape the grinding poverty of his
feudal countryside of the Philippines. The boy's father needs to find someone
who could read for him a letter sent from America by another son, the
narrator's older brother Berto, who had emigrated there many years earlier.
Since the family is illiterate and the letter is in English, it remains unread for
three years. The narrator says retrospectively: "The suspense was hurting him
and me, too. He wanted me to learn English so that I would be able to read
it to him. It was the only letter he had received in all the years that I had
known him, except some letters that came from the government once a year
asking him to pay his taxes." (San Juan 1983:41).
Almost two decades later, the narrator is in the U.S. He remembers the
letter, writes to his father asking to send it over to him so that he could
translate it. Six months later, he receives the original, translates it, and happily
sends both the original and the translation back to his father. Many months
later he receives mail from his hometown's postmaster saying that his father
had died some years earlier. The letter which, says the narrator, "had driven
me away from my village and had sent me half way around the world" has
remained unread by his father. The very end of the story reveals the text of the
letter; it reads: "Dear Father (my brother wrote): America is great country.
Tall buildings. Wide good land. The people walking. But I feel sad. I am
writing you this hour of my sentimental. Your son. — Berto." (San Juan
1983:45).
The Filipino translation of this story included in the second issue of the
Collegian is among the most competent and moving: the language is poignant
and the narrative flows from a boy's consciousness growing up to be a man
who has learned to face the harsh realities of life. It is a stark image of the rest
of the world's marginalised people, with a tragic-comic touch. At every turn
one senses how the translators — about a dozen young students of the
University of the Philippines — were carefully weaving their way in and out
the source and target texts. But when the translation of the story gets into the
letter proper, one discovers that the only material that gets translated is the
parentetical clause. The translation reads: "Dear Father (anang kapatid ko):
America is great country. Tall buildings. Wide good land. The people walking.
But I feel sad. I am writing you this hour of my sentimental. Your son. —
Berto." (Philippine Collegian 1975b: 15).
What happened here? The translators understood that the passage was calling
out for translation just like the rest of the story; but it called for a very
different translation from the rest of the text: it was asking for a virtual rather.
than an actual translation. The broken and ungrammatical English is essential

to the sense of fragmentation; a tender, concrete, painful and ironic effect that
goes beyond merely linguistic and aesthetic considerations. The complex and
textured nuancing produced by the careful translation of the parenthetical
clause and its insertion in the truncated English of the letter is a translation
strategy just as daunting and radical as it is most difficult to replicate. It
produces a passage which now, in turn, claims for a re-translation in the
reader's mind. "Mahal kong Ama (anang kapatid ko): Dakilang bansa ang
Amerika. Gusaling matataas. Malawak, mabuting lupain. Mga taong
naglalakad. Pero malungkot ako. Sumusulat ako sa iyo sa panahon ng aking
kalungkutan. Ang iyong anak. — Berto."
In Filipino, the passage becomes devoid of the shocking recognition of the
truth about the colonial experience; the entire passage becomes perfectly
acceptable and intelligible as ordinary language. The choice to translate the
parenthetical aside distances the translators from the speaker in the letter, as
the reader is reminded of the utterance's 'locality' by anang, an almost
dialectal, everyday contraction of 'according to.' Here, in a singular stroke,
translation splits subjectivities within double quotation marks, as it shifts codes
between linguistic fragments. Today the translation evokes a re-textualisation
of a specific moment of the Filipino people's colonial subjugation; the
disjunction of identity and speech is aestheticized in the very materiality of
language. What the passage requires, is an imagined juxtaposition of languages
in dialogic translation constructed around the longing for a 'language' that
binds even as disrupts.
Viewed in terms of the problematic history of the Philippines as a
peripheral outpost of U.S. hegemony, the partially translated passage is witness
to the reality that postcolonial people of the world continue to live 'in
translation', irreducibly distinct in their specific experience of colonial and
neocolonial bondage. For all the current government's claims of prosperity in
the face of its inability to provide local employment, the Filipino people are
condemned to a diaspora as migrant workers. In their lands of exile, they try
to learn the language of their employers and masters as they have done for
centuries, but their hopes are spoken in their native languages. Between the
hope of freedom and decent life and the reality of indignities they suffer
abroad, the Filipinos live in translation to survive.
References
Kamao. 1971. Kamao. Quezon City: Ateneo de Manila. (Reprint of Katipunan 1 (3 and 4),
Quezon City: Philippine Studies Department, Ateneo de Manila).
Lumbera, Bienvenido. 1975. "Ang Tagasalin at ang Kanyang Mambabasa". Philippine
Collegian 1975b, 3-4.
Lumbera, Bienvenido. 1994. "Interview with M.L. Torres Reyes", 28 December 1994.
Niranjana, Tejaswini. 1992. Siting Translation. History, Post-Structuralism and the Colonial
Context. Berkeley and Los Angeles: University of California Press.
Philippine Collegian. 1975a. Philippine Collegian, special issue 4 February 1975, 'Tinig ng
Ikatlong Daigdig". Diliman: University of the Philippines.
Philippine Collegian. 1975b. Philippine Collegian, special issue 14 August 1975, "Tinig ng
Pilipino". Diliman: University of the Philippines.
Rebolusyonaryong... 1992. Rebolusyonaryong Panunuring Masa sa Sining at Panitikan.
Quezon City: Kalikasan Press.
San Juan, E., Jr. 1983. Bulosan: An Introduction with Selections. Metro Manila: National
Book Store.
Tiongson, Nicanor. 1971. "Papaglingkurin sa Pilipinas ang Dayuhan". Kamao 1971, n.p.
Tiongson, Nicanor. 1994. "Interview with M.L. Torres Reyes", 29 December 1994.
Trend. 1970. Trend 2 (1), Quezon City: Ateneo de Manila.
Appendix A: Table of contents of Kamao. 1971.
Asia
China. Mao Tsetung: Ch 'angsha, Nyebe, Mga Ulap sa Taglamig; Lu Hsun: Taglagas 1935.
Vietnam. Ho Chi-Minh: Ang mga Paang Bakal, Payo sa Sarili, Mga Paghihigpit, Sa
Pagbabasa ng "Antolohiya ng Sanlibong Makata"; To Huu: Tandaan ang Aking mga Salita.
Korea. Yi Kwang-Su: Isang Anemona. Indonesia. Chairil Anwar: Ako. Turkey. Nazim
Hikmet: Isang Sigarilyong Hindi Ko Masindihan.
Latin America
Argentina. Ernesto Che Guevara: Awit kay Fidel. Cuba. Nicolás Guillén: Maipagbibili Mo
Ba?; Félix Pita Rodriguez: Riple Numero 5767; Alcfdez Iznaga: Maybahay at Kasama; Luís
Marre: Awit; Fayad Jamis: Buhay; Pedro de Oraa: Para Kanino?; Domingo Alfonso: Ang
Sining ng Tula. Chile. Pablo Neruda: Ang United Fruit Co., Ang mga Diktador, Gutom sa
Timog. Guatemala. Otto René Castillo: Mga Apulitikal na Intelektuwal; Marco Antonio
Flores: Rekiyem Kay Luis Augusto. Nicaragua. Fernando Gordillo Cervantes: Sa Isang
Yumaong Kabataan, Ang Presyo ng Isang Bansa. Peru. Javier Heraud: Ang Bagong
Paglalakbay.
Europe
Russia. Alexander Blok: Ang Labindalawa; Vladimir Mayakovsky: Mula sa 'Ubos-lakas,
Buong-dahas '. Germany. Bertolt Brecht: Awit ng Mangangalakal (Paglilibing sa Manunulsol
na nasa Kabaong-zink), Walang Sinuman O Lahat, Sa Posteridad.
Africa
Angola. Antonio Jacinto: Monangamba. Congo. Patrice Emery Lumumba: Bukanliwayway sa
Dibdib ng Aprika. Senegal. Leopold Sedar-Senghor: Panalangin para sa Kapayapaan; David
Diop: Makinig mga Kasama.
Appendix B. Table of contents of Philippine Collegian. February 1975.
Asia
Indonesia. Chairil Anwar: Dipo Negoro; Rivai Apin: Bantayog. Malaysia. Usman Awang:
Balita Mula sa Asya, Tatang Utih; Masuri S.N.: Ang Kaning ito na Aking Kinakain. Palestine.
Mahmoud Darweesh: Hinggil sa Pag-asa. Pakistan. Faiz Ahmed Faiz: Magsalita.
Philippines. Amado V. Hernandez: Ang Panday; Benigno R. Ramos: Gumising Ka, Aking
Bayan!. Bangladesh. Zakiul Huq: Paniniwala; Abu Bakr Siddique: Awit ng Bayan. Vietnam.
To Huu: Munting Luom; Anon.: Sa Tagsibol, Tayo'y Ikakasal. Turkey. Cedvet Kudret: Handa
sa Patay. India. Sahir Ludhianvi: Ang Taj Mahal; Shanmuga Subbiah: Pagkabasa ng Diyaryo.
Africa
South Africa. Peter Abrahams: Ako, Maykulay. Ghana. Kwesi Brew: Paghahanap. Liberia.
Roland Tombekai Dempster: Ito ba ang Aprika? Senegal. Birago Diop: Mga Kuwentong
Naging Duyan ng Aking Kamusmusan; David Diop: Sa Kanila 'y Inagaw nang Lahat; Leopold
Sedar Senghor: Itim Na Babae. Cameroon. Mbella Sonne Dipoko: Distiyero. Sao Tome. Aldo
do Espirito Santo: Nasaan ang mga Taong Sinaklot sa Unos ng Kabaliwan? Nigeria. Gabriel
Okara: Noong Araw; Wole Soyinka: Pag-uusap sa Telepono.
Latin America
Peru. Igor Calvo: Pabula; Cesar Vallejo: Masa. Cuba. Nicolás Guillén: Ang Burgesya.
Argentina. Leopoldo Marechal: "Pag naipaghiganti na ang malaking katampalasanang ito na
lumulukob sa Latin Amerika". Puerto Rico. Hugo Margenat: Mga Kawing. Chile. Pablo
Neruda: Mga Digma.
Appendix C. Table of contents of Philippine Collegian. August 1975.
Ilokano. Reynaldo A. Duque: Anakpawis; Fernando B. Sanchez: Hilaga; Donato B. Abanilla:

Ina, Narito Ako; Calixto A. Palino: Bayani; Cristino I. Inay: Sa Iyo, Bayan Ko.
Kapampangan. Juan Crisostomo Soto: Ang Watawat. Bikol. B. Alzaga: Tula Kong Kundiman;
O. C. Muni: Ang Puso Mo 'y Matutuyo. Sebuano. Gumer M. Rafanan: Yaong Itim Na Bathala.
Waray. Aniceto O. Llaneta: Ang Ibong Kanaway. Hiligaynon. Isabelo S. Sobrevega: Si
Pingkaw. Espanyol. Manuel Bernabe: Paglalakbay ng Kamatayan; Claro M. Recto: Sa Mga
Bayani ng '96. Ingles. Carlos Bulosan: Kasaysayan ng lsang Liham.
A call for descriptive Translation Studies on the
Turkish tradition of rewrites1
Saliha Paker/Zehra Toska
In this paper we shall first observe some of the most salient aspects of the
oldest Near Eastern translations of the collection of stories known in Arabic
and Turkish as the Kalilah wa Dimnah, and Kettle ve Dimne (KD) respectively.
This will be followed by a discussion on (a) the earliest known translation of
KD (by Kul Mesud) into Western (Old Anatolian) Turkish in the 14th century,
which was identified in the late 19th century but has only recently been the
subject of a full scholarly study (Toska 1989; 1991)2, and (b) the need for
descriptive research especially on the medieval rewrites that belong to the
formative period in the Ottoman-Turkish literary tradition. We think that the
concept of "rewriting" (Lefevere 1992:1-9, 47) is especially useful at the initial
stage of research, because it covers the multiplicity of forms in the Turkish
tradition that have been named and/or described as translations in literary
histories in one way or another, but not analyzed in the context of the
translational, literary, social, ideological expectations, constraints or norms that
underlie those forms.
It is generally agreed by modern literary historians that the 13th and 14th
centuries mark the beginnings of literary writing in Old Anatolian Turkish.
That most of the literature produced in this period appeared as rewrites of
Persian sources in the one or the other form, is an issue that has been
1
This paper forms part of a joint project, Studies in the Early History of Literary
Translation into Turkish, carried out at Bogazici University, Istanbul, with the
collaboration of Z.Toska (Dept. of Turkish Language and Literature), S.Paker (Co-
ordinator, Dept. of Translation & Interpreting) and N.Kuran-Burcoglu (Dept. of
Translation & Interpreting).
2
E.g. There is no reference to Kul Mesud's translation or other early Western Turkish
versions in the detailed "Genealogical Table of the Panchatantra" (Grube 1991: inside
cover).
recognized in literary studies on individual works but not explicitly addressed.
Persian and Arabic had already established their respective canons of written
literature. Turkish had not, and was only beginning to develop one through
various rewrites in prose and verse. One of the aims of this study is to draw
attention to this by examining, on the basis of currently available documentary
evidence, some aspects of Kul Mesud's KD and the network of relationships
between the text and the literary/linguistic/social system for which it was
intended.
The second half of the 19th century has long been recognized as an age
of translation in Ottoman-Turkish literature, because the European source
culture and languages were identified as "foreign" (Paker 1991). By contrast,
a very wide-ranging literary/linguistic/cultural transfer from Persian and Arabic
in various forms of translation in the medieval period (and even in the later
centuries) has largely been evaluated in terms of "influence" because it was
appropriated as part of the common Islamic tradition (Paker forthcoming). Our
concluding arguments will therefore emphasize how and why descriptive
translation studies can help us to see if we are dealing with a history of
"concealed" (Toury 1995:70f.) translation in Ottoman-Turkish literature, and
can thus contribute to the understanding not only of translation but of literary
history in the early period and in the classical, leading up to the 19th century.
The first translations of KD in the Near East
The origin of the beast fables bearing the title Kalilah and Dimnah (in Arabic),
has been traced back to the earliest (not the existing) written versions of the
Sanskrit Pancatantra, composed according to Hertel c.300 AD (de Blois
1990:1). The five books of the Sanskrit fables, and three stories from the 12th
chapter of the Mahabharata, the Indian national epic, apparently served as the
source-texts for the Pahlavi (Middle Persian) version, the first known
translation that was produced in the reign of the Persian (Sasanian) king
Khusroy I (531-597), also known as "Anoshagruwan" ,"Hüsrev Anu§irvan" in
Turkish (de Blois 1990:1, 13; Toska 1989:9).3 The first known Syriac version
3
François de Blois' excellent textual/critical study on the fables known in Arabic as
Kalilah wa Dimnah, aims to reconstruct the lost Middle Persian version by focussing on
the "oldest segment of the history of the book [...] in the Near East" (de Blois 1990:iii)
which, in fact, is of unique significance for the translation historian since it uncovers the
story of the first (6th century) translation, from Sanskrit to Pahlavi/Middle Persian, and
the translator's (Burzoy's) autobiography, both of which are incorporated in the translated
Saliha Paker/Zehra Toska 81
derived from the Pahlavi/Middle Persian is also from the 6th century, but it is
the Arabic translation by Ibn'al-Muqaffa which dates from the mid 8th century
(750) that has been the most influential as the Near Eastern source text. The
genealogy of the existing translations of the KD into languages of the West and
the East, including Neo-Persian, can be traced back to (the lost) Middle
Persian version, thanks to the extant manuscripts of the Arabic translation and
that of the Syriac version.
The (Arabic) title of the book of fables Kalilah wa Dimnah, bears the
names of the two jackals ("Karataka" and "Damanaka" in the Sanskrit
Pancatantra) who appear in the first chapter in the fable of "The Lion and the
Ox". The story begins with Dimnah's decision to ingratiate himself to the
Lion, king of the jungle, though Kalilah warns against this. Dimnah ensures
a friendly relationship between the Lion and the Ox, but soon becomes jealous
of the friendship, takes no heed of Kelile's warnings, turns the Lion against the
Ox, and makes him kill his friend. The Lion is left to regret his action. In the
Arabic version the Lion not only regrets and repents but brings Dimnah to
trial, proves his guilt, and condemns him to death. He is moved and
encouraged to do this by the mother-Lion (Toska 1989:45, 57f.). The second
framework story/chapter known as "The Judgement of Dimnah" is generally
accepted as an addition by the first Arabic translator, Ibn-al Muqaffa.
It is important to remember that the 'original' Pancatantra was composed
not as a collection of popular beast fables but as a book of practical knowledge
of the politics of life, a text written for the instruction of the ruling Indian
families on the art of survival above all through intelligent conduct in the
private and public domain (de Blois 1990:15ff). Therefore, it is not difficult
to see why and how the proliferation of translations over the centuries helped
to establish the KD as a classical model for the book of political/moral conduct
(or the "Mirror for Princes"). Proliferation also meant variations. Textual
critics working on more than one version and trying to establish a text, often
find it a nuisance to see the 'purity' of a text contaminated with interpolations.
But it is evident in the case of the KD that at least some of the translators also
served in the development of a highly popular narrative genre with the
variations they introduced in the structure and content of the work. Considering
the problem of the differences presented by various manuscript copies of Ibn-al
text as introductory chapters. In our view, these chapters in the first (known) version of
the KD, must be seen as reflecting (a) one of the earliest perceptions of translation as a
medium for the acquisition and dissemination of knowledge and (b) the exceptional
"visibility" of the translator (Burzoy) in the textual transmission itself.
Muqaffa's (8th century) Arabic, the oldest of which is dated five hundred years
(13th century) after the translation was made, de Blois states:
...the Arabic Kalilah wa Dimnah has to a large degree become a victim of its
own popularity... [E]ditors and copyists felt free to alter the text, to add new
stories and rewrite old ones, to combine material from various manuscripts, and
so on, in a way which would have been unthinkable in the case of a 'serious'
work, say on theology. For Kalilah wa Dimnah was generally (if not always)
considered to be a 'popular' work, a piece of entertainment, which one did not
need to approach respectfully. As a result, we cannot truly say that what we
possess today is Ibn-al Muqaffa's translation but rather a variety of Arabic texts
derived in one way or another from it (1990:3).
However, almost to prove our point, the implications of disrespect for the
'original' in the above argument are expressed in more positive terms in the
following observation on the Neo-Persian translation made (from the Arabic)
by Abu 1-Ma'ali Nasr Allah in the 12th century:
[This] version is a literary tour de force. The translator has stuffed it with
quotations from Arabic and Persian poems, from the Qur'an and Hadith, and so
on, which sounds rather quaint in the mouths of animals in the jungles of India.
In order to fit these quotations into the book, the translator has padded the prose
text, too, to a considerable extent. Nonetheless, the mixture of poetry and prose
makes [this] version stylistically more like the original Sanskrit Pancatantra than
most other versions of Kalilah wa Dimnah (de Blois 1990:5).
Kul Mesud's Kelile ve Dimne4
It is Nasr Allah's version that has served as source text for Kul Mesud's (14th
century) KD, which is among the very first translations into Old Anatolian
Turkish.5 The textual analysis on KD has established that Kul Mesud described
his text as a "translation" ("Türki'ye tercüme olundi"), a "turning into
Turkish" ("Türkce'ye dönderdüm");6 that in translating prose narrative, his
norms led him to a closer adherence to the source text than in rendering verse
4
The detailed descriptive analysis of Kul Mesud's KD, which formed part of the original
paper, had to be omitted for reasons of space in this volume. The full version of the
paper will appear in Turkish translation in the Journal of Turkish Studies, Hasibe
Mazioglu Festschrift, Harvard University.
5
The two existant copies of the translation are in the Süleymaniye Library (Laleli Section
1897), Istanbul and in the Bodleian Library (Marsh 180), Oxford.
6
Cf. Robinson (1991:134ff.) on the conception of translation as "turning".
units; that his consideration for the constraints of the popular poetic tradition
in Turkish as well as those of the Arabic/Persian metre, and his thematic
preferences, reflect a more flexible conception of verse translation; that the
textual designation "Turkish poem" for some verses is curious since it does not
serve to indicate whether the verse units are translations or the translator's own
compositions. While some of these "Turkish poems" are actually based on
those in the source text, if only thematically, others are not (Toska 1989:259).
In rendering the KD in Turkish, Kul Mesud's choice of Turkish
vocabulary in his prose and verse shows he took considerable care to produce
a text that would be accepted as functional by Umur Bey (c. 1309-c. 1347), the
young prince who commissioned it, and by his entourage. Umur Bey, prince
of Aydin on the Aegean coast, came from a line of princes known for their
patronage of the arts and especially of translations of books of canonical status
in Arabic and Persian culture. Including the KD, five translations were
dedicated to Umur Bey, Mehmed Bey, and Isa Bey (Toska 1989:236, 237). It
is also known that in neighbouring pre-Ottoman principalities such as the
Germiyan in the Kütahya region, north-east of Aydin, there was a similar
tradition of patronage. Zajacskowski has drawn attention to the cultural
interaction between the Aydin and Germiyan ruling nobility, remarking that the
"translation of the KD for the court of Aydin might have led to the translation
of another 'Book of Advice', the Marzuban-name, for Süleyman Shah" prince
of Germiyan and son-in-law of Umur Bey (Toska 1989:241). Like the KD, to
which it contains references, the Marzuban-name is a collection of beast fables,
translated by the poet Şeyhoglu from the Persian.
The linguistic constraint that such books of political/moral advice should
be read in Turkish not Persian, has ideological implications, and the patronage
of ruling kings and princes throws light on the linguistic expectations
underlying translations and on the link between language and ideology. The
pre-Ottoman principalities in western and central Anatolia were conscious of
their Turkoman identity and had political reasons for placing emphasis on
Turkish rather than Persian which flourished as the medium of literary writing
and the official language of the Seljuk Sultanate, then the most powerful state
in Anatolia.
We note the decree issued by Mehmed Bey, the ruling prince of Karaman,
when he set up his principality in Konya in 1277, whereby Turkish was to
replace Persian and Arabic in all forms of communication (Silay 1994:15).
What is primarily important about the Karaman prince's decree of the 13th
century is that in authorizing the use of Turkish, particularly of prose, in
official communication, it must have also provided impetus for translation
activity of a much broader scope, which evidently found its way into other
principalities. A translator's dedication from the 15th century shows that royal
patrons still demanded translations that were accessible in Turkish. Mercümek
Ahmed, who translated the Kabus-name, another "mirror for princes", by King
Kay Ka'us ibn Iskender, explains that it was in response to the following
words by the Ottoman Sultan Murad II (1421-1451) that he translated the book:
"This is a very good book [with]...much advice and useful information in it.
However it is in Persian. Someone rendered it into Turkish but it is not clear.
He did not translate it into plain language" (Silay 1994:15f.).
Linguistic constraints operating in a certain period are no doubt important
in the study of literary translations but they cannot be considered independently
of poetic/stylistic norms or expectations dominant in the same period. To
examine both, a descriptive study of the rewrites in the 14th century will have
to move beyond the sphere of the individual translator to look for links
between his work/s and those of other poets/translators. This point seems to
be particularly relevant in placing Kul Mesud's translation in a certain context.
Nothing is known about Kul Mesud apart from the fact that he completed
the translation of KD for the prince of Aydin before 1334 (Toska 1989:237).
His name does not appear in any of the known biographies. However, another
Mesud, a well-known poet designated as Hoca (Master) Mesud, is known to
have written Süheyl ü Nevbahar (1350), one of the first "mesnevi" romances
in Turkish, and the Ferhengname-i Sadi (1354) a "book of advice" derived
from the Bostan by Sadi, the Persian poet. Both of these works have been
considered translations, although no source text has yet been established for
Süheyl ü Nevbahar which could, after all, have been a pseudo-translation.
A descriptive study on the works of Kul Mesud and Hoca Mesud primarily
as translators could reveal hitherto unnoticed affinities between their texts. In
this group could be included the verse translation of the AD, (known as the
Gotha manuscript) that dates from the 15th century, dedicated to Sultan Murad
I (1359-1389) by an anonymous translator who states his work was based on
a previous prose version, which H. Ethé established as Kul Mesud's KD
(Toska 1989:19f.). In view of the extensive studies on the Near Eastern
transmission and translations of the KD which any modern bibliography can
reveal, it seems necessary also to examine subsequent Ottoman-Turkish
versions which call for research within the field of modern translation studies.
The most important of these is the 16th century Hümayun-name by Ali b. Salih
Çelebi who translated it from Husayn Vaiz Kashifi's (d.1505) well-known
Anwar i Suhayli, having spent twenty years improving on the Persian source
text before finally dedicating it to Sultan Süleyman the Magnificent. As Toska
noted with reference to V. Chauvin's study (1897) on the subject, the

Hümayun-name was regarded as a masterpiece of prose and translated into
French and other European languages in the 18th century, copied widely in
manuscript from the 16th century onwards and printed in Egypt and Istanbul
in the 19th century (Toska 1989:32). Diachronie studies on the Hümayun-name
corpus including the Marsh 61 manuscript (c.l5th/16century) and Ahmed
Midhat's version (1887), would help to explore the shifts in linguistic and
stylistic choices and the poetics of translation underlying them.
Comparing the early versions of the KD with the 14th/15th century
rewrites of other "mirrors for princes" (e.g. Camasb-name, Marzuban-name,
Kabus-name) would help establish the norms regarding the translation of
didactic narrative in the "books of advice" and formulate an early poetics of
translation. A point not to be overlooked is that some of the translators were
also well-known poets who had established themselves as such by means of
their works in Persian. Study of the nature of such 'original' works, their
relationship with the rewrites by the same poet, would also help in establishing
the poet/translator's characteristic norms, thematic/stylistic preferences, and
general conception of literature.
Projects centering on systemic, or more specifically, on polysystemic
analyses, have wider implications, especially in regard to the conception of
literary "imitation". They are particularly useful not only in investigating
aspects of Ottoman dependency on Persian but in discovering deviations from
sources in spiritual/ideological/cultural attitudes reflected in rewrites. They can
also draw our attention to the differences/oppositions in and between the
Arabic, Persian and Ottoman systems within the seemingly unifying umbrella
of the Islamic "system of systems" i.e. polysystem.7
Lefevere, in a unique and challenging discussion which draws a parallel
between the "Islamic system" with the European in the context of translation
studies, states that the "Islamic system demonstrates the futility of any attempt
at confining literature to a given language, even though it may be convenient
to refer to particular systems in this way. Rather, the real boundaries of
literary systems tend to be drawn by their common ideology, often extended
through conquest or imposed by authority... "; that the dominant poetics of the
Islamic system was developed in Arabic, a Semitic language and that the
7
Lefevere (1992:11) quotes Steiner on the "complex system of systems" without reference
to the "polysystem theory". Lambert's (1995) detailed, and timely, critical survey (which
also provides an excellent bibliography) is extremely useful in drawing attention to the
discussion on polysystemic studies and their importance for research.
fundamental components were taken over by Persian, Turkish and Urdu (each
belonging to a different language group), but not fully adapted ("not 'bent' to
'suit' each language") (1992:31; see also Silay 1994:34). Quoting Bombaci,
Lefevere points out that when Turkish "adapted itself to Arabic-Persian
metrical forms, it did violence to its own nature, since it is a language unsuited
to quantitative meters" (1992:31). While this argument can be accepted in
theoretical terms, it is certainly worth investigating how Ottoman poetry
developed and survived for over five hundred years in a language that "did
violence to its own nature". A question of considerable importance, it underlies
the problem of "literary diglossia" also mentioned by Lefevere (1992:17), and
recognized by all literary scholars; for it was not only the "aruz" metre and the
various genres that were adopted from the Persian system but also the
vocabulary conveying poetic images (see Silay 1994:31). Kul Mesud's
translations of Persian and Arabic verse are fairly simple examples of the ways
that could be found for adapting Turkish vocabulary to the metric requirements
of "aruz" in the 14th century. However limited it may have been, the
"movement for plain Turkish" ("Turki-i Basit") in the 15th century was an
attempt to avoid Persian/Arabic vocabulary in poetry (Silay 1994:16-19). To
what extent rewrites were prominent in this "movement" is a question that
should be looked into carefully, as the major area of investigation into the
process of lexical appropriations, primarily from Persian but also from Arabic,
is quite evidently that of rewrites.
Silay provides us with a highly illuminating modern experiment in
translating a Persian poem into Ottoman, as an example of "the easy
transaction between the Arabic, Persian and Ottoman poetic traditions"...
showing, "with minimal syntactic changes" ...the "possibility and
appropriateness of a direct and easy translation from Persian into Ottoman,
since the crucial element, the vocabulary is stunningly similar" (1994:33). But
should not the real focus of attention be on the existing translations of which
there seems to be such a large corpus, rather than probable versions? Silay
himself states that "While the current state of scholarly understanding of
Ottoman poetry is hardly adequate, on the basis of what has been done, we
know that the earliest works of medieval Ottoman poetry [13th-15th centuries]
are direct imitations of Iranian models" (1994:32). What is meant by
"imitation" in the Ottoman tradition, how it was perceived by the poet-
rewriters and early historians, the notion of"direct"ness, the nature and degree
of linguistic and literary appropriation, deviations from sources, all await
analytical description in order to help clarify the notion of literary dependency
on the Persian/Arabic systems.
Holbrook, in her outstanding work of modem scholarship on Ottoman

poetry, debates the question of originality in the romance Hüsn ü Aşk, an 18th
century masterpiece by §eyh Galib:
Modern criticism of Ottoman romances has concentrated on determining the
textual sources of works, paying less attention to their intrinsic qualities, and by
this preoccupation and its omissions critics have created the impression that
Ottoman poetry was more imitative of precedent, therefore (by 19th century
Romantic standards) less original than Western poetry. Explicitly stated, such a
conclusion seems unlikely if not absurd, yet we are bequeathed the assumption
that borrowing plots — from the Persian — was standard practice for Ottoman
romance writers. Such an assumption would be hard to prove. A great many
romances were written during the Ottoman centuries, and they tend to be long
and available only as unedited manuscripts. Critical analyses are rare, and many
seem to rely upon plot summaries or opinion taken from other secondhand
sources. A supportable general conclusion would rest upon study of the precise
nature of relationships with 'sources' just as numerous and long, and of how
these relationships were understood by individual authors.. Without a genealogy
of the idea, evaluation of one author's originality, let alone that of an entire
literary tradition, can be made only on the basis of current concerns, not in view
of standards according to which works were produced" (1994:76f.; emphasis
added;cf.Lefevere 1992:32).
From the perspective of translation studies, it is interesting that Holbrook
should recognize the need for "study of the precise nature of relationships with
'sources'" (i.e. source-text and target-text relationships) without actually
mentioning its relevance to the practice of translation or rewriting, - a point
which becomes even more conspicuous in the context of her following
argument: "Western [ Old Anatolian] Turkish poetry developed in the
fourteenth century by appropriating to heretofore largely oral usage Persianate
models of literacy... If this development can be called imitation, it is one of
a special kind, a matter of transition from orality to literacy rather than
imitation of one literary tradition by another" (Holbrook 1994:77). Here a
notion of "translation" is implicit in the verbal transfer from the oral to the
written, but in a sense that reduces the concept of "imitation" to an abstraction.
While the appropriation of the oral by the written is a significant
dimension to be taken into account, this cannot be a primary concern, given
the large body of rewrites produced as a result of the most conspicuous activity
in that century. It will be much safer to search for the roots of any Ottoman
concept/s of imitation in the concrete relationships between Persian sources and
actual translations/rewrites, in what was omitted from and/or added to those
texts, in the way poets/rewriters did or did not formulate their perception of
the literary activity they were engaged in, and in the way their contemporaries,
biographers, or later historians perceived, defined or concealed it.
References
De Blois, Francois. 1990. Burzoy's Voyage to India and the Origin of the Book of Kalilah wa
Dimnah. Prize Publication Fund Vol.XXIII. Royal Asiatic Society.
Grube, Ernst, J. 1991. (ed.) A Mirror for Princes from India. Bombay: Marg Publications.
Holbrook, Victoria Rowe. 1994. The Unreadable Shores of Love. Turkish Modernity and
Mystic Romance. Austin: University of Texas Press.
Lambert, José. 1995. "Translation, Systems and Research: The Contribution of Polysystem
Studies to Translation Studies". In: Yves Gambier (ed.), Orientations Europeennes en
Traductologie.TTR. VIII (1), 105-152.
Lefevere, André. 1992. Translation, Rewriting and the Manipulation of Literary Fame.
Paker, Saliha. 1991. "The Age of Translation and Adaptation 1850-1914. Turkey". In: R.
Ostle (ed.) Modern Literature in the Near and Middle East 1850-1970. London/New
York: Routledge. 17-32.
Paker, Saliha. Forthcoming. "The Turkish Tradition". In: Mona Baker (ed.). Encyclopedia of
Translation Studies. London & New York: Routledge.
Robinson, Douglas. 1991. The Translator's Turn. Baltimore/London: John Hopkins University
Press.
Silay, Kemal. 1994. Nedim and the Poetics of the Ottoman Court. Medieval Inheritance and
the Need for Change. Bloomington, Indiana: Indiana University Turkish Studies
Series: 13.
Toska, Zehra. 1989. Türk Edebiyatinda Kelile ve Dimne Çevirileri ve Kul Mesud'un Çevirisi.
Vol.1: Textual Analysis. Vol II: Modern Turkish Transcription of the Süleymaniye
Library Manuscript of Kul Mesud's Translation. Unpublished Ph.D. thesis. Forthcoming:
Department of Near Eastern Languages and Civilizations, Harvard University.
Toska, Zehra. 1991. "Kelile ve Dimne" nin Türkçe Çevirileri". In: G.Kut and G.A.Tekin
(eds.) Journal of Turkish Studies, 15, Fahir Iz Festschrift II, Harvard University, 355-
380.
Toury, Gideon. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia:
John Benjamins.
Translating plays or baking apple pies: A functional
approach to the study of drama translation
Sirkku Aaltonen
In my paper, I propose to outline a model for the analysis of the translation of

foreign drama for the stage. I am interested in the relationship between the
source text and its translation, in particular when and how the translation is
allowed to manipulate its source text, and I shall use the terms
"transformation", "intersection", and "borrowing" to describe the different
forms of manipulation. According to Andrew (1984:98ff.) translation
transforms its source texts when it follows "the letter of the source text" in that
it uses the same dramatic structures or the style of presentation. It intersects
the source texts when it foregrounds a particular aspect of it by changing the
order of the scenes, some of the characters or the setting. A translation may
also use the source text by borrowing a central idea, topic or theme in order
to weave a new play round it. I am interested in finding characteristics of the
receiving theatrical system which explain the use of these methods. It has for
long been clear that the linguistic system alone cannot sufficiently explain what
goes on in drama tanslation, but it has been far from clear what factors ought
to be included in the study.
My paper incorporates insights from theatre semiology into the study of
translated drama by applying a functional approach which was originally not
outlined for written artifacts at all. The approach is based on the assumption
that drama translation is a purposeful activity which involves advance planning.
The model which I am applying to the study of translated drama was originally
devised by Victor Papanek in 1970 for product design, and it views
functionality as a wide concept which includes many more aspects than mere
suitability for a particular purpose.
Papanek's model
All people are designers, wrote Victor Papanek, and maintained that design
and planning underlie all human activity (1973:21). Whenever a particular goal
needs to be achieved, some design and planning become essential. This applies
to a whole range of different tasks, such as the writing of a novel, executing
a wall painting, and composing a concerto. It is also needed in the clearing and
cleaning of a drawer, extracting a decaying tooth, baking an apple pie and
raising children. Naturally design is involved in finding a suitable form for a
bridge or a bottle of vitamin pills. Papanek defined functionality in such a way
that it takes into account both aesthetic considerations and associations.
According to him (1973:25), although a product is designed to serve a
particular need, it also reflects the social and historical circumstances of its
creation and results from a particular combination of material and method of
working. Functionality, as Papanek understood it, has the dynamic dimensions
and inter-relationships of Method, Need, Telésis, Association, Aesthetics, and
Use. In a well-designed product, all these aspects are in good balance.
A translated theatre text, like any theatre text, can be analysed and studied
through these different dimensions of functionality. All these dimensions
influence the shape a translation takes when it is created, only the emphasis
given to different aspects varies. The different dimensions may overlap in the
translation work, but they can still be used to explain certain regularities and
features of translated play texts. In what follows, I shall apply the various
dimensions of functionality, as described by Papanek's (1973: 25-39), to the
analysis of drama translations.
Method
One of the dimensions of a translated play text is the Method which grows out
of the combination of the material which is available and the circumstances
where the translation work takes place. It involves the various relationships
between the source text, the translator, and the translation. One perspective the
Method offers for the study of drama translations is whether the receiving
polysystem perceives the translator as a mediator or creator, which is related
to a number of characteristics of the theatrical polysystem.
Sirkku Aaltonen 91
Centrality of peripherality of translated drama
The translator's role in the theatrical system as creator or mediator is related

to the centrality or peripherality of translated drama in the receiving theatrical
polysystem. When translated drama occupies a central position in the domestic
theatrical system — for example when the system is still very young and
enough domestic material is not yet available — the distinction between a
source text and its translation is not considered as important as when translated
drama is peripheral to the system (Even-Zohar 1990:46). This was the case,
for example, in the 19th century Finnish theatre when foreign plays were
cannibalised into the Finnish polysystem through total acculturation. It was
important that there were play texts available and that they were in Finnish.
The mode of translation was borrowing and often only the central idea of a
source text was used and a new play woven around it. Foreign plays were
turned into new Finnish plays, and for example Macbeth and Erasmus
Montanus were rewritten as Finnish plays.
The copyrights
The translator's role as mediator or creator depends also on the question of

copyrights and how they are monitored. The copyright law outlines, sometimes
vaguely, sometimes more strictly, the limits of the manipulation in the
translation of works whose copyrights are still controlled by the playwrights
or their descendants. For these play texts the mode of translation is likely to
be transformation or intersection. If the patrons so wish, the law can be
enforced and a translation rejected. This happened in Finland in 1986, when
the translation of Lloyd Webber's Cats was not approved by the T.S. Eliot
Society in London. They had the translation withdrawn from the rehearsals in
the theatre. The patrons claimed that it had "violated the spirit" of the original.
The translation had to be revised before it was allowed on stage again.
Play texts whose copyrights have expired are more likely to be rewritten
in translation, and Shakespeare's plays for example are more likely to be
recreated afresh than those by some contemporary playwright.
A translator's status in the polysystem
The outcome of a translation process is also influenced by whoever has

completed the translation, as not all translators are granted equal power by the
theatrical system. In Finland, there is a two-tier system of translators in which
a translator's role as mediator or creator is linked with the power relations in

the receiving polysystem. Conventions have developed which determine who
is allowed to manipulate certain structures of the play and in what
circumstances. The first category of translators are those whose only
connection with the stage is the translation work. They are fairly powerless and
their relationship to the dramatic text is comparable to that of an actor. The
text sets the parameters of the work, and both the translator and the actor must
bow to the text. Their role is seen as that of mediators rather than of creators.
The second category are translators who work within the theatre, such as
dramaturges or directors. They exercise more power and retain this power
when they work as translators. As translators they are closer to being creators
than mediators. They can, if they so wish, make adjustments or interpret the
text according tp need.
The medium of transmission
The variation between creation and mediation tends to be linked with the
medium of transmission in that borrowing or intersection are not acceptable
when a play text is published in printed form in the literary system, whereas
the use on stage, on TV or in film makes manipulation possible.
Translation and its source
A particular Method may also grow out of the perception of the relationship
between a translation and its source text. The first steps of the Finnish national
theatre and drama translation offer several examples. One of the Finnish
theatre historians (Tiusanen 1969:59f.) mentions a play which was first
translated from German into French and then back into German, at which stage
the text had dropped the name of its German playwright from the credits.
French drama was translated into Swedish through German, and the same play
could arrive in Finland in different forms by different routes. This is still
possible, but in contemporary theatre both transformations and intersections are
credited to the playwright of the source text. For example, the Irish playwright
Sean O'Casey wrote his play Purple Dust in 1939-40, and it was translated
first into Finnish from English in 1966. In 1974, however, a new translation,
an intersection, of the play arrived which was based on a German translation
commissioned by the Berliner Ensemble. Both are regarded as translations of
the play which O'Casey wrote in the 1940s.
Sirkku Aaltonen 93
Use (Does it work?)
When theatre practitioners are asked to describe a "good" theatre text, their
immediate reaction is usually that "it must work". What they actually mean by
this is more diffiicult to establish. In my view, the use of a play text can be
viewed from two perspectives of which one is the compatibility of the play
with socio-cultural, generic and theatrical conventions and the other is the level
of the individual stages.
In order to be accepted by both the theatre practitioners and the audience,
texts can be expected to match the circumstances of their use in observing the
dramatic and performance conventions of the receiving system as well as the
competence of their audience of cultural, behavioural or ideological
conventions. In order to understand what is going on on stage, the audience
needs to be able to decode, if not all, at least a sufficient minimum of the signs
and sign systems within the text. In consequence, adjustments may be made in
the translation process in relation to the general cultural conventions covering
the language, manners, moral standards, rituals, tastes, ideologies, sense of
humour, superstitions, religious beliefs etc. There are also the specific
dramatic and performance conventions of a culture, society or Subculture as
well as the conventions of a specific dramatic Medium (TV, stage, radio) or
sub-genre (soap opera, comedy) which are followed.
The plays also need to meet the requirements of a particular theatre. The
fact that a play text is chosen for the repertoire results partly from its
suitability for the economic and human resources of the theatre, compatibility
with the repertoire and the assumed relevance for contemporary audience.
Adjustments may be necessary if only one setting is financially and technically
possible, if the number of actors is not sufficient for the play or if the
practitioners assume changes in the audience's reception of the play.
Need
The third of Papanek's functional dimensions is the Need for a particular

product or activity. Why do theatre texts, and some theatre texts rather than
others, get translated at all? Are they translated for the literary or the theatrical
polysystem? When Finland was in the process of receiving its first national
theatre, foreign drama was cannibalised into the indigenous system for several
reasons. Firstly, the advocates of a new theatre needed to prove that there was
a sufficient number of texts available for a national theatre. Secondly,
translation offered an opportunity to exercise the Finnish language and to
enrich its vocabulary with new expressions. At the time of the censorship, play
texts were less strictly monitored than literary works (an Act of censorship was
passed in 1850). The number of translations was therefore relatively high; they
were usually done by theatre enthusiasts who were struggling to find a balance
between ideological considerations and those concerning the poetics of drama.
Another view of the Need for translated drama which has affected the
outcome of translation work is offered by Susan Bassnett (1990:79) who has
distinguished between two distinct forms of drama translation which developed
in Europe by the nineteenth century. One was commercial translation, for
which the eventual performance was crucial, and the other was the aesthetic
translation of classical texts for the reader. The same division of dramatic texts
into those for theatrical use only and others available for the general readership
as well may still be found in contemporary Finland. Only costly rewritings of
the canon but very few modern playscripts are published in printed form.
Modern drama which does not interest commercial literary publishers may see
the light of day in the publications of the theatrical system, for example, as
supplements to the only national theatre periodical. The majority of both
domestic and foreign drama only exists as playscripts. These, typed and in A4
format, are available on request from the central library run by the Finnish
theatre union in Helsinki. They are hardly ever read by anyone other than
theatre practitioners (or academics for research purposes!).
Telésis
The Telésis of a translated text ties it to the socio-cultural circumstances of its

conception. Although the text will always mean different things to different
individuals and a multitude of meanings will always arise from the interaction
between the content of the signs it emits and the spectator's competence to
decode them, it all still happens in particular social and historical
circumstances. When John Millington Synge wrote The Playboy of the Western
World, it gave rise to riots in Dublin. It could never have the same impact
again in another time or another culture. The further the author recedes in
time, the less relevant become the original meanings, and the more different
the "message". The great advantage of stage drama lies in the fact that each
translation and performance can take the particular cultural, social, historical
and geographical situation of its audience into account and adapt the play to
these changing circumstances.
Sirkku Aaltonen 95
Associations
We have been conditioned psychologically from our very early childhood

onwards to make associations, and this ability is also effectively used in theatre
texts. It helps us to make sense of works of art which are new to us and in this
way integrate them into the receiving systems.
Martin Esslin (1994:109ff.) talks about signifiers, key or clef signs, which
operate over a long period of time and determine how other sign systems
within a given section or passage of a work are to be read. The opening words
of a play set the key in which the whole play, or at least that scene is to be
taken by the audience (for instance in a poetic language or verse).
Meanings also arise from recognised intertextuality, that is, references to,
parallels with or variations on, other drama or dramatic structures presumed
to be generally known in a particular milieu. In present-day popular drama this
recognition is one of the basic sources for an audience understanding and
enjoyment of the many series and serials on TV. The immense popularity of
drama is based at least partly on the very strongly established patterns of
structure with additional meanings arrived at by varying this pattern to some
extent in each episode.
On the level of language, associations are produced in a similar manner
and they may link theatre texts not only to each other and to the outside reality
of the audience. When Tony Roper's play The Steamie was translated into
Finnish in 1994, the translation used a phonological representation of the way
spoken Finnish runs words together, drops endings etc. This could be
interpreted by the actors in different ways, primarily as indications of the
social class of the dramatic figures. The English play created a representation
of a strong Glaswegian accent. When the protagonist Gar in the Finnish
translation of Philadelphia Here I Come (by Brian Friel) occasionally speaks
recognisable Helsinki slang, he is likely to be interpreted in the Finnish context
as trying to be tough.
Associations serve their most important function in the ways in which a
text is brought into contact with other texts. According to Jonathan Culler
(1975:140-152), the simplest level of this naturalisation process is the socially
given text which seems to derive directly from the structure of the world. In
drama translation the implications here are to the degree of acculturation in the
setting and milieu of the play.The second level of naturalisation is the cultural
stereotype or accepted knowledge which the culture itself recognises as
generalisations. Italians are noisy, economists and book-keepers are boring,
women may be witches, and stepmothers are wicked. These are examples of
stereotypes which are recognisable within our Western culture. The third type
of naturalisation is made up of conventions of the genre. Fourthly, there may
be either an implicit or explicit indication that what we are receiving is not a
generic convention.The audience may be drawn into the play, either concretely
in a performance, or when it is being addressed directly by dramatic figures.
The fifth and last category of naturalisation is the complex claim of reality
related to recognisable intertextuality. A play or a film borrows an idea and
uses another known text as its material. Medieval mystery plays draw on Bible
narratives, and Kurosawa's film The Throne of Blood on Macbeth,
Aesthetics
Aesthetics become dominant in the debates about some formal structures of

drama which may be manipulated to meet the dramatic conventions of the
receiving system or cause a play's rejection if the receiving system does not
accept the suggested innovation.
A great many debates have dealt with the aspects of whether the
translations should follow the poetics of their source texts in the distinction
between prose or verse. Aesthetics are also concerned in the changes in the
style of presentation. When O'Casey's Purple Dust was translated into Finnish
in 1974, it changed the structure of the 1966 translation in the direction of epic
theatre. Some of the dialogue was turned into songs, and the structure was
simplified by a more subtle characterisation of the dramatic figures. Aesthetic
considerations are involved in the decisions about the use of accents and
dialogues as well as taboo words.
To conclude, the different dimensions of functionality interact and are
interrelated. A particular Method may be linked with a certain Need, and a
given Use may be a consequence of a specific Telésis. Aesthetics and Need
may be effective simultaneously. I hope I have managed to illustrate how
different aspects of functionality may have an impact on the way theatre texts
get translated. Translated drama is a vast area and will provide interesting
insights into the power structures of a country's theatrical and cultural system.
It therefore deserves more attention than it is getting at the moment. I hope my
paper has also been able to answer the question about the similarities between
the activities of translating drama and baking apple pies. Both operations
require planning, and they have to be carried out in a purposeful way. They
involve considerations concerning the Method, Use, Need, Telesis,
Associations and Aesthetics which get different emphasis at different times.
Sirkku Aaltonen 97
References
Andrew, Dudley. 1984. Concepts in Film Theory. Oxford: University Press.

Bassnett, Susan. 1990. "Translating for the Theatre". Essays in Poetics 15, 71-84.
Culler, Jonathan. 1975. Structuralist Poetics. London: Routledge.
Esslin, Martin. 1994. The Field of Drama. London: Methuen.
Even-Zohar, Itamar. 1990. Polysystem Studies. Poetics today 11:1 (Spring).
Papanek, Viktor. 1973. Turhaa vai tarpeellista? (orig. Das Papanek Konzept 1970, transi.
Jyrki Saarikivi and Paula Leistén). Helsinki: Kirjayhtymä.
Tiusanen, Timo. 1969. Teatterimme hahmottuu. Helsinki: Kirjayhtymä.
Translation strategies and the reception of drama
performances: a mutual influence
Marta Mateo
A drama performance would not exist without an audience, and a translation

depends — for its success and indeed for its very existence as a translation —
on the interests and cultural assumptions of the receiving system. The receptor
then plays a major role in both activities and has also gradually made its way
towards the focus of attention of both Translation Studies and theatre semiotics.
One reason for this important presence of the receptor in translation and
in drama is their interactive nature: translations act as a form of intercultural
communication, making what is alien to a culture come into contact with what
is peculiar to it. They imply an appropriation of a source text by a target
culture and, since it is generally the receiving system that initiates the cultural
contact (Heylen 1993:22), a translator's decisions will be largely determined
by the translation and cultural norms prevalent in the target polysystem. As
regards drama, "[c]ultural assumptions affect performances and performances
rewrite cultural assumptions" (Bennett 1990:2). A theatre audience brings to
the performance a horizon of ideological and cultural expectations which will
interact with the theatrical event (Bennett 1990:108): the playwright and the
director will respectively shape their text and production to provoke a certain
response in an audience, and the experience of the production will in turn
establish or revise the spectator's horizon of expectations.
The crucial role that the receptor plays in translation and in performed
drama brings about a further similarity between these fields: as they are clearly
teleological actitivies — i.e. both the translator's and the producer's decisions
for their product are taken with a particular objective and receptor in mind -,
in neither case can we talk of a "right", "final" or "canonical" interpretation.
While there is only one source text, there is no limit to the number of target
texts that can be drawn from it, as each culture appropriates a foreign text
differently. Similarly, one single drama text may be performed in many
different ways, since each production realizes what each producer can or wants
to read and see in the text.
The reception of translated texts in general and that of source language
performances have therefore some interesting points in common. In this paper
we shall study the factors which make up the situation in which both activities
interact, i.e. translated drama. But we shall first have to look at the elements
which shape audience response in theatre performances, both in a target and
in a source language context.
As opposed to other forms of presentation, a stage performance is live,
and this has some important consequences for the communication process
involved, which Törnqvist (1991:13) has summarised as the following:
1. A stage performance involves a two-way communication process: the actors
may respond to reactions from the audience.
2. A theatre visit is a social event: the reactions are those of a mass audience.
3. Every stage performance is unique, unrepeatable.
4. A stage performance is determined by the spatial facilities available.
5. The spatial limitation necessitates abundant use of proxemics and the
distance from stage to auditorium requires special kinesics and para-
linguistics.
6. A stage performance has a plurimedial and unrepeatable nature and it is
therefore difficult to notate.
These characteristics of stage performances affect the three types of interactive
relations that are established with the audience in a theatre: "audience-stage
interaction in the field of fiction, audience-actor interaction, and interaction in
the audience" (Passow 1981, in: Bennett 1990:162).
The audience's role in the theatre is a very active one. While in other
modes of presentation — such as a printed text, a film or a television program-
me — all that the reader/spectator can do is read, listen or watch since the
communication is one-way, the theatre audience interferes with what is being
presented on stage to such an extent as to determine the success or failure of
a production on the very night of the performance. Mackintosh (1993:2) talks
of the "sense of danger" that accompanies the shared experience of a drama
performance: "[a]nything might happen" and that distinguishes live theatre
from eg. the experience of cinema.
Audience-stage interaction is established according to the spectators'
cultural assumptions, horizon of expectations and theatrical conventions on the
one hand, and the direct experience of a production with its own internal
horizon of expectations, on the other (Bennett 1990:180). The spectators'
response to the fiction presented on stage will depend on multifold factors and
is already shaped before the performance proper begins: things such as the
Marta Mateo 101
play's title, familiarity with the drama text or with the playwright, the amount
of information that can be read in the programme before the performance
actually starts, the price of the ticket or such practical questions as how the
spectator travelled to the theatre or what day of the week it is, determine the
audience's reaction together with some more crucial factors such as performan-
ce time or, more importantly, performance place, and the verbal and non-
verbal signs on the stage. The spectators come to the show with some
preconceived ideas about it, and when the performance starts and proceeds they
test those hypotheses against the fictional world that is being offered to them.
This is done within the time constraints of the performance, which imply that,
unlike the reader, the spectator will have to ckeck or confirm his/her
expectations as the actors perform: their reception will be marked by a sense
of temporariness, since they cannot go back to a previous stage of the play.
The audience also interacts with the actors. A "good" performance from
the actors will — hopefully — provoke a positive attitude on the audience's
part and an appreciative audience will in turn encourage the actors. This
interaction may also be affected by other factors such as the area designated
for the accommodation of the audience — the percentage of seats occupied will
affect the quality of the performance (Bennett 1990:140) — and the audience's
familiarity with the actors, since the spectators will then have some expecta-
tions about them and will not only be aware of the actor's fictional part but
also of his/her actual work during the performance. We may sometimes extend
this interaction to the director and even to the playwright: the potential
audience of a performance will exert a feedback effect on the text and on the
production when they are being prepared, since both the playwright and the
director will have an audience in mind when creating their work; this effect
sometimes comes from the real audience, which provokes changes in the
director's decisions or in the text itself after a preview or even during a run.
As regards interaction in the audience, we must bear in mind that the
audience forms a group as a whole and that individual spectators normally
expect confirmation of their reaction to the play from the other spectators'
responses and will usually suppress a reaction which goes against the general
trend (Bennett 1990:164). Group responses will also be affected by the seating
area and the number of seats occupied, since "[t]he experience of the spectator
in a packed auditorium is different from that of one in a half-empty theatre"
(Bennett 1990:140), as well as by the fact that some responses such as laughter
and applause are very infectious. Meyerhold tried to develop a code for the
notation of audience reaction and, among the common responses, he noted the
following: silence, noise, loud noise, collective reading, singing, coughing,
knocks, scuffling, exclamation, weeping, laughter, sighs, action and animation,
applause, whistling, catcalls, hisses, people leaving, people getting out of their
seats, throwing of objects and people getting onto stage (Stourac and McCreery
1986:20, in Bennett 1990:7).
The success of a drama performance depends on the skills of people
working on very different areas — play-writing, directing, stage-design,
acting, lighting, producing, music and sound, make-up, theatre management,
etc. — so that we may clearly state that performed drama is, with opera, a
particularly complex and plurimedial activity.
So far we have dealt with the reception of drama in general. We shall now
turn to the study of the implications that this may have for the transposition of
a source play to a target context. Since the text is only one of the elements of
the performance, the translator of plays for performance will have to base
his/her translation decisions on a very complex set of factors: the literary and
cultural dimension of the text together with the semiotic intricacy of the
production and the social characteristics of the target audience. We shall here
look at some of the extra-textual features that determine the reception of a
target text in performance and that therefore may play a key role in some of
the decisions taken during the translation process of a source play: the channel,
the theatre-building and the use of the stage.
The channel
The translation of plays sometimes entails a transposition from one medium to

another, since the mode of presentation of the target text may be different from
that of the source text. An English play may be turned into a Spanish TV
version and this will have some important consequences in the translation
process, as regards not only audience reception but also the potentialities
offered by the means of communication.
We may distinguish between reading a drama text, listening to a radio
play, and watching a TV version or a cinema film. The actual impact of the
text on the receptor differs considerably in the various modes of presentation:
a written text may be absorbed at leisure by a reader, while the impact of a
stage performance or a radio/TV/film version is immediate. This will bear on
the choice of textual material for the target text, since the translator for live
versions will have to select the verbal signs that match the actors' body-
language so that there is nothing to distract the receptor's attention from what
is in focus. The focus of attention marks another difference between the
Marta Mateo 103
channels: while the cinema director enjoys the power afforded him by the
editing camera, which enables him to select characters or things for the picture
at any moment, thus guiding the receptor's attention, "[o]n the stage one of the
challenges that faces a director, a writer and the actors is how to focus the
attention of the audience, how to bend the focus... when there are so many
other things to look at on the stage" (Pinter in: Törnqvist 1991:145). The
reading mode of reception coincides here with the listening mode: in both,
attention is usually just paid to speaking characters, while silent characters are
normally disregarded by readers and radio listeners. In a performance, how-
ever, the audience is always aware of all the characters on the stage, so that
their attention is not necessarily centred exclusively on the speaking characters.
An interesting example of the different effect that silent characters exert
on the audience in the various forms of presentation is provided by the opening
scene of Willy Russell's play Shirley Valentine (1986), which was made into
a film in 1989. While, on the stage, the audience immediately realizes that the
actress is talking to a wall, in the screen version, since this is something we
do not expect, we assume that there is someone else in the room whom Shirley
is talking to and who is not in picture at the moment. It is only after we have
heard Shirley say "Wall!" several times that we realize the wall is her
interlocutor, something we are aware of in the theatre from the moment the
curtain opens and she starts to speak. The translator may feel the need to give
a silent character from the source play some lines to speak in e.g. a target
radio version so as to make his/her presence felt. The change in the origin or
nature of some lines is sometimes due both to a new medium and to different
cultural conventions: in a 1967 Spanish TV version of Sheridan's The School
for Scandal, the characteristic 18th-century asides were either deleted or turned
into lines addressed to some other character in the play, in order to make them
fit in with the realism that the TV medium generally leans towards.
Stage directions are usually subject to some type of transformation in the
translating process, particularly if this is accompanied by a change in the
medium. The written stage direction of the drama text is usually converted into
props and into kinesic, proxemic or other non-verbal signs on the stage, but
it may sometimes be turned into an acting direction (an indication for the
actors) or even into a verbal sign to be uttered by a character — as in
Elizabethan plays. This may simply be due to directorial preferences but it is
often grounded on the facilities provided by the medium chosen for the target
text. As Törnqvist (1991:2) suggests, the polysemic range of the written text
is in a sense much wider than any performed version of it since we may
imagine various ways of realising a speech or a stage direction when we read
it, whereas we only receive the director's and the actors' interpretation of the
playwright's instructions when we watch or listen to it. The director's and the
translator's decisions as regards non-verbal signs are frequently mediated by
the channel. An example is provided again by the 1967 Spanish TV version of
The School for Scandal, in which long and descriptive stage directions have
been inserted for the settings so as to help to visualize the play. Special
emphasis has also been given to the actors' movements and gestures, all of
which turns it into a very visual and performable version, although it has also
become rather inconsistent from the plot's point of view.
The emphasis on the visual component is particularly characteristic of
cinema versions, which tend to show more of the environments in which plays
take place and frequently transpose parts of the dialogue into visual images. A
very important difference between stage performances and screen versions in
general is that the former tend to show fewer settings and make use of
continuous space — we remain practically within the same locale in each
act/scene-, while screen versions present discontinuous space, showing a great
variety of settings, particularly in the case of cinema (Törnqvist 1991:19). A
case in point is Shirley Valentine, a two-set play in which the action takes place
within the limits of the kitchen walls in the first act and moves to a Greek
summer resort for the second act. The confinement to a single set per act is a
symbol of the claustrophobic and dull life that the heroine leads, particularly
in the first act. This unity of setting would obviously seem uninspiring for
cinema versions and has therefore been divided for the film into the different
places that Shirley mentions in the account that she gives to the wall of her
past and present life.
Similarly, the unity of time and a limited number of characters are factors
that will bear greatly on a director's decision to transpose a play into another
medium: while they both favour adaptation for the radio and TV, they will
discourage the transposition of a play into a cinema film, which normally asks
for discontinuous time — moving backwards and forwards —, a large number
of characters and extras and a greater reliance on the visual than on the verbal
component. This may seem to have little to do with translation proper but in
fact the choice of medium may form part of the preliminary norms of the
translation process, as it will determine which plays lend themselves to one
mode of presentation or to another and will therefore decide which plays to
translate at a given moment. In the era of the reign of audiovisual media, the
choice of medium may explain why certain plays become known in a foreign
culture, while others occupying a more central position in the source culture
cannot go across the border: their being limited to stage performances might
Marta Mateo 105
prevent them from becoming known in a target context for which a different
mode of presentation is usually chosen.
Duration or running-time may be another factor to consider when deciding
whether or not to translate and/or transpose a play, and indeed most TV and
film adaptations of plays — particularly the classics — entail considerable cuts
in the dialogue. Thus, of all the four Spanish versions of The School for
Scandal, it is the 1967 TV production that has the shortest duration. This has
been achieved by cutting lines, drastically reducing scenes, removing some
secondary characters and replacing verbal signs with gestures.
We must not, however, overstate the influence of the medium chosen on
presentation decisions, particularly as regards stage directions. The different
solutions to the various versions of a play do not seem to depend entirely on
the choice of medium and frequently appear to be related to directorial
preferences: as Törnqvist has shown (1991:180-1), most decisions on gestures
and movements come from the director, while cuts in the dialogue are closely
associated with the channel chosen — stage performances tending to retain
more of the text than screen productions, and TV versions retaining more than
film versions. However, there is frequent overlap between directorial and
media differences, there being diverse approaches within each medium and
some directors' visions lending themselves better to one medium than to
another.
The theatre building
The performance of a play at a different theatre from that for which it was first
conceived — both in terms of context and of the structure of the theatre —
may entail a completely different reception of the play. The translation
strategies for a drama text will therefore be partly determined by the cultural
location and the design of the theatre at which it will be performed.
Due to the close communication between addresser and addressee in the
theatre, plays are usually subject to alterations so as to fit the established
theatrical conventions and cultural expectations of the target audience. This and
the fact that spoken language changes more quickly than written explain why
there are often several translations of the same play, as every new generation
would ideally require a new translation with which to share a new experience.
The drama translator will have to confront the same type of cultural
differences that are present in most other translation activities. What sets
drama apart from other genres is that the communication between the receptor
and the target text in the theatre implies a very complex process of shared
experience with an attached quality of immediacy. Receptors decode messages
in terms of their own experience and cultural expectations; in the case of
drama, their decoding and feedback reach the actors and director at the very
moment of the performance. This implies that the need to bridge the gap
between the two cultural systems involved will loom large in the mind of the
drama translator if the communication between the actors and the audience is
to succeed. Even so, a sense of danger will always be there.
An extreme case is provided by Fotheringham (1984:32-33), who recalls
a failed performance of The Legend of King O'Malley by Michael Boddy and
Robert Ellis in Australia. The performance was not a failure because of bad
acting or technical problems, but because of one teenage aboriginal girl sitting
three rows behind the fifty white people that formed the bulk of the audience
and to whom the play was clearly addressed. Social jokes about aborigines
were met with "an electrified silence" and proved that the play had been
written for the "white, urban, middle-class Australian". As Fotheringham
(1984:32) puts it, "the meaning of any play is modified by the structure of the
audience." This "structure" must sometimes be understood in more practical
terms than the cultural make-up of the audience, since it may be given by the
way the spectators are sitting in the theatre, which may easily determine their
response as a group:
In May 1989 RSC director Adrian Noble was reflecting upon his difficulties in
getting a laugh when directing Twelfth Night in Japanese at [a theatre] in Tokyo.
This theatre was new, it was well-equipped. At first he thought the reason why
the audience did not laugh was because in Japan laughter in public places is
thought impolite and therefore the audience had better hold their programmes up
to their faces and giggle quietly behind their fans. In a letter to this author he
recalled that he changed his mind when he realised that the problem lay
elsewhere: "The seats were very comfortable to sit in, but designed with a very
high back, so that one disappeared into the chair, as on an aircraft, and had no
sense of the surrounding audience at all. [...] [F]or a comedy like Twelfih Night,
[this] was very destructive, as it made it extremely difficult for the audience to
become welded into one group." (Mackintosh 1993:126-7)
It is therefore not only the context of the target theatre but also its structural
design that will determine audience reception and hence may influence
translation strategies. Each different type of playing space with its specific
stage-auditorium arrangement shapes the relationship established between
actors and audience, who are respectively provided with different acting
possibilities and theatrical expectations depending on their physical and
perceptual relationship.
Marta Mateo 107
Shakespeare's plays were written for Elizabethan theatres, whose thrust

stages meant that the audience played an active part in the communicative
process: they could hear and laugh at everything that took place on the stage
and it was usually the word that created the comic passages. As Mackintosh
(1993:10-14) explains, builders managed to get "as many people as close as
possible to the actor without jeopardising the actors' primary task of communi-
cating with every spectator, however distant. " The audience could practically
touch the actors, and their attitude towards the physical distance between
spectators was also very different from the standards of density we have
nowadays: it is believed that the first Rose in London held 2000 people, while
the second probably 2500 in a space which would not hold more than 400 or
500 today. The transposition of a play written with this arrangement in mind
to a different type of theatre — eg. one with an Italian design, which shows
a confrontation between actors and audience — will inevitably entail a different
reception of the play and may sometimes influence translation strategies.
An example of this is provided by a Catalan performance of The Merry
Wives of Windsor1 in the Teatre Grec in Barcelona, an open-air semi-circled
theatre with a seating capacity of 2500. Some of the recommendations made
by the director Carme Portacelli to the actors during rehearsals show how
different our standards of density are, compared to Elizabethan theatres. The
Teatre Grec has more or less the same capacity as a Shakespearean theatre but
the auditorium is much bigger and therefore considerably less densely packed,
so that the audience's response to the actors' lines is necessarily different2.
Thus Carme Portacelli has to recommend the actresses playing Mrs. Ford and
Mrs. Page to underline their quick replies in II.i, a scene in which they take
turns to read the identical love-letters that Falstaff has sent to them. If the
dialogue is not highlighted, the quick passages will be missed by the audience.
Audience capacity has here influenced actors' delivery, rather than translation
strategies as such — apparently, the director worked on Sagarra's target text
rather than with him —, but actors' delivery is inevitably determined by the
wording chosen by the translator, so that this still serves as an example of how
a performance element may bear on translation strategies when the translator
works for a particular production.
1
I am grateful to Eva Espasa, from the University of Vich in Barcelona, for this
information. The translation used for this performance was made by Josep Maria de
Sagarra. 1980. Les alegres casades de Windsor. Barcelona: Bruguera - Publications de
l'Institut del Teatre.
2
"Less densely packed auditoriums dilute the response received by the performer"
(Mackintosh 1993:171).
Something which varies from country to country and which is very rarely
remarked on is the viewpoint taken for stage-directions referring to space:
"right" and "left" may be interpreted either from the actor's or from the
spectator's position, and this is not usually specified by playwrights, probably
because each nation has its own conventions and there is no need to. But it
may create a problem for translators, who very often overlook this difference.
An exception is a translator of Strindberg into English, who turned the Swedish
playwright's indication for the entrance to the house in The Ghost Sonata —
"to the left upstage" — into English "at the rear, right" and explained the
change in the "Translator's Foreword": "For the purpose of conforming to the
American stage custom, I have reversed the author's directions — Right and
Left — to their opposites. Thus they are given here from the viewpoint of the
actor on the stage" (Arvid Paulson in: Törnqvist 1991:104).
The theatre stage
This shows that translators must be as careful with stage directions as with the
dialogue, since they form part of the whole network of signifiers in the drama
text. The value each direction has for the words spoken by a character at a
given moment and the cultural differences that may be attached to a certain
gesture, movement or prop must be taken into account in the analysis prior to
translation decisions, even if these finally entail deleting the stage direction
altogether.
The cultural value that kinesic and proxemic signs may have is shown in
a Spanish translation of The School for Scandal published in 1868, in which
the adaptation to the target culture was not only done on the textual level but
was also present in the kinesic signs: the translator inserted many stage
directions which conveyed Spanish people's supposedly emotional expressive-
ness, including roars of laughter and some hugging and crying which were not
present in Sheridan's source text and which added to the acceptability of the
target text.
The translation process may be affected by the unavailability of props
referred to in the dialogue in the new theatre and by the different emphasis that
cultures lay on the visual and verbal systems in drama. But, thanks to the
complexity of the semiotics of drama, the translator is then afforded the
possibility of resorting to any type of sign — linguistic, kinesic, musical, etc.
— to solve a problem raised in a different sign-system (Mateo 1995:25-28).
Arnott explains that Greek verbal imagery can easily be translated into visual
Marta Mateo 109
terms, yielding to "the modern theatre's greater reliance on visual effect"

(1961: 90): instead of having the chorus describe a tense meeting by resorting
to the language of tournament and the wrestling-ground, the debate scene could
be set in a boxing ring and the target text could then become simple and
concise.
The different use of the stage that cultures make can also determine
performances and translation decisions. The great number of scenes in
Elizabethan drama implied a comparatively empty stage and no real correspon-
dence between the stage directions (which were normally inserted in the
dialogue) and the props present on the stage, since audiences did not require
the realistic settings that modern spectators are used to. This means that
translators and/or directors sometimes feel the need to adapt the text in order
to make it fit in with the production facilities provided and meet the audience's
expectations of a realistic correspondence between setting and dialogue. In the
Catalan performance of The Merry Wives of Windsor mentioned above, the
director decided to change Falstaff's line "She shall not see me: I will ensconce
me behind the arras" (III.iii. 83) — which had been literally translated by
Sagarra into "No em veurà; m'amagaré darrera la tapisseria" — into
"m'amagaré aquí darrera" ("behind this/here") so that it corresponded with the
setting, which did not include any tapestry (arras). Curiously enough, whereas
Arden's edition explains in a footnote on arras that "[i]t does not necessarily
follow from the action here that there was an arras on the Elizabethan stage,
[...so that] if there were no arras, a pillar or any other available hiding-place
would have served,"3 the director of this Catalan production, which explicitly
rejected a realistic interpretation of the play, at times showed an excessive
desire for realism in this respect.
In that case the director had changed the translator's text. On other
occasions the decisions taken about the stage components of the play may
precede and have an effect on translation strategies. An 1861 Spanish
translation of The School for Scandal, clearly performance-oriented, seems to
have had the different settings of the play as the pivotal factor in translation
decisions: the translator rearranged the scenes in the different acts according
to the setting where they take place so that those sharing the same setting are
grouped together, which entails considerable changes in the plot and climax of
the play.
The viewpoint taken for one aspect of the play necessarily affects all the
3
In The Merry Wives of Windsor ed. by H.J. Oliver for The Arden Shakespeare, London:
Routledge, (1993/1971:83).
other elements, and the starting point for the approach taken need not always
be the text. It may sometimes be a paralinguistic feature of the dialogue:
apparently Chekhov's plays are usually translated into English with a Southern
American accent since the rhetoric of the language of the warmer regions
seems to fit the feelings of his plays better; this decision about dialect is then
followed by the corresponding setting, cultural references and so on. At other
times, it is the question of costuming that sets the viewpoint taken for the other
aspects of the translation and production (Hollander 1959:227).
Finally, another performance element which may affect drama translation
concerns the theatrical institutions that commission the plays for production.
Bennett (1990:119) explains that the strong reliance on lighting, theatre size
and technical effects of some modern London theatres — such as the National
Theatre and the Royal Shakespeare Company's Barbican — has shaped the new
drama commissioned by these theatres, so that plays written specifically for
them have tended to be "epic" dramas with large casts and multiple scenes.
This may also determine the type of foreign plays chosen for translation and
production in those theatres, which shows that preliminary norms need not
always come from purely literary or cultural values but may derive from
practical factors such as a theatre's production facilities or more financially-
oriented factors, like the need to make a mainstream theatre venue profitable.
References
Arnott, Peter. 1961. "Greek Drama and the Modern Stage". In: W.R. Arrowsmith/R. Shattuck
(eds.) 1961. The Craft and Context of Translation: a Symposium. Austin: University of
Texas Press, 83-94.
Aston, Elaine/Savona, George. 1991. Theatre as Sign-System. A Semiotics of Text and
Performance. London: Routledge.
Bennett, Susan. 1990. Theatre Audiences: a Theory of Production and Reception. London:
Routledge.
Elam, Keir. 1993. The Semiotics of Theatre and Drama. London: Routledge.
Fotheringham, Richard. 1984. "The Last Translation: Stage to Audience". In: O. Zuber (ed.),
29-39.
Heylen, Romy. 1993. Translation, Poetics and the Stage. Six French Hamlets. London:
Routledge.
Hollander, John. 1959. "Versions, Interpretations, Performances". In: R. Brower (ed.) 1959.
On Translation. Cambridge, Mass.: Harvard University Press, 205-231.
Mackintosh, Iain. 1993. Architecture, Actor and Audience. London: Routledge.
Mateo, Marta. 1995. "Constraints and Possibilities of Performance Elements in Drama
Translation". Perspectives. Studies in Translatology 1, 21-33.
Pulvers, R. 1984. "Moving Others: the Translation of Drama". In: O. Zuber (ed.), 23-28.
Törnqvist, Egil. 1991. Transposing Drama. Studies in Representation. Basingstoke: Macmillan
Education.
Zuber, Ortrun. (ed.) 1984. Page to Stage: Theatre as Translation. Amsterdam: Rodopi.
From saint to sinner: The demonization of Oscar
Wilde's Salomé in Hedwig Lachmann's German
translation and in Richard Strauss' opera
Rainer Kohlmayer
In this paper it is my objective to show how Hedwig Lachmann, the wife of

the well-known anarchist writer Gustav Landauer, intervened as translator of
Oscar Wilde's Salomé in such a decisive way as to predetermine the German
(and, indeed, worldwide) reception of the play from 1900 on up to the present
day.
Oscar Wilde wrote Salomé in French while he was in Paris in November
and December 1891. It was a bold attempt to break away from the paradigm
of England's cultural tradition and to return to the Greek sources of European
art. For Wilde this implied three things: verbal music, Aristotelian dramaturgy,
and anti-Christianity.
In his essay The Critic as Artist (1890) Wilde had condemned traditional
literature with its elaborate designs and advocated a return to the musical
language of Greek antiquity:
The Greeks [...] regarded writing simply as a method of chronicling. Their test
was always the spoken word in its musical and metrical relations. The voice was
the medium, and the ear the critic. [...] Yes: writing has done much harm to
writers. We must return to the voice. That must be our test [...] (Wilde
1987:1017).
The French language served Wilde, in a way, as a modern substitute for
Greek. "To me there are only two languages in the world: French and Greek"
(Ellmann 1987:352), Wilde told a French interviewer. The play Salomé, with
its multitude of leitmotifs, refrains, alliterations, assonances, and other stylistic
devices was — in Wilde's own words — "like a piece of music" (Wilde
1987:922). In this respect Salomé was quite in keeping with French Symbolist
literature of the 1890's (e.g. Maeterlinck).
On the other hand, the classicist macrostructure of Salomé, with its

Aristotelian unities and dynamic dramaturgy, was a far cry from the static
aimlessness of symbolist drama.
Wilde's aesthetic return to Greek classicism was, on a deeper level, a
return to the pre-Christian sources of Europe. Salomé was written as a heathen
act of deconstruction directed against the conventional — in the sense of
Puritan — interpretation of the Bible. It is a Nietzschean work of art in that
it propagates a profound revaluation of traditional values. Wilde's message
might be summed up briefly as follows: Christianity befell the ancient world
as a veritable disaster.
Wilde reinterprets the Biblical story of John the Baptist and Salomé in
such a way that it is no longer the Christian saint who falls prey to the dancer
Salomé or her mother Herodias, but it is rather Salomé who is destroyed by
John the Baptist.
Wilde portrays the arrival of Christianity in the ancient world as an
apocalyptic catastrophe. Thus, Jochanaan (as John the Baptist is called in
Wilde's play) announces triumphantly: "Les centaures se sont cachés dans les
rivières, et les sirènes ont quitté les rivières et couchent sous les feuilles dans
les forêts" (Wilde 1908:16).
For the centaurs and sirens, the mythical symbols of the male and female
union of humans and animals, the beginning of Christianity ushered in the
panic of the apocalypse. The destruction of ambiguities, of the unity of body
and soul, of the polyphony of antiquity, now continues resolutely in the
liquidation of Salomé. It is important to realize that the initiative for and even
the exact method of Salomé's execution at the end of the play stems from
Jochanaan. King Herod merely carries out the brutal demands for a lynching
that Jochanaan had hurled at Salomé from his prison, claiming his curses were
the pronouncement of divine judgement:
Voici ce que dit le Seigneur Dieu. Faites venir contre elle une multitude
d'hommes. Que le peuple prenne des pierres et la lapide... [...]. Que les
capitaines de guerre la percent de leurs épées, qu'ils l'écrasent sous leurs
boucliers. [...] et que toutes les femmes apprendront à ne pas imiter les
abominations de celle-la (Wilde 1908:50f.).
Just as Jochanaan demands, at the end Salomé is literally crushed under the
shields of the soldiers: "Les soldats s'élancent et écrasent sous leurs boucliers
Salomé [...] (Wilde 1908:82) — a striking symbolic image of collective lynch
law, anticipating phenomena such as the witch-hunting of the Middle Ages.
In view of Jochanaan's brutal call for Salomé's murder, her demand to
have him beheaded becomes not only an act of revenge but also self-defence.
Rainer Kohlmayer 113
The play is, in one respect, an extremely succinct biography of Salomé,

who appears on stage as an innocent child full of questions and a thirst for
knowledge, whose sudden love for Jochanaan is callously rejected and cursed,
who takes revenge and dies an incredible death of passion, while kissing
Jochanaan's severed head. Salomé's life is the rejection of self-denial; holding
on to love at all costs is her anarchic, even absurd act of rebellion and vitality.
Salomé's monologue of love ends with the words: "[...] le mystère de l'amour
est plus grand que le mystère de la mort. Il ne faut regarder que l'amour"
(Wilde 1908:80).
This is an unequivocal statement on the priority of love, but can also be
understood as a commentary on Jochanaan's misspent life, a prophet and man
who — according to this interpretation — did not truly live before his death.
The allusion to the Song of Salomon ("l'amour est fort comme la mort", 8,6)
and to the well-known Virgilian verse "omnia vincit Amor: et nos cedamus
Amori" (Bucolica, X, 60) adds weight to Salomé's words. Her biography is
presented as exemplary, as a legend of a heathen martyr, a saint of the senses,
so to speak.
The play Salomé is a refutation of the Bible. It is an ironic, anti-Christian
interpretation of the great turning point of history: the ancient world of love
is destroyed, and terror in the name of the Christian God begins its reign.
Hedwig Lachmann's translation
For the history of the German reception of the play, it was not the original
French but rather the English version that was of primary importance — a fact
that had remained undiscovered until today. The French edition of the play
appeared in 1893 in book form, the English translation by Lord Alfred Douglas
with illustrations by Aubrey Beardsley came out in 1894. Wilde criticized this
translation heavily and most likely corrected parts of it (Ellmann 1987:379-
381). On the whole, the translation only rarely does justice to the musicality
of the original. Most noticeable is the archaic solemnity of intonation. Wilde's
French was the language of the day; the English translation, in contrast,
historicizes the work. It is not only the prophet Jochanaan who speaks using
the syntax of the Bible translation of 1611, the official King James version:
"The Lord hath come. The son of man hath come" (Wilde 1987:555) etc.,
which could be most easily justified, but also Salomé and Herod. For example,
after Salomé's dance the 'French' Herod turns to the dancer in easy
confidence: "Approchez, Salomé! [...] Ah! je paie bien les danseuses, moi.
Toi, je te paierai bien. Je te donnerai tout ce que tu voudras. Que veux-tu,
dis?" (Wilde 1908:66). In English the passage is as follows: "Corne near,
Salomé [...]. Ah! I pay the dancers well. I will pay thee royally. I will give
thee whatsoever thy soul desireth. What wouldst thou have? Speak" (Wilde
1987:570).
"[T]hee", "royally", "thy soul" etc. is reminiscent of a fairy tale removed
from the present. This also applies to Salomé's manner of speaking. Her
everyday statement: "Viens ici. Tu a été l'ami de celui qui est mort, n'est-ce
pas?" (Wilde 1908:11) becomes solemn stage rhetoric in the English: "Come
hither, thou wert the friend of him who is dead, is it not so?" (Wilde
1987:573).
To Wilde himself, Beardsley's illustrations appeared to be
too Japanese, while my play is Byzantine. [...] My Salomé is a mystic, the sister
of Salammbo, a Sainte Thérèse who worships the moon; dear Aubrey's designs
are like the naughty scribbles a precocious schoolboy makes on the margins of
his copybook (Qtd. in Jullian 1969:218).
Beardsley's sketches had and have still today a strong influence on the stage
reception of Salomé. Salomé's costumes and posture are often designed in
imitation of Beardsley's drawings.
Apparently without Wilde's knowledge, the play was published in the June
1900 edition of the art magazine Wiener Rundschau under the title Salome.
Tragödie in einem Aufzug von Oscar Wilde (London). Deutsch von Hedwig
Lachmann. Mit Zeichnungen von Beardsley. This is the translation which set
a precedent for the whole German reception, including the opera by Strauss,
and was reprinted time and again. The place reference on the title page Oscar
Wilde (London) was probably intended to be a discret indication of the fact that
Lachmann's translation was based on the English version of the play. Later
editions omitted such allusions. The most recent Reclam edition (1990) asserts
explicitly and incorrectly: "Aus dem Französischen übersetzt von Hedwig
Lachmann."
Obviously, Lachmann followed the English text from beginning to end;
however, she must have used a partially corrected English version or the
French text as a reference, as several lexical errors in the English translation
were corrected.
Hedwig Lachmann (1865-1918) deserves the fame critics have bestowed
upon her since her translation appeared. Her text sounds like a powerful
German original; since it follows the English text rather closely, it is, on the
whole, rougher and more solemn than the French, but does not imitate the
archaizing, historicizing fairy-tale style of the English version. Lachmann
chooses an elevated form of contemporary spoken German that is dramatically

intensified through variation, compounding and a dynamic rhythm. Lachmann
does not strive to recreate the soft impressionistic music inherent to the French
original. The lexical variation is greater than in Wilde's original and strives for
more powerful rhetorical effects. Jochanaan's cistern prison is referred to as
"très malsain" (Wilde 1908:12) and "very unhealthy" (Wilde 1987:554) in the
French and English, whereas it becomes "ein mörderischer Ort zum Wohnen"
(Wilde 1900:193) in the German; "grands corps" (Wilde 1908:24), i.e.
"mighty bodies" (Wilde 1987:537), becomes "Leiber wie von Riesen" (Wilde
1900:195); for "battement d'ailes gigantesques" (Wilde 1908:39), i.e. "beating
of vast wings" (Wilde 1987:562), Lachmann selects the acoustically and
rhythmically more impressive "Rauschen von mächtigen Flügeln" (Wilde
1900:199). Towards the end of the play, Lachmann intensifies emotions even
more: Herodias' accusation "Pourquoi la regardez-vous toujours?" (Wilde
1908:40), i.e. "Why are you always gazing at her?" (Wilde 1987:562),
becomes "Warum stierst du sie immer an?" (Wilde 1900:200); 'Angst haben"
(Wilde 1908:53, 81) becomes "erzittern" (Wilde 1900:203, 212); "ma passion"
(Wilde 1908:80), i.e. "my passion" (Wilde 1987:574), becomes "dies brünstige
Begehren" (Wilde 1900:210).
Lachmann makes generous use of the specifically German possibility of
forming compounds in order to create dozens of new words: "Bernstein-
Augen" , "Sündenbecher", "Hycinthgesteine", "Schlangenknoten",
"Scharlachband", "Granatapfelblüten" etc. (Wilde 1900:195, 197). This
dynamic use of language has a magnificent rhetorical effect in the monologue
Salomé delivers while holding Jochanaan's head: "Und deine Zunge, die wie
eine rothe, giftsprühende Schlange war, sie bewegt sich nicht mehr, [...] diese
Scharlachnatter, die ihren Geifer auf mich spie" (Wilde 1900:210).
With regard to the deficiencies of Lachmann's translation, at least three
items must be noted or criticized, of which the first has to do with the gender
of "la lune" and "der Mond". The French (as the English) allows a smooth
transition to the female personification of the moon; Lachmann's German text
employs various aids such as "Mondscheibe" or "Ist es nicht ein seltsames
Bild? Es sieht aus wie ein wahnsinniges Weib [...]" etc. (Wilde 1900:191,
198). Nor did Lachmann take into account the differentiated forms of address
that are revealing with regard to the relationships between the characters,
which can be explained by the influence of the English version. In the French
text, there are subtle distinctions that could have been incorporated into the
German. For example, Salomé addresses only Jochanaan with his name and the
intimate form "tu" from beginning to end; she keeps all other characters at a
distance. Finally, Lachmann destroys the parallel symbolism of the centaurs

and sirens who hide in the rivers and in the forest: She turns the sirens into
"Nymphen", who lie "unter den Blättern des Waldes begraben" (Wilde
1900:193), which no longer conveys the same impression of flight and the
inversion of the ancient order of nature. In Wilde's original, the world of
antiquity is by no means "begraben", but rather continues to live on under the
surface.
In addition, there are several important interventions on the part of the
translator that directly influenced the productions of the play and Strauss'
opera. In the French text, Salomé enters the stage more childlike, in the
German more aware of her erotic appeal. First, Salomé's appearance in the
French text:
Je ne resterai pas. Je ne peux pas rester. Pourquoi le tétrarque me regarde-t-il
toujours avec ses yeux de taupe sous ses paupières tremblantes?... C'est étrange
que le mari de ma mère me regarde comme cela. Je ne sais pas ce que cela veut
dire... Au fait, si, je le sais (Wilde 1908:14).1
Salomé's manner of speaking is slightly lyrical due to rhythm, assonances, and
alliterations. The two signs for pauses (marked by dots) are also important. A
mere child, Salomé is confronted with Herod's behaviour, which puzzles and
irritates her; she describes it with the somewhat childlike image of a mole and
does not seem to understand or seems to suppress the intentions guiding him,
until she hesitatingly admits after a pause that "eigentlich, ja" she does indeed
know what his behaviour means. Lachmann's text:
Ich will nicht bleiben. Ich kann nicht bleiben. Warum sieht mich der Tetrarch
fortwährend so an mit seinen Maulwurfs-Augen unter den zuckenden Lidern? Es
ist seltsam, dass der Mann meiner Mutter mich so ansieht. Ich weiß nicht, was
es heißen soll. In Wahrheit — ich weiß es nur zu gut (Wilde 1900:193).
This Salomé is more deliberate and purposeful; the "Ich will" in the first
sentence, which Lachmann takes from the English (future tense), is the first
indication; the two pauses in the original French version are also missing in the
English, hence also in Lachmann's text. Lachmann's Salomé is clearly aware
of her erotic attraction, the reason why Herod looks at her the way he does.
Nor does she ask herself what this means, but rather "was es heißen soll", i.e.
what is intended by it; and Salomé's knowledge is stressed emphatically: "In
Wahrheit — ich weiß es nur zu gut".
1
"I will not stay. I cannot stay. Why does the Tetrach look at me all the while with his
mole's eyes under his shaking eyelids? It is strange that the husband of my mother looks
at me like that. I know not what it means. In truth, yes I know it" (Wilde 1987:555).
Almost in slow motion, Wilde portrays her development from a child to

a young woman growing aware of her sexuality during her first entrance;
Lachmann, in contrast, has a fully-fledged femme fatale appear on stage.
Salomé's heightened erotic awareness is accompanied by a certain
exculpation of Jochanaan and Herod. In the French (and in the English
translation as well) Jochanaan calls for Salomé's execution after their
confrontation by ordering her to be stoned, stabbed and crushed to death
(Wilde 1908:51); in Lachmann's text, however, he merely predicts these
different ways of dying: "Die Kriegshauptleute werden sie mit ihren
Schwertern durchbohren, sie werden sie unter ihren Schilden zermalmen"
(Wilde 1900:202).2
Through this translation of Jochanaan's demand as a mere prpphecy, he
is no longer personally involved in Salomé's execution. Herod makes the
decision to have her put to death alone; it is not recognizable that he is merely
executing the orders Jochanaan had given before his own death, or that, since
that moment, Herod seems to speak with the dead prophet's voice.
In her book on Wilde, published in 1905, Hedwig Lachmann revealed in
a ten-page narrative of the play the psychological drama that, as a subtext,
constitutes the basis of the translator's mise-en-scène. She speaks of
"elementarer Wildheit" (Lachmann 1905:39), of
grosse Typen von singulärer Art [...]. Die Charaktere sind einfach, stark, auf
e i n Gesetz, gleichsam auf eine Formel gebracht: den unbändigen, durch kein
Verstandes-moment geschwächten und gehemmten Eigentrieb des Individuums.
[...] Salome ist die willensstarke, unzerspaltene Natur, deren Lebensenergien im
vollen Einklang mit der Größe ihres Schicksals und ihres Verbrechens sind
(Lachmann 1905:47).
At the end, due to Herod,
[wird] dem Verbrecherischen im Drama ein Mass gesetzt, er repräsentiert
gleichsam die Grenzen der Menschlichkeit. Und mit einer wahrhaft grossen
Bewegung löst er sich in einem Moment von all dem Ungeheuerlichen, das um
ihn vorgeht, vollkommen ab und erhebt sich zur selbstsicheren Persönlichkeit,
indem er Salome das Todesurteil spricht (Lachmann 1905:49).
As becomes apparent, Lachmann focuses solely on Salomé's crime, not
on Jochanaan's verbal destruction of her, nor on his murderous appeal for a
lynching, nor on Herod's function as a proxy, nor on Salomé's final
2
"Let the war captains pierce her with their swords, let them crush her beneath their
shields" (Wilde 1987:565).
declaration of love. In this context, it is, I believe, significant that Lachmann

simply drops Salomé's final sentence "Il ne faut regarder que l'amour" (Wilde
1908:80), i.e. "Love only should one consider" (Wilde 1987:574). It is the
only sentence in the text that Lachmann does not translate; she probably did
not think this modest statement was suitable for the magnificent subtext she
imagined.
The psychological drama of the "tierhafte Wildheit Salomes" (Lachmann
1905:48), underlying Lachmann's translation, projects a grandiose, anarchic
vitality into this character: the eruption and taming of 'woman' as natural
force.
The dialectic relationship between Jochanaan and Salomé does not come
to her mind. In and since Lachmann, as Jochanaan does not call for lynch law
with regard to Salomé, Jochanaan embodies solely the passive purity of a
Christian prophet detached from the world, so that, for the theatrical reception
of the characters as well, the simple dichotomy of whore and saint had to
impose itself. In the words of an early theatre critic: "Der Täufer fällt der Gier
eines Mädchens von dirnenhaften Instinkten zum Opfer [...]. Salome fällt als
Opfer ihrer Gier [...]" (Qtd. in Jaron et al. 1986:525).
The theatrical and musical reception of Lachmann's text
Lachmann's version was a sensational success, first performed on 15

November 1902, in Berlin, in a private production at Max Reinhardt's Kleines
Theater. All of the leading critics from all over Germany were invited, in
addition to writers, artists and musicians such as Stefan George and Richard
Strauss. In the role of Salomé, Gertrud Eysoldt projected the combination of
"Weib und Tigerin" that a critic had believed to have recognized in Beardsley's
book illustrations for Lachmann's translation as early as 1900 (Becker
1901/02:203). Almost all critics agreed that Gertrud Eysoldt brought the
anarchic, 'animal' traits of Salomé to life grandly, it being evident that, in
particular, Lachmann's text, i.e. subtext, played to Eysoldt's demonic style of
acting. Thus, the critic of Vorwärts describes Salomé's first entrance as
follows: "Langsam, in leisem Selbstgespräch voll lüstern-listiger Gedanken
schleicht Salome aus dem Festsaal herbei. Es kitzelt sie, daß Herodes, der
Mutter Gemahl, mit verliebten Augen an ihr, der Tochter hängt" (Schmidt
1902).
Similarly, Paul Block mentions Salomé's "frühreife Jugend" and that

Salomé "ahnungsvoll nach Erfüllung wilder Begierden lechzt. Sie weiß, daß
der Stiefvater sie mit heißen Blicken anschaut" (Block 1902).
The development of Wilde's Salomé from the French original into
English, via Beardsley's drawings, Lachmann's translation, Reinhardt's
production, i.e. Eysoldt's acting, shows a process of increasing radicalization
and brutalization of the character of Salomé, who is removed from the
historical context of early Christianity and shaped with increasing clarity into
an icon of eruptive sexuality.
This development reaches a peak in Richard Strauss' opera, which was
first performed on 9 December 1905, in Dresden, with sensational success (39
curtain calls) and since then has been part of the repertoire of the leading opera
houses of the world. Initially, Strauss had wanted to have a libretto written by
Anton Lindner, the music critic who had published Lachmann's translation of
Salomé in 1900. On 30 April 1902, however, the premiere of Debussy's
Pelléas et Mélisande, which received much attention in Paris, took place. This
opera was a novelty due to the fact that Debussy had set Maeterlinck's prose
text to music without the aid of a librettist. The success of this first 'literary
opera' in the modern sense probably led Strauss to do without the customary
libretto and also set Lachmann's text to music directly as well, incorrectly
assuming, however, that Wilde's French original had been "translated literally
by Ms. Lachmann," as he had written to his publisher on 5 July 1905 (Qtd. in
King et al. 1991:75).
Strauss shortened the text almost by half, but in his composition he
followed the melodic and rhythmic phrasing of Lachmann's text exactly, true
to the principle adopted from Wagner that the melody must emerge from the
word. The dynamic style and all the other changes written into the translation
by Lachmann, or which were already present in the English text, thus
constituted a verbal programme that inevitably influenced the creative process
of composing. For example, Strauss follows the rhythm of Lachmann's
opening sentence, with the strong initial accent on the word "schön": "Wie
schön die Prinzessin Salome heute Abend ist!" (Wilde 1900:191).3 The
translator had followed the English version literally, while the French original
had placed the main stress at the end: "Comme la princesse Salomé est belle
ce soir!" (Wilde 1908:5). For Strauss, it was also a matter of course that the
3
In the book edition of her translation (1902) Lachmann changed the opening sentence
into: "Wie schön ist die Prinzessin Salome heute Nacht!". Richard Strauss' libretto of
1905 is based on Lachmann's text as published in 1902.
root syllable stress of the individual words be respected. Hence, he wrote to
Romain Rolland concerning the composition of Salomé:
Im 4/4 ist jedes erste und dritte Viertel fast stets notwendig ein Accent, dem nur
die Wurzelsilbe jedes Wortes anvertraut werden kann. Seit Wagner natürlich!
Vorher nahm man es nicht so genau, wenn nur die Melodie schön war (Qtd. in
King et al. 1991:79).
Lachmann's interpretation was heightened and transformed by Strauss into a
monumental world of sound by an orchestra consisting of more than one
hundred musicians, a work which suggestively evokes the "mental underworld"
(Schmidgall 1977:281) of the main characters. Jochanaan, however, whom
Strauss despised (Del Mar 1962/1:250), must make do without a subconscious.
Alfred Kerr's ridicule that, in Strauss, Jochanaan had become "fast ein
Kreuzritter mit Marschmotiv; ein Gottesmann in B-Dur, ein deutscher
Jochanaan; im Kern ein blonder Prophet" (Kerr 1954:271) must be passed on
to Hedwig Lachmann, who had paved the way for Jochanaan's musical
sanctification.
Through his condensing of the text, musical commentary and symphonic
interludes (Mahling 1991:9If), Strauss focused attention even more on
Salomé's wildness as a femme fatale than Lachmann had. For example,
Salomé's first — 'childlike' — monologue is truncated to a few seconds, while
the pause before her first meeting with Jochanaan or her rising thoughts of
revenge after Jochanaan's maledictions are expounded upon with strikingly
forceful music.
Six months after the Dresden premiere, Strauss began working on a
French version of his opera. He did not, however, wish to have his libretto
translated, but preferred to replace Lachmann's text with Wilde's French
original, naively assuming that Hedwig Lachmann's "literal" translation would
be a quantité négligeable that had not left any traces on his musical
interpretation. His correspondence with Romain Rolland on this topic is not
without a comic element; for example, Strauss criticizes Debussy's post-
Wagnerian score of Pelléas, which Rolland had recommended to him as an aid
in learning the technique of French phrasing, because of its lack of agreement
between the word stress and musical accent (Strauss/Rolland 1951:44). For
these comments, he is in turn rebuked by Rolland: "Vous êtes trop orgueilleux
en ce moment, en Allemagne. Vous croyez tout comprendre, et vous ne vous
donnez aucune peine pour comprendre. Tant pis pour vous, si vous ne nous
comprenez pas!" (Strauss/Rolland 1951:47).
Strauss was willing to learn, and let Romain Rolland teach him the
differences between German and French stress and phrasing. After three
months of hard work, he had adapted all of the phrasing of the voice lines to
Wilde's French text, and Romain Rolland corrected the score for him. The
changes in the composition that, for example, were necessary to transform
sentences such as "Wie schön ist die Prinzessin Salome heute nacht! " back into
"Comme la princesse Salomé est belle ce soir!" were considerable.
Strauss' 'original version' in French, however, was not able to establish
itself. In 1909, he had Lachmann's text translated back into the French.
Singers performing internationally then only had to relearn the French text and
not the whole voice part. It was only in 1989/90 that Strauss' and Rolland's
French version was taken out of the drawer and performed at the Opéra de
Lyon. The most recent French production at the Opéra de Paris-Bastille in the
spring of 1994, however, uses Lachmann's German text once again.
In summary, it can be stated that Hedwig Lachmann played a decisive role
in the reception of Salomé in German by transforming Wilde's French
symbolist-impressionist "piece of music" (Wilde 1987:922) into a pre-
expressionist drama.
References
Becker, Marie Luise. 1901/02. "Salome in der Kunst des letzten Jahrtausends". Bühne und
Welt 4, 157-165, 201-208.
Block, Paul. 1902. "'Salome* und 'Bunbury'. Zwei Werke von Oskar Wilde im Kleinen
Theater". Berliner Tageblatt 16.11.1902.
Del Mar, Norman. 1962. Richard Strauss. A Critical Commentary on His Life and Works.
Vol. I. London.
Ellmann, Richard. 1987. Oscar Wilde. London: Hamish Hamilton.
Jaron, Norbert et al. 1986. Berlin — Theater der Jahrhundertwende. Bühnengeschichte der
Reichshauptstadt im Spiegel der Kritik (1889-1914). Tübingen: Niemeyer.
Jullian, Philippe. 1969. Oscar Wilde. London: Constable.
Kafitz, Dieter (ed.) 1991. Drama und Theater der Jahrhundertwende (Mainzer Forschungen
zu Drama und Theater 5). Tübingen: Francke.
Kerr, Alfred. 1954. Die Welt im Drama. Köln, Berlin: Kiepenheuer & Witsch.
King, Julia et al. 1991. Salomé. Booklet for the First Recording of the French Version by
Richard Strauss Using Oscar Wilde's Original Text. Mainz: B. Schott's Söhne.
Lachmann, Hedwig. 1905. Oscar Wilde. Berlin, Leipzig: Schuster und Loeffler.
Mahling, Christoph-Hellmut. 1991. "'Tönendes Schweigen' und Ausdruckstanz. Bemerkungen
zu zwei Komponenten des Musiktheaters der Jahrhundertwende". In: D. Kafitz (ed.), 87-
99.
Schmidgall, Gary. 1977. Literature as Opera. New York: Oxford University Press.
Schmidt, Conrad. 1902. "Theater". Vorwärts 18.11.1902.
Strauss, Richard. 1905. Salome. Drama in einem Aufzuge nach Oskar Wilde's gleichnamiger
Dichtung in deutscher Übersetzung von Hedwig Lachmann. Musik von Richard Strauss.
Berlin: Adolph Fürstner.
Strauss, Richard et Romain Rolland. 1951. Correspondances. Fragments de Journal. Paris:
Éditions Albin Michel.
Wilde, Oscar. 1900. Salome. Tragödie in einem Aufzug. Deutsch von Hedwig Lachmann. Mit
Zeichnungen von Beardsley. Wiener Rundschau 4 (12), 189-212.
Wilde, Oscar. 1902. Salome. Tragödie in einem Akt. Übertragung von Hedwig Lachmann.
Leipzig: Insel-Bücherei.
Wilde, Oscar. 1908. Salomé. A Florentine Tragedy. Vera. London: Methuen.
Wilde, Oscar. 1987. Complete Works. With an Introduction by Vyvyan Holland. London and
Glasgow: Collins.
Wilde, Oscar. 1990. Salome. Tragödie in einem Akt. Mit Illustrationen von Aubrey Beardsley.
Aus dem Französischen übersetzt von Hedwig Lachmann. Nachwort von Ulrich
Karthaus. Stuttgart: Reclam Universalbibliothek.
Translation as a process of power: Aspects of cultural
anthropology in translation
Michaela Wolf
Interdisciplinarity, according to Roland Barthes, consists in creating a new

object which does not belong to anybody (Barthes 1984:100). In this sense two
considerations are especially pertinent to this paper. Firstly, the term
"translation" will be treated here in a broader sense, which means it is used
in the sense of "rewriting" or even "cultural textualization". Therefore there
may not even exist an original text that is to be translated into a target
language and target culture. Secondly, translation studies are not to be
understood as a subdivision of linguistics, the study of literature, cultural
studies or any other sort of science, but as a discipline which operates mainly
through the encounter with these and other disciplines, e.g. cultural
anthropology. Or, as Mary Snell-Hornby has defined it (1986:12): "Wenn man
von den bestehenden Wissenschaften als Kategorien ausgeht, dann wäre die
Übersetzungswissenschaft als interdisziplinäre, multiperspektivische Einheit zu
verstehen...".
As a consequence, when "translating between cultures" there is a lot of
overlap between ethnography and translation, ethnography being understood
as a part of cultural anthropology, and therein mainly as an act of
representation or rather textualization of something observed. This paper will
try to locate these overlaps and — focussing on the matter of power — will
raise questions relevant to both these disciplines. Finally, propositions will be
made which may serve to encourage a prolific collaboration of both disciplines.
Ethnography and translation as culturally specific communication
The interaction of scientific disciplines which have some scientific and

methodological approaches in common has not always been self-evident. As far
as translation studies are concerned, it was recognised at a very early stage that
there were common issues with anthropology (see Nida 1945; Göhring 1978).
The real breakthrough, however, did not occur until it was understood that
translation is an act of culture-specific communication (Vermeer 1986), and
that ethnography, as textualization of oral discourses, is a social act, addressing
the "dialectical reformulating of the 'other'" (Ulin 1991:69).
Thus, "translating between cultures", in ethnography as well as in
translation, means interaction, intercultural activity. This paper will concentrate
on the kinds of activity which refer to the transfer between "Third" and "First
World". The basis for discussion in the translatological context will be texts
which refer to literary and political contents (without specific examples), as in
these fields the asymmetries of the transfer become specially visible (cf. Wolf
1995).
The discussion on asymmetrical power relations in the "translation
between cultures" is made evident in a series of publications in translation
studies as well as in ethnography — mainly in a colonial or post-colonial
context. Power as a social principle of development and integration, as it is
understood by Michel Foucault (1972; 1976), will serve as a basis for
demonstrating these asymmetries.
The "will to power"
In line with the above-mentioned claim of Foucault's concept of power as a

desire for limitless authority, power is bound to play a decisive role in
theoretical systems of knowledge. As a matter of fact, power and knowledge
are the two main topics in Foucault's philosophy. Following Nietzsche,
Foucault equates the will for knowledge with the will to power. His historical
reconstruction of cultural knowledge systems — which to him are discourses
(another major hallmark of his philosophy) or "discursive practices" — shows
how these systems of knowledge played an important role in the development
of Modern European thinking in the 18th and 19th centuries. Finally, Foucault
demonstrates by means of his historical analysis that the aim of this striving for
power is to control the threatening dangers of discourse (Foucault 1972:10).
Power is therefore seen as a means of control, of subjection, and of
repression.1 Therefore it is the mechanisms of political power rituals which
1
At a later stage, Foucault relativizes his hypothesis of repression, pointing out the
manifold mechanisms of power (cf. Foucault 1976:125).
Michaela Wolf 125
create non-egalitarian and asymmetrical relations (Dreyfus/Rabinow 1987:16).

It will now be demonstrated how these asymmetrical power relations are
revealed in translation and ethnography, or rather, in the interactive activities
that occur when "translating between cultures".
Asymmetrical power relations in the translation process
As far as asymmetrical power relations in the translation process are

concerned, the research conducted by Sapir and Whorf in linguistics and
cultural studies have shown that the respective language has a profound impact
on how we perceive and experience our world and our actions. If language is
understood as a basis from which a society experiences itself and others,
Western expansion into non-western societies always entails seizing power of
the respective language or languages. Thus, when colonising, Western societies
not only subjugated these societies economically and politically, but also
linguistically. For the field of translation this means (in the colonial context)
that the discourses of Western institutions are perpetuated in the discourses of
societies of the so-called "Third World" and thus perpetuate colonial structures
(Niranjana 1992:3).
It was mainly Lawrence Venuti who tried to reveal this interdependency
in translation. Venuti's theory is grounded on genealogy, i.e. Foucault's theory
of power practice. He shows how, due to the bias held by the publishing
sector, texts are subjected to "acculturation":
[...] and when such strategies are implemented, they inescapably perform a
work, of acculturation, in which a cultural other is domesticated, made
intelligible, but also familiar, even the same, encoded as it is with ideological
cultural discourses circulating in the target language (Venuti 1991:127).
According to Venuti, the demand that the "other" should be adopted as the
"other" means to see translation as a "locus" of heterogeneity and not of
homogeneity (Venuti 1992:13).
This request for heterogeneity can also be witnessed in the translation
of literature in francophone countries. Discourse about translation in Quebec
not only deals with keeping the French language free from interference; it
mainly discusses how the "otherness" of the text can be transferred to the
target culture in order to enlarge its cultural and linguistic repertoire.
Regarding literary translation, André Lefevere has used the "system
theory" (in particular his "patronage" system) to show how external
mechanisms determine the publication of literature in foreign languages. For
Lefevere translation is the "rewriting of an original text" (Lefevere 1992: VII).
All these rewritings reflect a certain ideology and poetic, and as such they
manipulate literature to function in a given society in a given way. For
Lefevere, the patronage system consists of several power factors such as
institutions or individuals which can hinder or promote processes like writing,
reading or rewriting of literature.
As we shall see, very similar restraints can be observed in ethnography,
regarding the problems that arise when textualizing the observations made in
fieldwork.
Asymmetrical power relations in ethnography
In ethnography the representation of societies with different value systems from

those of the observing and describing scientists has always been a problem, but
was never included in practical and critical reflections of ethnographers until
the 1920s and 1930s.
For ethnography decolonization brought about radical changes. Among
the immediate consequences were new methods. In the awareness that the mere
fact of having worked with or in a foreign culture does not necessarily imply
having knowledge of it, it has been requested (and observed) that ethnographic
authors should speak for the people they have observed. In other words, they
should act as their spokesmen. This request, in turn, brought about new rules
of scientific writing. In the area of syntax, for example, the use of the first
person singular should be avoided, in the area of content ethnographers should
come up with as many voices as possible that were essential during field
research. This approach is also known as "de-centred ethnography".
As we can see, the discussion on de-centred ethnography takes place
mainly in the field of "representation". By this we not only mean a specific
method of observation and recording, but also the social discourse of the
informant. Thus what counts is the attempt to express other discourses in one's
own form of discourse, as well as to identify and capture the differences in
language. Here we can locate an overlap with translation in the traditional
sense. It is paramount to break with a tradition that tries to integrate the
"other" in an objectifying and imperialistic way. As far as realisation of this
method is concerned, opinions are divided (as in the translation of
corresponding texts). This is also due to the fact that, as Stephen Tyler puts it,
each representation is an act of political oppression (Tyler 1993:288). In this,
Tyler, who is generally considered a representative of postmodernism in
Michaela Wolf 127
ethnography, goes beyond the field of ethnography and makes a reference to

the example of "representative government". He continues:
[...] selbst wenn Ethnographien als Reproduktionen des ursprünglichen
Dialogs zwischen uns und den Anderen, den Angehörigen einer anderen
Kultur, geschrieben werden, bleiben sie dennoch bedingt durch das Verfahren
des Schreibens und durch Kontexte und können keinen Fortschritt in Richtung
auf eine Wiederherstellung der ursprünglichen Präsenz des Anderen außerhalb
jenes Textes bewirken, der sie repräsentiert (Tyler 1993:289).2
Thus the author often assumes a critical role which makes him seem the only
real purveyor of truth in the text. How does this apply to ethnography and
translation?
Ethnography and translation in the same boat
If "faithful" representation is the key word in ethnography and translation

(provided that the question is: "How are cultural phenomena of the 'other'
represented in the target language/culture?"), does this not involve values or
value systems which — due to the asymmetrical power relations described
above — turn this endeavour a priori into a failure? And what about
textualization in different languages? Is it possible to understand "other"
realities by means of (in our case) a Western meta-language?
In the following I will try to work out the overlap of these questions in
translation and ethnography, first in the context of a general discussion, then
with regard to the processes of power involved.
Translators and ethnographers are, as we have seen, "intercultural
mediators". Their role is growing increasingly important, and what is required
is profound cultural knowledge gained by fieldwork. The methods necessary
to acquire this knowledge are numerous and cannot be analysed here. In
translation studies such models range from the so-called "cultural grammar"
(Nida 1945) to detailed training programmes for translators and interpreters
which also include perspectives of new job descriptions like "cultural advisor"
(Ammann 1990).
Let us start with the fact that ethnographer and translator are practically
2
In another article, Tyler goes even further by saying: "Because the text can eliminate
neither ambiguity nor the subjectivity of its authors and readers, it is bound to be
misread, so much so that we might conclude (...) that the meaning of the text is the
sum of its misreadings" (Tyler 1986:135).
"the first readers" of the other culture as is presented in the foreign

culture/language text. Both the translator and the ethnographer have to
represent the other in a primary process. One has to bear in mind that the term
"text" is used here in its widest sense: according to Clifford Geertz, just as
literary texts, social and cultural activities as well as events and forms of
expression can be regarded as text, as "fantasy products built by social
material". Geertz therefore defines culture as a "montage of texts" (Geertz
1973:253).
Superficially of course the process of translation seems quite different
for ethnographers and translators. The ethnographer, in a first step, has to
interpret the social discourse of his informants, i.e. he tries to find out what
they "meant" to say. In a second step, this interpretation is systematised and
textualized for a target public of the "First World". This is, as we can see, at
least a two step translation process. The translator, however, is already
confronted with a written text, which she/he also has to decode and reconstruct
in a manifold process. It would be quite revealing to go into more detail on the
overlap of hermeneutic understanding.
As far as the role of the first reader in translation is concerned, it
should be mentioned that for the translator learning to translate means
"learning to read", i.e. to produce meanings which are acceptable for the
cultural community the reader belongs to (Arrojo 1992:76). In translation
studies, the skopos theory offers an approach to solve this problem. Assuming
that the decision which skopos to choose in a particular translation context is
a cultural issue, we have to take into account the cultural conventions which
determine each translation strategy. This implies a major responsibility for the
translator, because each translated text for a target public that has no access to
the original, is the source for a different and new way of reading. The
ethnographer cannot escape this responsibility either: she/he is often the only
person who has done fieldwork in a certain region at a certain time and
translates the "results" into a (mostly Western) language and culture.
Ethnography and translation in the context of "Third" and "First World"
are thus positioned between systems of meanings which are marked by power
relations. "Translating between cultures" consequently means that "other"
meanings are transferred to (con)texts of the industrialised world which is
coined by its institutions, traditions, and its history. As far as the perpetuation
of power relations through institutional practices are concerned, both
disciplines have several studies at their disposal (see e.g. Venuti, ed. 1992).
Most of them concentrate on the discussion of (post)colonial discourses,
influenced by para-textual factors.
Michaela Wolf 129
Let me draw on Lawrence Venuti's model, which clearly reveals an

overlap of translation and ethnography. In his analysis, Venuti follows the
transformative process of Louis Althusser and compares translation with a
production process, in the course of which the raw material (the foreign
language text) is transformed through a transformative act (a certain theory)
into a product, the translation (Venuti 1986:186). The raw material as well as
the transformation process are conditioned by ideology,3 because they are both
rooted in cultural history and are both products of social forces. The reception
of this product is now essential: the less ambiguous a translated text is, the
more readable it is, and consequently the more "consumable" on the book
market. Venuti goes as far as to equate consumability with ideology, which can
be seen as an external determinant of translation. This is also im-posed by
editors and publishers partly in response to sales figures. Under these
circumstances the choice of translation strategy has ideological implications.
Talal Asad, the anthropologist who probes deeply into the matter of
power in the discursive process of translation, speaks of the "strong languages"
of industrialised countries and the "weak languages" of developing countries.
In the process of this transfer, the "Third World" not only takes on other ways
of production, and a different (Western) life style, but also changes its
language. These "half-transformed" ways of life and language favour
ambiguities which an unskilful Western translator may simplify in the direction
of his own "strong" language (Asad 1986:158).
If we follow up on these ongoing life styles and language usage in the
colonial framework, we will recognise that asymmetrical power relations will
be further aggravated by the process of translation. This is due mainly to the
fact that translations into a "Third World" language are marked by Western
philosophical terms (in the widest sense). They are embedded in powerful
discourses of historiography, literature, education, etc. (Niranjana 1992:33).
It is particularly remarkable how in the course of decolonialization this
"hegemonic apparatus" shifts from the ruling colonial power to the local elite
and their production of literature (see Mehrez 1992:123).
For the ethnographer this means that within the framework of
decolonialization, as already mentioned, local elite require an autochtonous
ethnography. On the other hand, the collaboration between ethnographers,
missionaries and colonial administrators was traditionally sponsored by colonial
3
Following Althusser, Venuti defines as follows: "ideology, a term I shall define
generally as an ensemble of social representations, values and beliefs that are realised
in lived experience and in the last instance serve the interests of a definite class"
(Venuti 1986:186).
governments. Later on, sponsoring was often taken on by multinational

organisations (Feuchtwang 1973:86, quoted by Niranjana 1992:77). Thus, in
this discipline too, "consumability" was pre-programmed.
Ethnography and translation: quo vadis?
What does this mean for representation in ethnography and translation? What
are the approaches in the two disciplines that deal with these problem sets?
What is the potential for co-operation provided that common scientific
questions can be raised? An answer to all these questions, in the framework of
this paper, can only be fragmentary and limited to a few aspects.
In the context of post-structuralism we will recognise that in the
discussion of representation, the status of an ethnographic text is being
increasingly questioned. Classical concepts like text, author or meaning are
challenged. More attention is being paid to the construction of knowledge in
the writing process. In translation studies, the principle of mimesis was
abandoned long ago, and Roland Barthes, Jacques Derrida or Jean-François
Lyotard are being increasingly used as models (cf. Arrojo 1992; Niranjana
1992; Hirsch 1995; Tyler 1986:137).
As far as text production is concerned, we see models in ethnography
and translation studies which are amazingly similar to one another. Post-
modern ethnography prefers discourse to text, dialogue to monologue.
Ethnography is now supposed to be the result of the collaboration of all
partners. The discourse to be produced should be the result of a reciprocal,
joint, dialogic process. Ideally, the product is supposed to be a "polyphonic
text". Holz-Mänttäri (1986), from the field of translation studies, believes that
the translator, in the ideal case, should contact the initiator (the one who
commissioned the translation) and the target public in order to specify the
translation project. The production of text is therefore no longer — always
ideally speaking — the sole decision of the translator/ethnographer (this is true
for both even if the former works with the author of a text and the latter
cooperates with the informant), but evolves as a joint project with the active
participation of all partners.
Here we can locate a common domain of activity, where
interdisciplinary collaboration could bring fruitful results for both partners,
with special regards to the methodological elaboration or differentiation of
discourses inspired by Western thought and to the elaboration of endogenous
cultural discourses. If we take into consideration that — still in the context of
Michaela Wolf 131
transfer between "Third" and "First World" — in ethnography as well as in

translation, interpretive, dialogic, polyphonic processes based on experience
play a role, we can detect methods of analysis which characterise asymmetrical
power relations and discover resulting discriminations. In this context Hatim
and Mason (1994:160) mention three different levels. On the lexical and
semantic level the discriminatory lexicalisation can be identified, which will
help to detect the motivation behind the use of certain lexical items. On the
syntactic level, in ideologically sensitive texts for example, generalisations can
preserve "surface similarities".
Finally, it should be pointed out that the often questioned and painfully
deficient competence of ethnographers as well as of translators will be
increased if the ethnographic result (the translated text) is discussed and
negotiated with the research subject (the author of the original text).
As a basis for translation, the analysis of culture-specific and social
phenomena of the societies involved is indispensable for the perception of the
structure and the meaning of language. This results in a clear demand for more
intensive interdisciplinary collaboration. The identification of common
problems in ethnography and translation could be a first important step in this
direction. What should be taken special note of, in case of further analyses, is
the phenomenon of the "other" in a semiotic as well as a semiologic respect
(cf. Todorov 1982). Furthermore the relationships between text-author-reader
still leave a series of questions unanswered. As far as the specific question of
asymmetrical power relations in translating between cultures is concerned, a
new concept of translation is necessary which needs to create a new awareness
of the relationship between "strong" and "weak" languages. Discourses in
different cultures are not autarchic but develop within social fields of power
and privilege. In order to detect these asymmetries, analyses of the economic
and political processes in the source and target society could be increasingly
employed for translation between cultures, which would subsequently reveal
the constraints in the production and the reproduction of texts. Such an
interdisciplinary operation with an intercultural approach must aim at finding
a new global reading for different symbolic referential systems4 which is
4
In the context of the production of postcolonial literature in Arabic countries, Mehrez
says: "It was crucial for the postcolonial text to challenge both its own indigenous,
conventional models as well as the dominant structures and institutions of the coloniser
in a newly forged language that would accomplish this double movement. Indeed, the
ultimate goal of such literature was to subvert hierarchies by bringing together the
'dominant' and the 'underdeveloped', by exploding and confounding different
symbolic worlds and separate systems of signification in order to create a mutual
interdependence and intersignification" (Mehrez 1992:122).
directed at perceiving language as an incessant production of meanings:

"Instead of aiming at domesticated transparency and hidden foreign-ness, the
translator should rather let the reader be aware of the linguistic and cultural
differences and the plurality of meanings" (Koskinen 1994:451).
This is not an elaborated model for "translating between cultures" —
and probably such a model cannot be produced at all. Many aspects have had
to be neglected, as for example the analysis of ideology which is indispensable
for the discussion of power relations. I have merely tried to develop an overlap
and some problems common in translation and ethnography, to hint at
possibilities for more intensive interdisciplinary collaboration and to recapture
existing relations of argumentation.
References
Ammann, Margaret. 1990. "Fachkraft oder Mädchen für alles? Funktion und Rolle des
Translators als Dolmetscher und Begleiter ausländischer Delegationen". In: H J .
Vermeer (ed.), 15-30.
Arrojo, Rosemary. 1992. Oficina de tradução. A teoria na prâtica. São Paulo: Editora Ática.
Asad, Talal. 1986. "The Concept of Cultural Translation in British Social Anthropology". In:
J. Clifford and G.E. Marcus (eds.), 141-164.
Barthes, Roland. 1984. Le bruissement de la langue. Paris: Seuil.
Berg, Eberhard and Fuchs, Martin (eds.) 1993. Kultur, soziale Praxis, Text. Die Krise der
ethnographischen Repräsentation. Frankfurt/Main: Suhrkamp.
Clifford, James and Marcus, George E. (eds.) 1986. Writing culture. The Poetics and Politics
of Ethnography. Berkeley: University of California Press.
Dreyfus, Hubert L. and Rabinow, Paul. 1987. Michel Foucault. Jenseits von Strukturalismus
und Hermeneutik. Frankfurt/Main: Athenäum.
Foucault, Michel. 1972. L'ordre du discours. Paris: Gallimard.
Foucault, Michel. 1976. Histoire de la sexualité, I: La volonté de savoir. Paris: Gallimard.
Geertz, Clifford. 1973. The Interpretation of Cultures. New York: Basic Books.
Göhring, Heinz. 1978. "Interkulturelle Kommunikation, Die Überwindung der Trennung von
Fremdsprachen- und Landeskundeunterricht durch einen integrierten Fremdverhaltens-
unterricht". In: W. Kühlwein and A. Raasch (eds.), 9-14.
Hatim, Basil and Mason, Ian. 1994. Discourse and the Translator. London: Longman.
Hirsch, Alfred. 1995. Der Dialog der Sprachen. Studien zum Sprach- und Übersetzungsdenken
Walter Benjamins und Jacques Derridas. München: Fink.
Holz-Mänttäri, Justa. 1986. "Translatorisches Handeln — theoretisch fundierte Berufsprofile".
In: M. Snell-Hornby (ed.), 348-374.
Koskinen, Kaisa. 1994. "(Mis)translating the Untranslatable — the Impact of Deconstruction
and Post-Structuralism on Translation Theory". Meta 39 (3), 446-452.
Kühlwein, Wolfgang and Raasch, Albert (eds.). 1978. Kongreßberichte der 8. Jahrestagung
der Gesellschaft für Angewandte Linguistik GAL. Stuttgart: Hochschul verlag.
London: Routledge.
Mehrez, Samia. 1992. "Translation and the Postcolonial Experience: the Francophone North
African Text". In: L. Venuti (ed.), 120-138
Nida, Eugene. 1945. "Linguistics and Ethnology in Translation-Problems". In: Word 1, 194-
208.
Michaela Wolf 133
Niranjana, Tejaswini. 1992. Siting Translation. History, Post-Structuralism, and the Colonial
Context. Berkeley: University of California Press.
Snell-Hornby, Mary. 1986. "Übersetzen, Sprache, Kultur". In: M. Snell-Hornby (ed.), 9-29.
Snell-Hornby, Mary (ed.) 1986. Übersetzungswissenschaft — Eine Neuorientierung. Zur
Integrierung von Theorie und Praxis. Tübingen: Francke.
Todorov, Tzvetan. 1982. The Conquest of America: The Question of the Other. New York:
Harper and Row.
Tyler, Stephen. 1986. "Post-Modern Ethnography: From Document of the Occult to Occult
Document". In: J. Clifford and G.E. Marcus (eds.), 122-140.
Tyler, Stephen. 1993. "Zum 'Be-/Abschreiben' als 'Sprechen fur'. Ein Kommentar". In: E.
Berg and M. Fuchs (eds.), 288-296.
Ulin, Robert C. 1991. "Critical Anthropology Twenty Years Later. Modernism and
Postmodernism in Anthropology". Critique of Anthropology 11 (1), 63-89.
Venuti, Lawrence. 1986. "The Translator's Invisibility". Criticism 28, 179-212.
Venuti, Lawrence. 1991. "Genealogies of Translation Theory". Traduction, Terminologie,
Rédaction TTR 4 (2), 125-150.
Venuti, Lawrence (ed.) 1992. Rethinking Translation. Discourse, Subjectivity, Ideology.
London: Routledge.
Vermeer, Hans J. 1986. "Übersetzen als kultureller Transfer". In: M. Snell-Hornby (ed.), 30-
53.
Vermeer, Hans J. (ed.) 1990. Kulturspezifik des translatorischen Handelns. Vorträge anläßlich
der GAL-Tagung 1989. Heidelberg: Universität Heidelberg.
Wolf, Michaela. 1995. "Interkultureller Transfer in entwicklungspolitischen Texten.
Überlegungen zu einer bewußtseinsorientierten Translation". TextConText 10 (1), 5-
23.
Astérix — Vom Gallier zum Tschetnikjäger:
Zur Problematik von Massenkommunikation und
übersetzerischer Ethik
Mira Kadric/Klaus Kaindl
Die Idee des Nationalismus erlebt im zu Ende gehenden 20. Jahrhundert in

Europa einen neuen Aufschwung. Vor allem in Krisenzeiten kippt das Gefühl
nationaler Zusammengehörigkeit allzu leicht ins Pathologische und macht einer
nationalistischen Haltung Platz, die alles Fremde — sei es nun ethnischer,
religiöser oder sonstiger Natur — für die bestehenden wirtschaftlichen und
politischen Probleme verantwortlich macht. Dieses Szenario kann man seit dem
Zusammenbruch der kommunistischen Regimes in Ost- und Südosteuropa
mancherorts beobachten, wobei die Staaten des ehemaligen Jugoslawien das
wohl blutigste Beispiel liefern.
Im Ge- und Mißbrauch nationalistischer Ideologien für die Gründung bzw.
Konsolidierung von staatlichen Gebilden können auch die Mittel der Massen-
kommunikation, wie Rundfunk, Fernsehen und Presse, aber auch die
literarischen Formen der Massenproduktion, wie etwa Comics, eine wesentli-
che Rolle spielen, da sie aufgrund ihrer Verbreitung und ihrer Möglichkeiten
fast alle Bevölkerungsschichten erreichen und damit auch bis zu einem
gewissen Grad beeinflussen können. Angesichts der potentiellen Bedeutung und
auch der immensen Verbreitung von Comics erstaunt es, daß sich die
Übersetzungswissenschaft diesem Gebiet bisher nur sporadisch gewidmet hat.1
Dies umso mehr, als besonders der deutschsprachige Comicsmarkt aufgrund
der relativ geringen Eigenproduktion überwiegend auf Übersetzungen
angewiesen ist, sodaß von einem krassen Mißverhältnis zwischen der enormen
übersetzerischen Tätigkeit in diesem Bereich und der kaum vorhandenen
theoretischen Auseinandersetzung gesprochen werden kann.
1
Zu den wenigen Arbeiten können Spillner (1980), Hartmann (1982), Grassegger (1985)
und Schwarz (1989) gezählt werden.
Gerade aufgrund der semiotischen Komplexität sowie der massenmedialen
Verbreitung, die aus Comics einen festen Bestandteil der Alltagskommunika-
tion machen, ergeben sich für den Übersetzer eine Reihe von Anforderungen,
die eine eingehendere Beschäftigung lohnenswert erscheinen lassen.
Elemente der übersetzerischen Kompetenz
Neben den allgemeinen Komponenten der übersetzerischen Kompetenz, wie

etwa dem analytischen Erfassen verschiedener Auftragssituationen, dem
Recherchieren und Auswerten von auftragsbezogenen Unterlagen, der
Erschließung von terminologischem Material und der Fähigkeit, übersetzerische
Entscheidungen argumentativ darzulegen, gliedert sich die translatorische
Transferkompetenz in eine Reihe von Teilfaktoren, wobei wir uns auf die
folgenden konzentrieren wollen:
— übersetzungstechnische Kompetenz — semiotische Kompetenz

— textsortenspezifische Sprachkompetenz — ethische Kompetenz
Diese Teilkompetenzen, die prinzipiell bei allen Übersetzungen nötig sind,

manifestieren sich je nach Text und Übersetzungsfall auf spezifische Weise. Im
folgenden sollen diese Anforderungen in Hinblick auf die Comics-Übersetzung
näher beschrieben werden. Als wesentliche Faktoren zur Differenzierung
dienen dabei die besondere Beschaffenheit des Textes und seine Verwendung
als kommerzielle Massenware.
Übersetzungstechnische Kompetenz
Unter der übersetzungstechnischen Kompetenz verstehen wir hier die

Fähigkeit, jene Probleme zu lösen, die formalbedingt sind und besonders
häufig bei multimedialen Texten auftreten. Als Beispiele seien hier etwa
Operntexte genannt, in denen die Silbenzahl der Wörter durch die musikalische
Einbettung determiniert wird oder Comics, bei denen sich aufgrund der
Integration des sprachlichen Teiles in Form von z.B. Sprechblasen in einen
Bildtext in der Übersetzung Platzprobleme ergeben. Häufig werden diese
Probleme bzw. die zu ihrer Lösung notwendige Teilkompetenz in der
einschlägigen Literatur überbetont, so auch von Grassegger, der in bezug auf
die Comicsübersetzung schreibt: "Für die Übersetzung ist [...] am wesentlich-
sten, daß durch die vorgegebene Größe der Sprechblasen der 'Spielraum' also
Mira Kadric/Klaus Kaindl 137
der dem zielsprachlichen Text zur Verfügung stehende Raum begrenzt ist"
(1985:11). Daß gerade diese Teilkompetenz, die zur Lösung von, was den
kognitiven Aufwand betrifft, einfachen Problemen dient, so stark betont wird,
ist wohl auch Ausdruck einer gewissen Geringschätzung der Comics-Sprache,
die von einer literarisch-ästhetischen Position heraus zumeist als "Pängsprache"
kritisiert wird. Ohne eine Wertung vornehmen zu wollen, kann sicherlich
festgestellt werden, daß sich die Sprache der Comics von anderen literarischen
Textsorten wesentlich unterscheidet, wodurch vom Übersetzer eine auf Comics
ausgerichtete textsortenspezifische Sprachkompetenz verlangt wird.
Textsortenspezifische Sprachkompetenz
Die Sprache in Comics manifestiert sich in verschiedenen Formen,2 die jeweils

unterschiedliche Funktionen erfüllen und somit auch in der Übersetzung
unterschiedliche Probleme stellen können. Während die zumeist am Bildrand
in Kästchen vorkommende Sprache die epische Komponente des Comics bildet,
durch die der Erzähler kommentieren, einführen etc. kann, stellen Texte in
Sprechblasen die gesprochene Sprache dar, wobei durch den sogenannten
"Dorn" angezeigt wird, daß es sich um die verschriftlichte Form verbal-
auditiver Sprachvorkommen handelt. Die verbale Aussage wird durch
zusätzliche Mittel ergänzt, wobei hier vor allem die typographische Gestaltung
zu erwähnen ist. Durch Schriftart, Größe und Duktusverhältnis, Umrisse,
Farbigkeit und typographische Zusätze wie etwa Piktogramme können
differenzierte Aussagen über die emotionalen Zustände der einzelnen
Comicsfiguren getroffen werden. Die Typographie im Comic trägt somit dazu
bei, "Aspekte der komplexen Kommunikationssituation eines natürlichen
Sprechaktes" (Wienhöfer, 1979:343) visuell darzustellen. Für den Übersetzer
bedeutet dies, daß er nicht nur den sprachlichen Text, sondern auch seine
visuelle Gestaltung bei der Interpretation des AT und der Produktion des ZT
beachten muß, ein Aspekt, der in der Translationswissenschaft noch viel zu
wenig beachtet wurde.3
Ebenfalls zur Gruppe der sprachlichen Zeichen sind die Onomatopoien zu
rechnen, die vor allem dazu dienen, Geräusche sprachlich zu imitieren; ihre
2
Hünig unterscheidet insgesamt 7 Kategorien sprachlicher Erscheinungen, die jeweils
unterschiedliche Funktionen erfüllen können: gesprochen dargestellte Sprache, gedacht
dargestellte Sprache, sprachliche Imitation von Geräuschen, Kommentare des Erzählers,
Etikette wie z.B. Plakataufschriften im Bild, Textbegrenzungsanzeiger und Editorials
(vgl. 1974:221).
3
Eine der wenigen Ausnahmen stellt Schopp (1994) dar.
Gestaltung und auch ihr Einsatz sind dabei kulturspezifisch. In der Übersetzung
werden diese Zeichen z.T. im Original belassen, was besonders bei Abenteuer-
und Science Fiction-Comics aus dem englischsprachigen Raum der Fall ist,
z.T. werden sie auch gemäß den zielkulturellen Konventionen vertextet, etwa
bei Comics aus dem franko-belgischen Raum aber auch bei den sogenannten
Funny-Comics wie Donald Duck und Micky Maus. Damit wird auch hier
wieder deutlich, daß der Übersetzer nicht immer nur allein für die Wiedergabe
des sprachlich Gemeinten verantwortlich ist, sondern auch darüber hinaus auch
für die in das Bild integrierte "visualisierte Akustik" (Müller, 1979:191).
Semiotische Kompetenz
Der bildliche Text im Comic erfüllt laut Oomen (1975:257) drei wichtige
Funktionen: "er bettet die Sprache in die Situation ein, er läßt die Sprach-
abläufe zum Handlungsspiel werden und er kommentiert die Abläufe." Mit
anderen Worten: Das bildhafte Element liefert über den verbalen Text hinaus
Informationen, die den sprachlichen Teil ergänzen, fortführen, widersprechen
oder bestätigen können. Dadurch wird dem Leser ein Teil der Visualisie-
rungsarbeit abgenommen. In seiner Verwendung und seinen Funktionen können
dabei sowohl Parallelen zum Film als auch zum Theater festgestellt werden.
Ähnlich wie im Film kann durch Strukturierung des bildlichen Teils, wie etwa
die Perspektivierung, Reduzierung auf das Wesentliche etc. die Interpretation
gesteuert und in eine bestimmte Richtung gelenkt werden. Gleichzeitig können
durch visuelle Zeichen das Äußere bzw. die emotionale und geistige Ver-
fassung der einzelnen Figuren charakterisiert werden. Sie können auch den
situativen Raum, in dem die Handlung abläuft, mehr oder weniger detailliert
gestalten.
Sowohl die personenbezogenen als auch die raumbezogenen Zeichen
können dabei kulturell determiniert sein. Werden diese kulturspezifisch
geprägten visuellen Elemente, wie es heutzutage oft gemacht wird, in der
Übersetzung übernommen, so erkennt der Zielleser zumeist lediglich die
graphische Darstellung. Die Information, die dem Ausgangsrezipienten
aufgrund seines soziokulturellen Erfahrungshintergrundes darüber hinaus
vermittelt wird, geht jedoch verloren. Bildliche Zeichen, die auf soziokulturelle
Inhalte hinweisen und somit mehr bedeuten als sie konkret darstellen, sind
gerade in Astérix häufig zu finden. Durch topographische, politische, kulturelle
und historische Anspielungen wird immer wieder das soziokulturelle Wissen
der Leserschaft gefordert,4 ein Wissen, über das der Leser der Übersetzung
aufgrund seines anderen kulturellen Erfahrungshintergrundes allerdings nicht
verfügt.
Ethische Kompetenz
Die hier genannten übersetzerischen Teilkompetenzen ergeben sich alle aus der
spezifischen Textgestalt des Comics. Gerade bei Comics als Medium der
Massenkommunikation spielt jedoch auch, wie wir zeigen wollen, die ethische
Kompetenz des Übersetzers eine wesentliche Rolle.
Ein wesentliches Kriterium für den Erfolg von Comics stellt das
Identifikationsmoment des Lesers mit der erzählten Geschichte dar. Guttmann
stellt dazu fest, daß "Massenzuspruch, gemeinhin Erfolg, von Comics graduell
auf der intensiven affektiven Bindung des Lesers, auf der Involvierung des
Rezipienten mit dem Comic-Geschehen beruht" (1984:35). Diese affektive
Identifikation wird für den Comicsleser durch die Darstellung der Inhalte und
Figuren in Form von Stereotypen und Klischees erleichtert. Dabei kann die
Wirkungsmöglichkeit allerdings nicht auf den rein ästhetischen Bereich
eingeschränkt werden, da es ein wesentliches Charakteristikum von Massen-
kommunikationsmedien ist, "die Ängste, psychischen Widersprüche, un-
eingestandenen Hoffnungen, [...] Allmachtswünsche und Ohnmachtserfahrun-
gen, Aggressionsbedürfnisse und Verzweiflungsgefühle" des Alltagslebens
darzustellen (Doetinchem/Hartung 1974:147). Daraus folgt, daß Mittel der
Massenkommunikation auch ideologische, politische und religiöse Werte
vermitteln bzw. den Leser in diesen Bereichen beeinflussen können. Für
Guttmann ergibt sich daraus ganz allgemein bei der Beschäftigung mit Comics
als Massenmedium auch eine "ethische Problemstellung" (1984:6). Eine solche
ist natürlich auch bei der Übersetzung vorhanden, wird allerdings unseres
Wissens nirgends systematisch thematisiert.
Eine Theoretisierung der Übersetzung sollte jedoch auch den ethischen
Aspekt translatorischen Handelns berücksichtigen, denn ethisches Verhalten ist
nicht nur privater Natur, sondern bedarf auch der theoretischen Begründung.
Dabei sind unseres Erachtens drei Aspekte zu unterscheiden.
1. Das berufliche Ethos: Dieses wird zumeist in den verschiedenen
Berufsverbänden thematisiert und betrifft einerseits die Verantwortung des
Übersetzers für die bestmögliche Arbeit, andererseits beinhalten diese Kodizes
auch Verhaltensnormen in Hinblick auf eigene Interessen des Übersetzers (wie
4
Zahlreiche Beispiele hierfür finden sich in Stoll (1974).
etwa Kostenfragen) und auf Interessen von Kollegen (z.B. Werbeverbot).
2. Die retrospektive Ethik: Diesen Aspekt findet man in der Debatte um
die Treue gegenüber dem Autor des Originals wieder. Dieses Verantwortungs-
bewußtsein des Übersetzers ist somit nach rückwärts gerichtet, es geht darum,
die Intentionen des AT-Produzenten zu wahren, nicht jedoch um die Frage,
welche Auswirkungen eine Übersetzung in einer Zielkultur haben kann. Gerade
letzterer Aspekt scheint uns jedoch für übersetzerische Entscheidungen
besonders bedeutsam und wesentlicher als die beiden bisher genannten.
3. Die prospektive Ethik: Hier geht es um die Verantwortlichkeit des
Übersetzers gegenüber den Folgen seiner Übersetzung. Übersetzen als
translatorisches Handeln impliziert somit auch die Bereitschaft, gegebenenfalls
universalmoralische Verantwortung für die Folgen seiner Handlungen zu über-
nehmen. Dazu bedürfte es jedoch einer umfassenden Ethikkonzeption, die
bisher im Bereich des Übersetzens noch fehlt,5 sodaß der Übersetzer lediglich
sein eigenes Gewissen als moralische Instanz befragen kann. Wie notwendig
eine Einbindung ethischer Fragestellungen in die übersetzungswissenschaftliche
Debatte ist, soll anhand einer kroatischen Übersetzung der Comicsserie Astérix
aufgezeigt werden.
Astérix: Vom Gallier zum Tschetnikjäger
Als Astérix 1959 in Frankreich auf den Markt kam, war das Alltagsleben vom
konservativen Geist de Gaulles geprägt. Diese Situation wird von den beiden
Autoren Goscinny und Uderzo auf satirisch-burleske Weise dargestellt. Die
verschiedenen Lebensbereiche, die in Astérix behandelt werden und von "Krieg
und Militär" über "Häusliche Beziehungen" bis zu "Kriminelles" und "Fremde
Länder" reichen (vgl. Guttmann, 1984:125f), werden mittels Sprachparodien,
Anachronismen in der bildlichen und sprachlichen Darstellung sowie der
parodistischen Darstellung von Klischeevorstellungen und Stereotypen
dargestellt.
Wenn nun in Zusammenhang mit Astérix von Ideologie zu sprechen ist,
so ist dieser Begriff hier nicht in politischem Sinne zu verstehen, sondern
meint eine Reihe von "Einstellungen, Haltungen und Wertungen, die [...] als
Anschauungsmodelle von einer großen Anzahl von Personen geteilt werden"
5
Dieses Manko einer prospektiven Ethik wird besonders deutlich, wenn etwa Vermeer die
Hauptaufgabe des Übersetzers darin sieht, "to promote the achievement of the skopos"
(1994:11), bei den ethischen Implikationen dabei jedoch vor allem den Berufsethos, d.h.
Bezahlung und Arbeitsbedingungen (vgl. 1994:13), thematisiert.
(Fisch 1979:73). Obwohl es das erklärte Ziel der Autoren war, ihr Publikum
zu unterhalten und nicht in irgendeiner Form zu beeinflussen, war es klar, daß
diese satirische Darstellung des französischen Ideensystems sehr wohl auch
politisch instrumentalisiert werden könnte.
Ein solcher Fall von (rechts)ideologischen Manipulation fand kürzlich in
Kroatien statt. Dabei ging es darum, die affektive Identifikation des Lesers mit
der Comics-Handlung für politisch motivierte Ziele auszunützen.
Bis zum Zerfall Jugoslawiens gab es für den ganzen serbokroatischen
Sprachraum (Bosnien, Kroatien, Montenegro, Serbien) eine einheitliche
Ausgabe der Comicserie Astérix. Nach der Unabhängigkeitserklärung Kroatiens
wurde diese "jugoslawische" Version durch eine kroatisierte Fassung ersetzt,
die als das "Unternehmen des Jahres des kroatischen Verlagswesens" im
kroatischen Fernsehen vorgestellt und gepriesen wurde.
Der neue Astérix ist dem "Heimatkrieg"6 angepaßt. Die Anspielungen auf
den Heimatkrieg werden zwar nicht durchgehend und konsequent gemacht,
dennoch wird versucht, den Leser durch gezielte kulturelle, sprachliche und
politische Anspielungen in seinen negativen Gefühlen gegenüber den Serben
bzw. seiner nationalistischen Einstellung zu bestärken.
Im französischen Original wird erzählt, wie die Römer versuchen, hinter
das Geheimnis der Zauberkräfte der Gallier zu kommen. Dazu entsenden sie
einen Spion in das gallische Dorf; dieser wird zwar entlarvt, kann aber
entkommen und den Römern von dem geheimnisvollen Zaubertrank des
Druiden erzählen. Daraufhin wird der Druide und in der Folge auch Astérix,
der gerade keinen Zaubertrank zu sich genommen hat, gefangen genommen.
Statt des gewünschten kräfteverleihenden Mittels braut der Druide für die
Römer jedoch ein schnellwirkendes Haarwuchsmittel. Der Lagerkommandant
verspricht ihnen die Freiheit, wenn sie ein Gegenmittel herstellen. Den beiden
Galliern gelingt es, die Römer zu übertölpeln und der gerade eintreffende
Julius Caesar kann nicht anders, als ihnen die Freiheit geben.
Die bildhaften Elemente wurden in der Übersetzung generell nicht
verändert. So stellt die Eröffnungslandkarte weiterhin Frankreich dar und auch
die kroatische Episode Asterix Gal beginnt wie gewöhnlich mit der Be-
schreibung des gallischen Dorfes. In sprachlicher Hinsicht werden jedoch
zahlreiche ideologische Manipulationen vorgenommen, von denen hier einige
beschrieben werden sollen.
Um die geheimnisvollen Kräfte der Galliern zu bekommen, entscheiden
sich die Römer, einen Krieger als Gallier zu verkleiden und ihn als Spion
6
So wird der jugoslawische Bürgerkrieg in Kroatien genannt.
einzusetzen. Er wird in Ketten gelegt und in die Nähe des gallischen Dorfes
gebracht, damit die Gallier ihn "befreien". Astérix tut dies mit dem Ausruf:
F: "Par Toutatis, on y va"! (Goscinny/Uderzo, 1961:13)
K: "Za dom i slobodu" (Goscinny/Uderzo, 1992:13)
["Für die Heimat und die Freiheit"]
Der Schlachtruf "Za dorn" wurde während des unter Hitlers Protektion
ausgerufenen "Unabhängigen Staates Kroatien" von der Ustascha, dem
kroatischen Pendant zur Waffen-SS, verwendet und entsprach dem deutschen
"Heil Hitler". Dieser Gruß wurde im Balkankrieg von den rechtsextremen
Kroaten wieder verwendet. Obwohl dieser Gruß in der Astérix-Übersetzung den
Zusatz "...i slobodu", also "... und die Freiheit" erhält, weiß jeder kroatische
Leser, daß dies die Anspielung auf den Ustascha-Gruß ist.
Nachdem Astérix und der Druide Miraculix in römische Gefangenschaft
geraten, versuchen die Römer mit Foltermethoden, aus den Galliern das
Geheimnis ihrer Zauberkraft herauszulocken:
F: Que Ton lie ce Gaulois sur cette table! Que l'on convoque le bourreau! (1961:31)
K: Vežite malog Gala na stol. Dovedite mučitelja Čedusa Seselijusa! (1992:31)
[Bindet den kleinen Gallier auf den Tisch! Holt den Folterknecht Tschedus
Seselius!]
Der namenlose römische Folterknecht bekommt in der Übersetzung den Namen
Tschedus Seselius, eine Anspielung auf den Tschetnik-Führer Šešelj. In der
Übersetzung macht man sich dabei die zufällige Korrespondenz zwischen der
bildlichen Darstellung dieser Figur und der stereotypen Physiognomie eines
Tschetniks, der in der Regel als Bartträger bekannt ist, zunutze. Der
dümmliche Gesichtsausdruck des Folterknechts läßt den verhaßten Tschetnik-
Führer dabei als lächerlich erscheinen7 (siehe Abbildung 1 im Anhang).
Nachdem sich der Druide bereit erklärt hat, den Zaubertrank zu brauen,
werden einige Legionäre in den Wald geschickt, um die für die Jahreszeit
kaum zu findenden Erdbeeren zu holen. Der Lagerkommandant kann die
Rückkehr der Legionäre kaum erwarten; als er sie erblickt, ruft er:
F: Ave, ave les enfants! Alors vous avez les fraises? (1961:33)
K: Ave, ave drugovi vojnici. No, gdje su jagode? (1992:33)
[Ave, ave Genossen Soldaten. Na, wo sind die Erdbeeren?]
7
Auch in der typographischen Gestaltung unterscheiden sich die beiden Fassungen.
Während im Französischen durch Fettdruck und Schriftgröße eine mit der wilden Gestik
des Römers korrespondierende Schriftart gewählt wurde, vermittelt die dünne Schrift der
Übersetzung nicht die Gefühlsintensität, mit der der Römer seine Äußerung macht.
"Genossen Soldaten" ist bekanntlich eine in den ehemaligen kommunistischen

Ländern verwendete Anrede, die heute — insbesondere in diesen Ländern —
verpönt ist und nur noch abfällig gebraucht wird. Damit wird unterstrichen,
daß die Feinde neben allen anderen schlechten Eigenschaften auch noch
Kommunisten sind.
Auch Wortspiele, auf denen sich u.a. die Qualität der Serie Astérix
gründet, werden in dieser Übersetzung für die ideologischen Zwecke
manipuliert. Wenn Astérix und Miraculix im römischen Lager festgehalten
werden, um den Römern den Zaubertrank zu brauen, verabreicht Miraculix
ihnen statt dessen ein starkes Haarwuchsmittel. Darauf folgt eine ganze Reihe
von spöttischen Redewendung bzw. Wortspielen, in denen Lexeme Haar/haarig
vorkommen. Wenn z.B. der verzweifelte Lagerkommandant ein Gegenmittel
fordert und dafür sogar die Freiheit für die beiden Gallier bietet, antwortet
Asterix:
F: Il a un poil dans la main!...
...Parfois il a un cheveu sur la langue aussi. (1961:42)
K: Čedo dlaku mijenja, ali ćud nikada! (1992:42)
[Tschedo wechselt sein Haar, den Charakter nie]
Bei dieser Äußerung handelt es sich um eine Anspielung auf die bekannte und
allgemein verwendete Redewendung, die lauten sollte: "Vuk dlaku mijenja, ali
éud nikada", was soviel wie "Der Wolf wechselt sein Fell, den Charakter nie"
heißt; mit andern Worten: ein Serbe bleibt immer ein Serbe.
Auch auf dem Gebiet der Realia erfolgen Anpassungen an die Zielkultur.
So wird Miraculix für die Preisgabe von Geheimnissen Ruhm und Geld
versprochen:
F: Druide, si tu parles, je ferais de toi un homme riche et puissant! — Non!
Tu auras des sesterces! Des tas de sesterces!!! — Non! (1961:24)
K: Progovoriš li, učinit eu od tebe čovjeka od bogatstva i moći! — Pih!
Imat ćeš denariusa! Brda jugodenariusa!! — Pih!
[Wenn du redest, mache ich einen reichen und mächtigen Mann aus dir! — Pff!
Du wirst Denarius haben! Berge von Jugodenarius! — Pff! (1992:24)
Bekanntlich war der Dinar die gesamtjugoslawische Währung. Nach der Un
abhängigkeitserklärung der einzelnen Teilrepubliken wurde das Thema der
Währungsänderung immer häufiger dikutiert.8 Der Dinar wurde, nicht zuletzt
wegen der hohen Inflationsrate in Restjugoslawien, zum Inbegriff der
8
Inzwischen wurde in Kroatien eine neue Währung eingeführt, die Kuna, die auch in der
Zeit zwischen 1941 -1945 die Währung des Unabhängigen Staates Kroatien war.
Wertlosigkeit und des alten, maroden Systems. Diese Auffassung wird durch
die Übersetzung — Sesterzen im Vergleich zu Bergen von Jugodenarius —
bestätigt. Und deshalb reagiert der Druide sogar auf das Angebot von "Bergen
von Jugodenarius" nur mit einem verächtlichen "Pih". Im Original antwortet
er bekanntlich mit einem "non". Der Gesichtsausdruck, vor allem die Mund
partie, ist dabei bewegungslos, seine Verärgerung wird lediglich an den gerun
zelten Augenbrauen deutlich. Da die Artikulation des verächtlichen "Pih" eine
andere Mundstellung erfordern würde, entsteht hier in der Übersetzung ein
unstimmiges Verhältnis zwischen sprachlicher und visuell wahrnehmbarer
Reaktion (siehe Abbildung 2 im Anhang).
Auch durch die Wiedergabe der gesprochenen Sprache, in der vor
liegenden Fassung in Form verschiedener serbisch-kroatischer Dialekte, wird
die Geschichte auf die Situation der Leser zugeschnitten. Im folgenden Beispiel
droht der römische Lagerkommandant den gefangenen Galliern, sie "aufspießen
zu lassen":
F: Cette fois-ci, Gaulois je vais vous faire embrocher! (1961:47)
K: A sad Gali, ću da vas pržim na roštiljče. (1992:47)
[Und jetzt werde ich euch auf dem Spieß braten, Gallier]
Um der Aussage zusätzliche Kraft zu verleihen, wird sie in südserbischem
Dialekt gesprochen bzw. geschrieben, wobei das Serbische dieser Äußerung
eine zusätzliche Dosis an Gefährlichkeit und Gemeinheit verleihen soll.
Mit diesen Beispielen sollten neben den zahlreichen Anforderungen, die
die Übersetzung von Comics an den Übersetzer stellt, vor allem die ethischen
Implikationen dieser Tätigkeit aufgezeigt werden. Von den Comicsherstellern
wurde die moralische Dimension bereits in den 50er Jahren erkannt. Verleger
in den USA, Frankreich und Deutschland verabschiedeten Ethik-Kodizes, in
denen moralische Maßstäbe für die Produktion von Comics festgeschrieben
wurden. Damit ist zwar noch keine Durchsetzungsgarantie verbunden, durch
die Institutionalisierung von ethischen Normen wird jedoch die Chance einer
Bewußtmachung und Sensibilisierung erreicht. Von einer solchen Institutionali
sierung ist man im Bereich der Übersetzung noch weit entfernt. Gerade die
Skopostheorie, die translatorisches Handeln als zielgerichtete Tätigkeit versteht,
birgt ohne eine Einbindung ethischer Aspekte die Gefahr in sich, insofern
mißbräuchlich angewendet zu werden, als sie dahin gehend interpretiert werden
kann, der Zweck heilige jedes Mittel. Um dies zu verhindern, bedarf es einer
translatorischen Ethik zum einen im Sinne einer allgemeinen Ethik auf der
Makroebene, in der die Legitimität einer anzufertigenden Übersetzung geprüft
werden und zum anderen im Sinne einer angewandten Ethik auf der Mikroebe
ne, wo es um die ethische Bewertung übersetzerischer Einzelentscheidungen
geht. Inwieweit der Übersetzer moralische Verantwortung übernehmen muß,

hängt letztlich von seiner konkreten Stellung im translatorischen Handlungs-
gefüge ab. In der Comicsproduktion erweist sich das Übersetzen als extrem
arbeitsteiliger Prozeß, wobei dem Übersetzer z.T. lediglich die Aufgabe
zukommt, eine Rohfassung zu erstellen, während die endgültige Version vom
Verlag verfaßt wird. Ethische Urteile können daher niemals pauschal gefällt
werden, sondern sollten auf der Grundlage eines übersetzerischen Ethik-Kodex
fallspezifisch von der Stellung und der damit zusammenhängenden Ver-
antwortung des Übersetzers für seine Arbeit ausgehen.
Bibliographie
Doetinchem, Dagmar v. und Hartung Klaus. 1974. Zum Thema Gewalt in Superhelden-Comics.
Berlin: Basis.
Fisch, Heinrich. 1979. "Ideologie und Ideologiekritik". In: Fischer Kolleg Sozialwis-
senschaften. Frankfurt/M.: Fischer, 13-75.
Goscinny und Uderzo. 1961. Astérix le Gaulois. Neuilly-sur-Seine: Dargaud.
Goscinny und Uderzo. 1992. Asterix Gal. Zagreb: Izvori.
Grassegger, Hans. 1985. Sprachspiel und Übersetzung: eine Studie anhand der Comic-Serie
Astérix. Tübingen: Stauffenburg.
Guttmann, Karl H. 1984. Asterix. Erfolg und Ideologie. Multiple Rezeptionsebenen als
Qualitätskriterium populärer Mediennutzung. Wien: unveröffentl. Diss.
Hartmann, Regina. "Betrachtungen zur arabischen Version von Astérix. Ein Übersetzungsver-
gleich ". Linguistische Berichte 81, 1-31.
Hünig, Wolfgang K. 1974. Strukturen des Comic-Strip. Ansätze zu einer textlinguistisch
semiotischen Analyse narrativer Comics. Hildesheim/New York: Olms.
Knigge, Andreas C. 1986. Fortsetzung folgt. Comic-Kultur in Deutschland. Frankfurt a.
M./Berlin: Ullstein.
Müller, Monika. 1979. Visualisierungs- und Verbalisierungsmechanismen in Comics. Salzburg:
unveröffentl. Diss.
Oomen, Ursula. 1975. "Wort, Bild, Nachricht". Linguistik und Didaktik 6 (24), 247-259.
Schwarz, Alexander. 1989. Comics übersetzen — besonders ins Deutsche und besonders in der
Schweiz. Lausanne: CTL.
Schopp, Jürgen. 1994. "Typographie als Translationsproblem". In: M. Snell-Hornby et al.
(eds.), 349-360.
Snell-Hornby, Mary, Pöchhacker, Franz, Kaindl, Klaus (eds.) 1994. Translation Studies —
An Interdiscipline. Amsterdam: Benjamins.
Spillner, Bernd. 1980. "Semiotische Aspekte der Übersetzung von Comics-Texten". In: W.
Wilss (ed.) Semiotik und Übersetzen. Tübingen: Narr, 73-86.
Stoll, André. 1974. Asterix, das Trivialepos Frankreichs. Bild und Sprachartistik eines
Bestseller-Comics. Köln: Dumont.
Vermeer, Hans J. 1994. "Translation today: old and new problems". In: M. Snell-Hornby et
al. (eds.), 3-16.
Wienhöfer, Friederike. 1979. Untersuchungen zur semiotischen Ästhetik des Comic Strip. Unter
der besonderen Berücksichtigung von Onomatopoese und Typologie. Zur Grundlage einer
Comic-Didaktik. Dortmund: unveröffentl. Diss.
Anhang:
Abbildung 1
Französisch:
Kroatisch:
Abbildung 2
Ethics of translation
Andrew Chesterman
Traditional discussions of ethics in translation studies have dealt with a rather

motley set of questions. These have generally been concerned either with the
duties of translators or with their rights. Typical issues have been:
(a) the general concept of loyalty, to the various parties concerned;
(b) the acceptable degree of freedom in the translating process, plus the
issue of whether translators have the right or duty to change or correct or
improve the original (cf. the ethics debate in the 1994 ITI Proceedings [Picken
1994], and e.g. Robinson 1991);
(c) linked to both these, the argument about the translator's invisibility,
understood as an ideal of neutrality or anonymity and recently challenged by
many scholars;
(d) whether translators have the right to refuse to translate a text they find
"unethical" (a right encoded in codes of translational ethics in some countries);
(e) what rights translators have regarding translations as intellectual
property, e.g. compared with the rights of original authors (e.g. Venuti 1995).
(f) There has been some discussion of the translation commissioner's
power and ideology in initiating the selection of texts to be translated: see e.g.
Lefevere (1992) on patronage. Issues have also been raised concerning the
relation between translators and the various authoritative bodies who legislate
or otherwise determine the positions to be taken within a given culture
concerning the above questions (e.g. Pym 1992).
I propose a rather different view of translation ethics, based not on the
concepts of duty or right but on that of value. Both duties and rights are
secondary notions: they depend on notions of value, which are therefore
primary. The framework I shall describe governs actual translation action after
the point when a translator has agreed to do a translation. That is, it excludes
issue (d) above, which I take to be more a matter of personal ethics generally,
not translation ethics in particular. The framework also excludes issue (e),
which comes into play after a translation has been submitted. And also
excluded are commissioner's ethics (cf. issue (f) above), and the wider
questions of cultural dominance and subservience. True, commissioner and
translator may sometimes coincide in the same person; yet in this case we can
still distinguish between the initiator's ethics of text selection and the
translator's ethics governing the act of translation itself. For convenience, I
shall include "team of translators" under the singular term "translator"; I shall
not therefore discuss here the additional ethical issues of cooperative translation
by committee.
My starting point is in deontic logic, that branch of philosophy that deals
with normative concepts. Translation studies since Toury (1980, and most
recently 1995) has been increasingly interested in norms and in applications of
the theory of action, and the approach to an ethics of responsibility which I
shall outline below fits in well with current concerns in translation theory.
Deontic logic makes a basic distinction between three levels of concepts:
praxeological concepts (concepts such as choice, decision, desire, freedom,
will, that have to do directly with actions); normative concepts (norms); and
axiological concepts (values). Deontic actions (those the agent feels "ought"
to be done) are governed by norms, and norms themselves are governed by
values. A norm, after all, is accepted as a norm because it embodies or
manifests or tends towards some value. Values are thus examples of regulative
ideas.
Actions have to do with changes in states of affairs. An action may either
be productive, in that it brings about such a change; or it may be preventive,
in that it prevents the occurrence of a change that would otherwise have taken
place. (For a formalized presentation of these and other deontic points, see e.g.
von Wright 1968.) Corresponding to these two classes of action there are also
two classes of non-action (i.e. of omission or forbearance): leaving a state
unchanged (no productive action), and letting a change take place (no
preventive action).
An analysis of the concept of change can be based on three elements: (a)
the initial state of affairs, before any action A; (b) the end state, after A has
taken place; (c) the hypothetical state, which would have prevailed if A had not
taken place.
A translator (or team of translators, passim) is someone who effects
changes of certain kinds in certain states of affairs, a decision-making agent.
The translator's task at every decision point is simply to make a comparison
between two states: the predicted end state (or: the most likely end state)
resulting from A and the hypothetical state resulting from not-A, and act
Andrew Chesterman 149
according to the result of this comparison. This comparison is a value

judgement: of two states, which comes closest to manifesting or promoting a
given value or values? Deontic logic accepts that such judgements are not
automatic, and not universal either: every deciding agent is a unique human
being in a unique personal life-situation with a unique state of knowledge and
cognition, with a unique personal history. Norms may be shared by a
community, a profession or a culture; so may their underlying values; but the
understanding and application of these norms and values in decision-making is
inescapably individual. In Gadamerian terms, we could say that this deontic
logic recognizes not only the influence of the constraint of tradition, but also
that every individual has a personal horizon. Decisions (such as translation
decisions) are not made in a void, but in a particular life-situation. (Hence, of
course, no two translations of a given text need ever be the same: cf. Quine
[1960] and the indeterminacy argument.)
So: what are the values underlying translation decisions? In deontic logic,
values are concepts that govern and underlie norms. The suggestion I shall put
forward is built on the analysis of translation norms I have proposed elsewhere
(Chesterman 1993). Assume first that translation activity is governed by four
fundamental kinds of norms:
(a) Expectancy norms: a translator should translate in such a way as to
conform to readership expectations about the translation product. Being
thus product norms, expectancy norms are logically prior to the three
process norms that follow; the point of process norms, after all, is to
guarantee that the quality of the outcome of a process meets people's
expectations.
(b) The relation norm: a translator should act in such a way that an
appropriate relation is established and maintained between the source text
and the target text.
(c) The communication norm: a translator should act in such a way as to
optimize communication, as required by the situation, between all the
parties involved.
(d) The accountability norm: a translator should act in such a way as to
be accountable to all the parties involved.
I suggest that each of these norms is governed by a primary ethical value;
"primary", because there is some obvious overlap between the values, and
because the correspondences between norms and values are not absolute but
relative. The correspondences nevertheless seem striking enough to be worth
noting, and may at least provoke further thought.
Clarity
Expectancy norms are primarily governed by the value of clarity. This is of

course an old rhetorical value, and not special to translation as such. Recall the
Gricean maxim of manner, and Leech's "Clarity Principle" (1983). Popper
([1945] 1962:307f.) takes clarity to be the basic value of any linguistic
communication, since without it rational communication in social life becomes
impossible. By this token, clarity is a genuine ethical principle, not just a
linguistic one.
True, clarity can of course be interpreted differently in different cultures;
and postmodernists may be suspicious of the concept's desirability, or even of
its possibility. Yet I will stick my neck out and claim that clarity will survive
as an ethical linguistic value long after the postmodernist textual anarchists are
dead and buried. It could be argued that one aspect of clarity is aesthetic, and
that beauty and clarity are closely related, but I will not pursue this point here.
Another related concept is that of relevance: clients or recipients of a
translation certainly expect a translation to be done in such a way that the
resultant text is relevant to their needs or expectations. Relevance, however,
is more a technical value than an ethical one; as a technical value, it pertains
both to the expectancy norms and to the relation and communication norms:
recall the wordings "an appropriate relation" and "as required by the situation"
in the definitions for the relation and communication norms, above — the
notion of relevance is already inherent there.
In terms of productive action, the clarity value may justify translation
strategies that enhance psycholinguistic processing, such as a choice of iconic
vs. non-iconic versions. In terms of preventive action, the clarity value may
justify translation strategies that seek to avoid ambiguity or unnecessary
obscurity etc. During the editing process, the clarity value is particularly to the
fore.
Paradoxically perhaps, the clarity value may also justify the breaking of
expectancy norms. Values are prior to norms, after all. And if reader
expectations concerning a particular text-type are indeed that the translation
will be clumsy, virtually unreadable gobbledygook, then a translator who
decides to over-ride these expectations and produces a clear and readable text
is breaking this norm for a very good reason: in order to approximate more
closely to an underlying value (or values).
Truth
Of my four basic norms, the relation norm is the only one that deals
exclusively with translation, for the others are of course relevant to other kinds
of communication too. The value behind this norm has traditionally been
defined as fidelity or faithfulness or loyalty, and this in turn has usually been
interpreted as some kind of equivalence — to the original text, the author's
intention, the original effect, etc. However, equivalence itself is then usually
defined in terms of identity or sameness, which has meant that the translator
is typically bound to fail to achieve it.
I emphasize that the relation norm is a linguistic one, between two texts
(not between a person and a text, as suggested by terms like fidelity and
loyalty — see below), one of which is some kind of representation of the other.
I suggest that the value underlying this relation is that of truth. Roughly
speaking, we say that something is true if it corresponds to reality. "Truth" in
this sense is a quality characterizing a relation between, say, a proposition and
a state of affairs. The proposition is not "the same as" the state of affairs it
describes, but the relation between the two can nevertheless be a true one, not
false. The truth relation has many forms: passport photos bear "a true
likeness", a report of an event can be "true", a photocopy can be "a true
copy", and so on. Similarly, translations can relate to their originals in many
different ways, each of which can be called a "true resemblance" (recall
Wittgenstein's family resemblances). A translation will be rejected by the
target community (or by the client) if it is not considered to bear any kind of
"true" resemblance to the original (or, the text in question will simply not be
called a translation).
In deontic logic, the truth relation is perhaps best represented in terms of
preventive action. In a given state of affairs, unwanted change must be
prevented as far as possible. A relation of some kind must be maintained
between the two texts, and it must be one that the receiving culture (the one
in which translations are defined as translations) accepts as being "true" in
some appropriate way. Whatever the relation is, it must not be false. (Compare
the Gricean maxim of quality, that one should forbear from saying what one
knows is false.)
Douglas Robinson's provocative book The translator's turn (1991) has a
long section on translation ethics that has precisely to do with this truth
relation. What Robinson is arguing is that there is indeed a vast variety of
relations that can validly subsist between a source text and its translation, and
that translators should be aware of the full range of possibilities. The point is
stressed also by Toury (1995), and in descriptive translation studies generally.

One of Robinson's more extreme examples is that of Jordan's Cotton
Patch New Testament, which places the text in the context of the black
liberation movement. This translation strategy means that the target text is
often what might be called "very free indeed", containing many additions and
semantic alterations. But the text has nevertheless been accepted by the target
culture as a translation of a kind, and Jordan states his aims and translational
principles quite openly. There is thus no theoretical reason why the relation
norm should not be met in this way, for the relation between source and target
has de facto been accepted as a possible true one. Mention might also be made
here of so-called feminist translations (cf. e.g. Lotbinière-Harwood 1991),
which also offer a new understanding of what can constitute a true relation.
The test of the translation pudding is always in the eating, not in some
theoretical preconception: to the extent that a translation is accepted by the
target community as being valid, the relation it establishes between source and
target is de facto one possible true one. Such acceptance may of course vary
widely across periods, cultures and sub-communities, and even between
individual readers. Translators are free to translate how they feel (to
paraphrase Robinson); but their professional and financial survival will depend
on the degree to which their mode of translating is accepted by their clients
and readers, i.e. on the degree to which translators conform to the expected
norms. This is also the pragmatic answer to the injunctions of scholars such
as Newmark that translators should always neutralize sexist language: yes, in
principle one might agree. But if a client nevertheless insists on sexist
language, you either have to bite the bullet or simply decline to translate
certain texts at all. The path taken will also depend on the status of the
translator in question, and on the status of the translator profession in the
cultures concerned.
Trust
Clarity is a value pertaining to the quality of a text itself. Truth (here) is a

value pertaining to the relation between two texts, source and target. The third
value, underlying the accountability norm, pertains to a relation between
people: trust.
In Steiner's (1975) hermeneutic motion, trust represents the first stage: the
translator (or initially the client, in fact) must trust that the original is worth
translating. This is followed by aggression (into the source text) and
incorporation (of the translation into the target culture), and finally by
restitution. This last seems to be the only point at which ethics enters the
process: restitution is necessary in order to redress the balance, to balance the
books between source and target. However, it is not clear how the translator
is to achieve this redressing of the balance: restitution comes across rather as
something that simply happens, in that the very fact of translation endows the
original text with a greater value. Steiner's translation ethics, and his view of
trust, seem unnecessarily narrow.
It is worth comparing the concept of trust with that of loyalty. Loyalty is
commonly used in two ways in translation theory: to describe the translator's
relation to the original writer or the source text, and also the relation to the
target readership (e.g. Nord 1991:29). In our deontic framework, the relation
with the source text is governed by the value of truth, which I take to be a
quality of the relation between that which represents and that which is
represented; in translation, this is an intertextual value. Trust, on the other
hand, is an interpersonal value.
One point of difference between trust and loyalty concerns the relative
status of the people involved. To be loyal to something or someone is to
maintain firm support, friendship or service. Yet this something or someone
is often understood to be "higher" than whoever is being loyal: one speaks of
being loyal to the king, of an army being loyal to the government, of being
loyal to a cause. Loyalty is commonly thought of as allegiance, as duty to a
liege or master. Its prevalence in translation studies perhaps goes back to the
days when the source text and/or its writer were raised on a pedestal above all
the other factors involved in translating, with the translator in a servant's role.
Trust, on the other hand, describes something more like a relation between
equals, and specifically between people. As a translator, I trust that the original
writer has something to say that is worth translating (cf. Steiner's point,
above); I also trust that the client will pay me; and I trust that my own readers
will read my translation in good faith, trusting in turn that there is "something
there".
More importantly, whereas loyalty is presented as a requirement of
translators alone (not the other parties in the communicative act), trust is a
value that must be subscribed to by all parties concerned. The client must trust
the translator, and so must the original writer if he or she is present; so must
the readers. Without such mutual multidirectional trust, communication fails.
In fact, trust is precisely the value which motivates loyal behaviour: one is
loyal in order not to lose trust; it is not the case that one trusts in order not to
lose loyalty. Trust is therefore the underlying and primary value here.
This is where Pym (1992) makes a central contribution to translational
ethics. Taking a sociological perspective, Pym is interested in extending the
cope of translation ethics to include broader questions, such as who decides
what shall be translated and who validates the norms governing translation
action in a particular culture; who accredits licensed translators, for instance?
Pym's basic argument is that translators have a higher loyalty than to source
or target organs: the whole accountability of professional translators is
grounded in the profession itself, in other professionals. Translators check each
other's work, drawing on past translations for guidance. They derive their
norms from the existing professional context, but the profession itself is not
bound to a particular culture. Like the international scientific community,
translators are a community that survives via its own system of checks and
balances: we validate each other. In Pym's words: "Translators' prime loyalty
must be to their profession as an intercultural space" (1992:166).
If we then ask why a translator must be accountable in the first place to
the profession, the answer is of course trust. Trust is the glue that holds the
system together. Translators, in order to survive, must be trusted as
translators. They will be trusted (a) if the profession is trusted, (b) if they are
deemed to be bona fide members of the profession, and (c) if they have done
nothing to forfeit this trust.
Trust is typically lost rather than gained: one's default reaction is to trust
someone unless events undermine this. In deontic action theory, then, we can
again make use of the concept of preventive action: we aim to translate in such
a way as to prevent a change in the default state of affairs in which trust exists.
Initiation into the profession, perhaps via a special examination or the like
(administered usually by other professionals who are presumably trusted in turn
both by the profession and by society at large), counts as establishing the trust,
and the purpose of the accountability norm is to set a standard so that this trust
will be maintained. In other words, we seek to leave the status of trust
unchanged, or at least unimpaired.
The value of trust is directly relevant to the translator's visibility. It used
to be argued, by some, that the translator should be invisible, a window
through which the original could shine unimpeded. But if you accept that trust
is one of the fundamental values of translation ethics, visibility often seems
more important than invisibility. It goes without saying that a translator's name
should always be mentioned, as a minimum degree ofvisibility; but translator's
prefaces to longer literary translations are also valuable, particularly when the
translator is seeking to challenge rather than conform to readers' expectations
(as in the examples mentioned in the previous section).
Pym associates the importance of loyalty to the profession with the aim of
translation as such. The translator profession exists in an intercultural space,
and the aim of translation is simply to improve intercultural relations. In terms
of the trust value, this could be paraphrased as "creating more trust".
Understanding
The fourth value, underlying the communication norm, is understanding. Like

the other three, I take this too as an ethical value, particularly in the light of
the analysis offered by Ebeling (1971), in which understanding virtually
becomes a manifestation of, or metaphor for, love. Like trust, understanding
also has to do with relations between people.
Ebeling offers a hermeneutic theory of translation, a theory embedded in
what he calls a theological theory of language and centred around the concept
of understanding. His starting point is the speaker's right to speak
{Ermächtigung). For a translator, this means that I start any translation task
with the question: do I have the right to translate this? And then: will my
readers trust that I have the right to translate it? (My client de facto does seem
to trust me, at least.)
The second element in Ebeling's analysis is responsibility
{Verantwortung): speakers (translators) are responsible for saying the right
word at the right time, and for knowing when to remain silent. A third key
concept is the speaker's need to challenge the hearer to understand
{Verstehenszumutung). And finally there is the understanding itself
{Verständigung), which is the goal of all communication, the value that makes
communication worth attempting: empathy, love.
In terms of action theory, we could say that the goal of a translational
action is to produce understanding, to effect a change of state from
non-understanding to understanding. Formulated thus as productive action, this
may sound trivial; it also raises queries about the possibility of complete
understanding anyway. But a formulation in terms of preventive action allows
a more realistic and also more fruitful approach. In terms of preventive action,
we seek to promote the value of understanding by reducing or minimizing
misunderstanding.
This is refreshingly non-utopian: we acknowledge that misunderstanding
cannot be eliminated entirely, that total absolute communication is impossible,
that horizons can never fuse completely, only overlap to varying degrees. This
approach is in fact an application of Popper's inverse utilitarianism: seek not
the way of greatest happiness but that of least suffering. As science proceeds
by eliminating false theories, so social advancement is best effected piecemeal,
via the gradual elimination of defects, argues Popper (e.g. 1972). As regards
the value of understanding, we might extend Popper's view to cover
"communicative suffering": a translator would then seek to minimize this,
rather than to attain some impossible ideal.
Communicative suffering is of two main types. One is qualitative, having
to do with misunderstanding as such, as a consequence of ambiguity,
obscurity, confusing style etc.; this brings us back to the value of clarity. The
other is quantitative, having to do with the number of "understanders": to
reduce misunderstanding in this sense is to increase the number of (potential)
receivers of the original message. It is often pointed out that translation
automatically extends the potential number of receivers of a message, thereby
decreasing quantitative misunderstanding. Nystrand (1992) explores this notion
further, with respect to all writers, not just translators.
One corollary of this general argument would be that translators should be
aware of whether they are translating for native or also non-native readers of
the target text: this may affect their translation strategies, choice of lexis and
idiom, etc. To exclude readers, albeit non-native readers, might even be
construed as denying them their right to understand.
The four values of clarity, truth, trust and understanding thus suggest a
basis for a fairly comprehensive translation ethics. An approach from deontic
logic further suggests that it is useful to think of translation not just as
productive action but also as preventive action. Deontic logic furthermore
offers the possibility of a formalized description, if such is deemed desirable.
Whatever the ethical framework, since translators are, by definition,
agents of change, it is instructive to wonder about the values that guide the
norms we follow. What right do we have to interfere with the state of the
world? What values shall we rank higher than the value of the trees that will
be killed for the sake of our texts?
References
Chesterman, Andrew. 1993. From 'is' to 'ought': translation laws, norms and strategies.
Target 5 (1), 1-20.
Ebeling, Gerhard. 1971. Einführung in theologische Sprachlehre. Tübingen: Mohr.
Leech, Geoffrey. 1983. Principles of pragmatics. London: Longman.
Lefevere, André. 1992. Translation, rewriting and the manipulation of literary fame. London:
Routledge.
Lotbinière-Harwood, Susanne de. 1991. Re-belle et infidèle. La traduction comme pratique de
réécriture au feminin / The body bilingual. Translation as a rewriting in the feminine.
Montréal/Toronto: Les éditions du remue-ménage / Women's Press.
Nord, Christiane. 1991. Text analysis in translation. Amsterdam / Atlanta, GA.: Rodopi.
Nystrand, Martin. 1992. Social interactionism versus social constructivism: Bakhtin,
Rommetveit and the semiotics of written text. In: A.H. Wold (ed.) The dialogical
alternative. Towards a theory of language and mind. Oslo: Scandinavian University
Press, 157-173.
Picken, Catriona (ed.) 1994. Quality — assurance, management, control (ITI Conference 7,
Proceedings). London: ITI.
Popper, Karl R. [1945] 1962. The open society. London: Routledge and Kegan Paul.
Popper, Karl R. 1972. Objective knowledge. An evolutionary approach. Oxford: Clarendon
Press.
Pym, Anthony. 1992. Translation and text transfer. Frankfurt am Main: Lang.
Quine, Willard van O. 1960. Word and object. Cambridge, Mass.: MIT Press.
Robinson, Douglas. 1991. The translator's turn. Baltimore: Johns Hopkins University Press.
Steiner, George. 1975. After Babel. London: Oxford University Press.
Toury, Gideon. 1980. In search of a theory of translation. Tel Aviv: Porter Institute.
Toury, Gideon. 1995. Descriptive translation studies and beyond. Amsterdam: Benjamins.
Venuti, Lawrence. 1995. Translation, authorship, copyright. The Translator 1 (1), 1-24.
von Wright, Georg H. 1968. An essay in deontic logic and the general theory of action (Acta
Philosophica Fennica 21). Amsterdam: North-Holland.
Part II
Translation and Beyond —

Aspects of Communication
News translation as gatekeeping
Erkka Vuorinen
Originally introduced by Kurt Lewin in his social psychological work during

World War II, the metaphor of gatekeeping was established in communication
research in the early 1950's, most notably by David Manning White (1950).1
The metaphor has frequently been used to examine the often complex route of
news texts from the initial producer (or news event) to the end user (i.e., the
newspaper reader, television viewer, radio listener, etc.) and the selections
and modifications taking place along the way. A news story's route may be
particularly complex in the case of international news. For instance, an agency
news text published in the foreign news pages of a newspaper is likely to have
passed a long chain of processing stages including 1) an international news
agency's regional bureau, 2) the agency's central bureau, 3) the national news
agency in the receiving country, and 4) the newspaper itself (Bell 1991:47).
Hence, on its way to the final recipient (reader), it may have travelled through
the computer terminals (or hands) of as many as a dozen or more people (Bell
1991:48; Rosenblum 1981:110). Consequently, as Teun van Dijk points out,
the final news items are "the ultimate results of a complex sequence of text
processing stages" (1985:6, emphasis omitted).
Generally put, gatekeeping could be defined as the process of controlling
the flow of information into and through communication channels. The
controlling function is carried out by gatekeepers located at certain strategic
areas, or gates, in the information channel. The gatekeepers decide what
messages or pieces of information shall go through a particular gate and
continue their journey in the channel and what not ("in" or "out" choices), and
1
Lewin himself also suggested that his theory of channels and gatekeepers is applicable,
among other things, to the "traveling of a news item through certain communication
channels in a group" (Bass 1969:71; see also White 1950:383; Shoemaker 1991:9).
Lewin developed the gatekeeper concept as a means of understanding how cultural
habits, specifically the population's food habits, could be influenced.
in what form and substance these messages are allowed to pass. Here is an
example of a schematic model depicting the step-by-step flow of an
international news story through a series of gatekeepers. The model was
presented by John T. McNelly in 1959:
Figure 1. The step-by-step flow of an international news story (McNelly 1959:25)
According to the model, a story (S) is written about a newsworthy event (E)
by a foreign correspondent (C1). The story (Sn) then passes through a chain of
other gatekeepers, or intermediaries (Cn), each of whom may edit, rewrite or
cut it, combine it with a related story, or otherwise shape it. The story may
also be eliminated. In addition to foreign correspondents, the gatekeepers en
route may include editors, rewritemen, deskmen, telegraph editors of
newspapers, or radio or television news editors. Ultimately, the story reaches
the receiver (R), who may pass on an oral version of the story to other people
(Rn). The broken arrows represent feedback.
Gatekeeping and translation
As may be seen, the gatekeeping metaphor strongly suggests that different

types of text manipulation occur during the passage of the text. How, then,
does translation, which is often required in international news transmission, fit
into the picture? If we, for example, consider the Translator's Charter
published by the International Federation of Translators, the kind of
manipulation procedures mentioned above, if carried out by a translator,
would seem to constitute both a legal and moral violation of the basic norms
Erkka Vuorinen 163
of translation: "Every translation shall be faithful and render exactly the idea
and form of the original — this fidelity constituting both a moral and legal
obligation for the translator." (Translator's Charter, Section 1 point 4)
Very little has been written or said about translation in the context of
international news transmission. Typically, in studies dealing with international
mass communication, translation is either completely ignored or only
mentioned in passing. In those cases where it is identified as one of several
processing operations, the implicit assumption seems to be that translation is
something different from "editing," "modifying", or "editorial selection and
processing". Consider the following quotes:
(...) much of the world's news undergoes a process far more radical than editing
within the same language. Translation between languages is a major language
function of the international agencies. (Bell 1991:66)
News processing is the handling and adapting of news copy. It consists of
copy editing, translating or modifying for local needs, heavily or lightly as
policy dictates. (Bass 1969:72, orig. emphasis)
A piece of news destined for a foreign audience typically must run an
obstacle course of reportorial error or bias, editorial selection and
processing, translation, transmission difficulties, and possible suppression
or censorship. (McNelly 1959:23)
Since none of the writers chooses to elaborate on the particular nature and
consequences of translation, it remains unclear what the translation process
actually involves. The statements quoted are also mutually contradictory in the
sense that, while Bell seems to suggest that translation results in considerable
changes, Bass and McNelly distinguish translation from other, inherently
manipulative textual operations. At the same time, the latter two also seem to
imply that translation is essentially a non-manipulative and reproductive
operation — which is, of course, fully in line with the above excerpt from the
Translator's Charter.
In his article dealing with news translation in Japan (1988), Akio Fujii
looks at translation from the gatekeeping point of view. Adopting the well-
known Westley-MacLean model of the mass communication process as his
point of departure, he revises the model to account for the translation of
Japanese news into English. The extended model is shown in Figure 2.
Figure 2. Westley-McLean model, as modified by Fujii (1988).
The model depicts a process in which a news reporter (or a news organization)
(C1) transmits to a group of receivers (B1) a message formed by C 1' s selections
from messages from a news source (A) and C !, s selections and abstractions
from objects (Xn) in his/her own sensory field (X3C1, X4C1), which may or
may not be Xs in A's sensory field. According to Westley and MacLean (Fujii
1988:33), C is one who can select the abstractions of objects X that are
appropriate to B's needs, satisfactions or problem solutions, transform them
into some form of symbol containing meanings shared with B, and transmit
such symbols by means of a channel or medium to B. Messages to C1 are
indicated by x', and messages from C1 by x". Whenever translation is
involved, a translator or translating organization (C2) steps in, and produces a
new message (x"') for a new group of receivers (B2). The new message is
constituted by C2's selections from the message from C1, and selections and
abstractions.from objects in C2's own sensory field (X3C2, X4C2). Not included
in the figure are arrows indicating possible feedback between the actors.
Drawing on an example, Fujii (1988:36) identifies four gatekeeping
functions performed by news translators:
1. Controlling the quantity of message, i.e., cutting the original;
2. Message transforming, i.e., altering the expressions of the original (e.g., by
replacing a date with a weekday);
3. Message supplementing, i.e., adding expressions/information to the
original;
4. Message reorganization, i.e., changing the structure of the original.
What makes Fujii's article somewhat confusing, however, is his conclusion
where he argues that performing the four gatekeeping functions "goes beyond
the work of mere translation," and that the operations "could well elevate the
Erkka Vuorinen 165
status of an English-language news reporter from that of a translator to at least

that of a 'copy desk'." (1988:37) In other words, Fujii suggests that
translation is somehow an essentially less prestigious occupation than other
editorial tasks and therefore has a lower status in the editorial hierarchy.
Moreover, he seems to imply that translation is, in fact, a mere transcoding
operation which does not (or should not) involve any gatekeeping procedures,
but which should reproduce the source text as faithfully as possible.
Nevertheless, in the light of the real-life example provided by Fujii,
which he uses to describe some of the modifications (such as transformations
of culture-specific references) that have been necessary to produce a
functionally adequate English-language news story out of a Japanese one, one
may ask whether the kind of "mere translation" presupposed by Fujii exists at
all, or, if it does, whether it makes any sense in a mass communication
context? In other words, why postulate a form of translational activity that
would not really serve any purpose at all?
Shoemaker's multilayered model of gatekeeping
It would seem that the problems pertaining to Fujii's approach are to some
extent connected with the gatekeeping model adopted. Namely, the Westley-
McLean model is rather an abstract one and pays little explicit attention to the
various external factors governing the gatekeeping process. To account for this
aspect of gatekeeping, there is a more recent — and concrete — model
available, presented by Pamela J. Shoemaker in her book on various strands
of theory and research in gatekeeping (1991).2 Shoemaker discusses
2 Shoemaker's model also seems to avoid some of the shortcomings that earlier
gatekeeping models have been criticized for. The early gatekeeping studies were in some
disagreement over the proper level of analysis. Some scholars, like David Manning
White (1950), stressed the role of the individual gatekeeper's personal values and
subjective decisions in the process, while others, like Walter Gieber (1956), saw the
technical and organizational constraints of news production as more important governing
factors. According to O'Sullivan et al. (1994:126f.), in most gatekeeping studies the
pressures influencing or prejudicing the gatekeepers' decision process have been seen to
stem from 1) the gatekeepers' subjective value system, likes and dislikes; 2) their
immediate work situation; and 3) legal, commercial and bureaucratic controls
constraining their work. The phenomenon is, however, more complex than that, which
is probably why the gatekeeper concept has sometimes been rejected as "oversimplified
and of little Utility" (O'Sullivan et al. 1994:126f.), or "sociologically inadequate and as
implying a passivity alien to journalism as a process of construction" (Schlesinger
1992:308).
gatekeeping theoretically at five different levels of analysis: the individual
level, the communication routines level, the organizational level, the
extramedia and social/institutional level, and the social system level. Drawing
on her discussion, she (1991:70-75) presents a new, more comprehensive
multilayered gatekeeping model, shown in Figures 3a, 3b and 3c. It should be
noted that the models seen in Figures 3b and 3c are not independent models
but enlargements of portions of Figure 3a.
Figure 3a. Gatekeeping between organizations (Shoemaker 1991:71).
Figure 3a depicts the overall process of gatekeeping between communication

organizations. The circles in the figure represent individual gatekeepers, who,
in this case, function in "boundary roles," i.e., they interact with other
organizations and outside people, filtering inputs and outputs (see Shoemaker
1991:17, 56-57). The vertical bars in front of the gatekeepers are gates, and
the arrows on both sides of the gates are forces that affect the entrance of a
message into the gate and its journey onward. One or more channels lead to
and from each of the gates and gatekeepers. Each channel may carry one or
more messages or potential messages. The squares in the middle represent
communication organizations, such as wire services, newspapers, etc., and the
small rectangles on top are social and institutional factors governing the
process. The whole process is embedded in social system ideology and
culture.
Erkka Vuorinen 167
Figure 3b. Gatekeeping within an organization (Shoemaker 1991:73).
Figure 3b is a detailed version of gatekeeping within an organization. As the

figure shows, within complex organizations the gatekeepers acting in boundary
roles may pass selected messages on to one or more internal gatekeepers.
These internal intermediaries may then make their own selections and shape
a message in a variety of ways, after which the processed message is sent
further to boundary role gatekeepers for final selection, shaping and
transmission. For instance, a staff translator working within the organization
of a newspaper may act as an internal gatekeeper. S/he receives a text
selected, and possibly pre-edited, for translation by a journalist working in a
boundary role, and having produced a target text, s/he transmits it for further
processing to another internal gatekeeper or to a boundary role gatekeeper for
final transmission to the audience (Vuorinen 1990:123ff.). Depending on how
the work is organized, translation may also be carried out by a gatekeeper
acting in a boundary role.3

Gatekeeping within an organization is embedded in the communication
routines and organizational characteristics of the organization.
Figure 3c. Intraindividual gatekeeping processes (Shoemaker 1991:74).
Figure 3c is a further enlargement of 3a, demonstrating intraindividual

gatekeeping processes. It identifies different psychological processes and
individual characteristics that can affect gatekeeping. The intraindividual level
is embedded in the individual's life experience.
Summarizing the complex gatekeeping process, Shoemaker writes,

The individual gatekeeper has likes and dislikes, ideas about the nature of
his or her job, ways of thinking about a problem, preferred decision-
3
The "groupthink phenomenon" included in Figure 3b refers to a "mode of thinking that
people engage in when they are deeply involved in a cohesive ingroup, when the
members' strivings for unanimity override their motivation to realistically appraise
alternative courses of action..." (Janis 1983, as quoted in Shoemaker 1991:28). The
symptoms of groupthink may include overestimation of the power and morality of one's
own group, closed-mindedness, and pressures on the group's members towards
uniformity (Shoemaker 1991:29).
Erkka Vuorinen 169
making strategies, and values that all impinge on the decision to reject or
select (and shape) a message. But the gatekeeper is not totally free to follow
a personal whim; he or she must operate within the constraints of
communication routines to do things this way or that. All of this also must
occur within the framework of the communication organization, which has
its own priorities but also is continuously buffeted by influential forces
from outside the organization. And, of course, none of these actors — the
individual, the routine, the organization, or the social institution — can
escape the fact that it is tied to and draws its sustenance from the social
system." (1991:75; cf. also Boyd-Barrett 1980:73)
What becomes evident on the basis of Shoemaker's model is that translation
which takes place in an institutional setting cannot be examined as isolated
from the whole individual, institutional, social, and cultural framework
surrounding it. On the contrary, it is an inseparable part of this framework,
and, at the same time, it is governed by a multitude of factors both internal
and external to the organization/institution in question. Consequently,
distinguishing translation from other text processing operations, such as
editing, or postulating, as Fujii does, a "mere translation" or "translation
proper" which aims at a total reproduction of the source text, seems to be
arbitrary in the sense that it does not take into account the complex
interrelatedness of the various processes and factors (cf. Delabastita
1989:214).
But is it translation?
Naturally, it could still be argued that the processing of foreign-language

news, if it results in considerable changes as to the form and substance of the
original texts, should not be termed 'translation' at all. In fact, such a claim
is not uncommon among practitioners in the field. For instance, several
Finnish students who have interviewed journalists handling international news
material at the Finnish News Agency have reported that the journalists do not
seem to perceive their work as translation but as "editing" or "production of
Finnish stories based on foreign items" (e.g., Kukkonen 1989:3; Offor
1993:35). The attitudes seem at least partly to be connected with the questions
of prestige and hierarchy referred to above, i.e., with perceiving editing as a
process that involves active decision-making and creativity, whereas translation
is seen as passive or even slavish imitation. The persistent twist in the logic
underlying such a distinction lies in seeing news text production in general as
a series of operations that responds to the various requirements and constraints
set by the communication situation, while translation — or the mythical

"translation proper" — is regarded as an operation responding first and
foremost to the source text,4 irrespective of the functional qualities of the
result.
It is, however, highly questionable whether this kind of pure (or "mere")
translation exists at all in professional translation. Rather, the notion would
seem to be an abstraction that is useful for the purposes of, for instance,
communication and journalism scholars, as well as practicing journalists in
their attempts to define and describe journalistic activities. More specifically,
it may be in the journalists' interests to posit a (low-prestige) "mere
translation" so as to help them to define their own position within the
professional and organizational power structures.
In any case, as Dirk Delabastita points out, such a narrow definition of
translation, is "in danger of being applicable only to very few, well-selected
cases, and of being unsuitable for a description of most actual facts"
(1989:214). To avoid such a narrow definition, it may be assumed that the
kind of textual operations commonly connected with editing or rewriting may
take place as part of any translation assignment, too. That they may be on
average more pronounced in news translation than, say, in the translation of
legal documents, has to do with the particular production conditions and goals
(the gatekeeping framework) characteristic of news translation. The difference
is, however, only a quantitative, not a qualitative one. Accordingly, I propose,
contrary to Fujii, that various gatekeeping operations, such as deletion,
addition, substitution, or reorganization, be considered part and parcel of the
normal textual operations performed in any translation, and particularly in
news translation, in order to produce functionally adequate target texts for a
given use.
References
Bass, Abraham Z. 1969. "Refining the 'Gatekeeper' Concept: a UN Radio Case Study".
Journalism Quarterly 46, 69-72.
Bell, Allan. 1991. The Language of News Media. Oxford UK & Cambridge MA: Blackwell.
Boyd-Barrett, Oliver. 1980. The International News Agencies. London: Constable.
Delabastita, Dirk. 1989. "Translation and Mass Communication. Film and T.V. Translation
as Evidence of Cultural Dynamics". Babel 35 (4), 193-218.
4
Cf. also Brian Mossop (1990:351), according to whom "The existing literature on
translation most often refers to the process of finding equivalents as if this were
something done by the translator as an individual rather than the translator as an agent
of an institution. "
Erkka Vuorinen 171
Fujii, Akio. 1988. "News Translation in Japan". Meta XXXIII (1), 32-37.
Gieber, Walter. 1956. "Across the Desk: A Study of 16 Telegraph Editors". Journalism
Quarterly 33, 423-432.
Kukkonen, Tiina. 1989. Translatorinen ja journalistinen toiminta Suomen Tietotoimiston
ulkomaantoimituksessa. Unpublished MA thesis. Tampere: University of Tampere,
Department of Translation Studies.
McNelly, John T. 1959. "Intermediary Communicators in the International Flow of News".
Mossop, Brian. 1990. "Translating Institutions and 'Idiomatic' Translation". Meta XXXV (2),
342-355.
O'Sullivan, Tim; Hartley, John; Saunders, Danny; Montgomery, Martin; Fiske, John. 1994.
Key Concepts in Communication and Cultural Studies. 2nd edition. London and New
York: Routledge.
Offor, Marja-Riitta. 1993. Kulttuurispesifinen adaptaatio uutissähkeiden kääntämisessä.
Unpublished MA thesis. Turku: University of Turku, Department of Translation Studies.
Rosenblum, Mort. 1981. Coups and Earthquakes. Reporting the World for America. New
York, etc. : Harper Colophon Books.
Scannell, Paddy; Schlesinger, Philip; Sparks, Colin (eds.) Culture and Power. A Media,
Culture & Society Reader. London/Newbury Park/New Delhi: Sage.
Schlesinger, Philip. 1992. "From production to propaganda". In: P. Scannell; P. Schlesinger;
C. Sparks (eds.), 293-316.
Shoemaker, Pamela J. 1991. Gatekeeping (Communication Concepts 3). Newbury Park: Sage.
The Translator's Charter, published by the International Federation of Translators.
van Dijk, Teun A. 1985. "Introduction. Discourse Analysis in (Mass) Communication
Research". In: T. A. van Dijk (ed.), 1-9.
van Dijk, Teun A. (ed.) 1985. Discourse and Communication. New approaches to the analysis
of mass media discourse and communication. Berlin: de Gruyter.
Vuorinen, Erkka. 1990. Kääntämistyön ohjautumisesta sanomalehden toimitusprosessissa —
esimerkkitapauksena Turun Sanomat. Unpublished MA thesis. Turku: University of
Turku, Department of Translation Studies.
White, David M. 1950. "The 'Gate Keeper': A Case Study in the Selection of News".
Advertising — A five-stage strategy for translation
Veronica Smith/Christine Klein-Braley
In the training of translators, very little attention appears to be paid to the

translation of advertisements. Yet this type of translation is far more
widespread than is immediately obvious. Why is this?
Attempts by companies to transcend national markets have led to an
increasing globalisation of products and services. Companies operating
worldwide, however, may need to adapt either their products or services, or
their marketing strategies, to win and hold on to new markets. Hence "think
globally, act locally" has become a slogan for companies with cross-border
markets. Adapting an advertising campaign to meet the conditions of a
regional market creates problems: the costs of mounting advertising campaigns
are so high that budgets may not permit individual campaigns for each
separate market: "Because photography, artwork, television production and
color printing are very costly, performing all of these in one location and then
overprinting or rerecording the voice track in the local language saves
money." (Wells et al. 1992:681)
More important, companies have discovered the power of brand images
as a communication strategy and want to capitalise on them. Hence global
companies often use the same visual images worldwide, either in print or
video format, with accompanying texts tailored to local needs. For this reason
it is very common to find apparently identical advertisements in a number of
different cultures.
For translation studies, the translation of advertisements provides us with
a microcosm of almost all the prosodic, pragmatic, syntactic, textual, semiotic
and even ludic difficulties to be encountered in translating (cf. Smith and
Klein-Braley 1985:81ff.). By analysing such short but complex and
structurally complete texts we can derive valuable insights into possible
strategies and methods for dealing with these phenomena in other longer texts,
whether literary or non-literary.
Our aim in this paper is to develop a taxonomy of strategies for the
analysis of translated advertisements. We have restricted ourselves to print
advertisements in this contribution so that the strategies can be exemplified.
Nevertheless our framework of analysis can be applied with minor adjustments
to other media. Our examples are taken from publications in English and
German.1
Setting the scene
To the consumer, advertising is a very high profile activity but in fact in terms
of the overall marketing strategy, it represents only one very small step in this
process which involves: "planning and executing the conception, pricing,
promotion, and distribution of ideas, goods, and services to create exchanges
that satisfy individual and organisational objectives." (American Marketing
Association 1985:1)
The relatively minor role that text plays within the context of the whole
marketing communication process is probably the reason why the major
handbooks of advertising devote very little attention to matters of language.
The translation of advertisements receives virtually no mention at all.2 This,
in turn, is probably the reason why the language in some international
advertising is very poor (cf. Klein-Braley and Franklin 1989; Snell-Hornby
1992): there appears to be little awareness of the importance of checking
culturally dependent discourse and pragmatic features in the translation as the
final step in the production of an advertisement in another language.
The translation is more likely to be poor if the prestige, importance or
dissemination of the text is low: in-house literature, hotel brochures and tourist
information material frequently suffer from major linguistic howlers, as
anyone working in the field of translation studies can testify. These trans-
lations are not the result of the best professional practice. When we get to the
1
It is important to point out that English may well need to be modified depending on
whether the target market is the USA, UK, Australia etc.. Similarly the German
language market is not a monolithic one, but consists of Austria, Germany and
Switzerland. A comparison of two tubes of a particular brand of toothpaste (Elmex)
purchased in Germany and in Austria showed that although the colour scheme and
appearance of the tubes was identical, the information offered in the texts differed
considerably.
2
The translation of brand names is an important issue. Rolls Royce could only with
difficulty be persuaded not to name a new car "Silver Mist" (German Mist = manure).
There are many more examples in the literature (cf. Ricks 1983; de Mooij 1994:109).
Veronica Smith/Christine Klein-Braley 175
highest level of visibility, however, i.e. the international newspaper and

magazine market, the language work is usually impeccable (except for
inevitable misprints). Indeed, most firms seem to follow the recommendations
(cf. de Mooij 1994) that advertising texts must be produced by native speaker
copywriters or copywriter/translators whose expertise goes beyond straight
translation.
It is precisely the fact that these texts are produced by highly qualified
experts which make them so interesting to analyse: it is legitimate to assume
that the decisions taken with regard to such features as text fidelity, translation
of jokes, puns, metaphors, transfer of cultural content have been made
consciously and explicitly. It is easy to shoot down the inept and inexperienced
translator, but much more rewarding to evaluate the performance of the top-
flight professional. Moreover these texts can be assessed as authentic texts of
language X in their own right. They may have started life as translations, but
they have to sell their products as original advertisements.
What is advertising? One definition (Alexander 1965:9) states:
"Advertising is defined as any paid form of non-personal communication about
an organization, product, service, or idea by an identified sponsor."
Products differ in the advertising strategies they demand. Advertisements
are usually geared to one of two generally accepted response dimensions,
namely thinking versus feeling, or cognitive versus emotional response. A
cognitive strategy based on the presentation of information is used for products
and services such as purchasing a new car or selecting an insurance policy
where rational thinking and economic considerations prevail. Other products
at a more mundane level using an informational approach are household items,
such as washing powder, toilet paper, and food.
Emotional or affective strategies are used for purchases such as cosmetics
and jewellery and also those which appeal to immediate sensory gratification:
cigarettes, alcohol, sweets. For these types of products psychological and
emotional motives such as fulfilling self-esteem or enhancing one's ego would
be stressed in the advertising. In general, emotional appeals are intended to
evoke positive mood states and positive attitudes to the product. They are also
remembered better (Belch and Belch 1992:353ff.).
The German and UK markets
These approaches need to be modified in accordance with the target market for
the advertisement. Legal restrictions, for instance, may be in operation. In
Germany Diet Coke has to be sold under the name of Coca Cola Light
because the word diet can only be used for products which fulfil certain
medical requirements (Belch and Belch 1992:746). Advertisements comparing
two products are not permitted in the German market. Different codes of
behaviour may operate: for instance, nudity is more acceptable in some
cultures than it is in others (Belch and Belch 1992:764).
One linguistic difference which must be taken into consideration is the
degree of context demanded by the language in question. Wells et al. explain
how differences between low-context and high-context cultures can influence
textualisation:
Advertising messages constructed by writers from high-context cultures might be
difficult to understand in low-context cultures because they do not get right to
the point. In contrast, messages constructed by writers from low-context cultures
may be difficult to understand in high-context cultures because they omit
essential contextual detail. (Wells et al. 1992:675)
Although Wells et al. label English as a mid-context language, and German as
a low-context one, the degree of contextuality needed in translating advertise-
ments seems to change according to circumstances. The following
advertisement for Heineken beer was very successful in the British market:
Brewers don't have to be good talkers (Exhibit 1 : HEINEKEN)
It did not, however, enjoy the same success outside the English market
because the translators failed to add context in the translated versions (Belch
and Belch 1992:742).
A German translation reading "Brauer müssen nicht gut reden" leaves
much open to the imagination. The reason is that the intonation, which
disambiguates the spoken text, is not available to the consumer reading the
print advertisement. One might ask why the marketing experts allowed
translations of this advertisement to be printed without considering this point,
but since they knew the English original they were judging it from an
"insider" position. Moreover they probably heard the text spoken. A
translation for the outsider needs to change the text to include the
"intonation", producing something like "Braumeister müssen nicht auch noch
gut reden können". This conveys the joke of the original - "provided their
beer is good enough" (cf. Smith & Klein-Braley 1985:62ff.).
The reverse situation is encountered with an advertisement placed in the
Economist by Kommunalverband Ruhrgebiet (Exhibit 2) in which the English
reader needs more context. Although disjunctive syntax is typical for
advertising language (Leech 1966:113), basic disambiguating information is
needed in Exhibit 2 in order to establish the theme-rheme organisation and

overall coherence for the reader: What are the problems for which solutions
are needed? Which are the questions for which answers will be provided?
Why is it a good thing to have so many universities etc.? What is the focus?
In particular, why should the addressee have an interest in becoming part of
it?
WHERE LIFE MEETS BUSINESS: IN THE NORTH-WEST OF GERMANY

power POOL
POOLING SCIENCE AND INDUSTRY: THE RUHR
Solutions for energy supply, environmental protection and economy
issues are needed the world over. Research carried out here in the Ruhr
provides the answers. There can hardly be another industrial region in
the world which has 15 universities, 48 research centres and 17
technical institutions. A focus which stimulates communication and
synergy. Why aren't you part of it?
Exhibit 2: KVR
These examples (Exhibits 1-2) imply less that languages differ significantly in
the degree of contextualisation needed in the absolute, but rather that the
translation process usually involves adding context for the L2 target audience.
This is why, as de Mooij (1994:213), points out, translations are generally
longer than the original text: "A television commercial with lengthy copy will
be difficult to adapt: translation of an English text into French requires on
average a 15% increase in time; if it is translated into German this figure rises
to 50%."
The need to take this expansion of text into consideration often leads to
compression of the message in the translated version. At the same time the
focus of information often shifts as we see in Exhibit 3. The text of these
advertisements for Canon photocopiers is accompanied by a "Peanuts-style"
strip cartoon of two children bragging about what their respective fathers have
achieved in terms of status and office equipment. The English body copy,
which is assumed to be the original, is considerably longer than the German
version. The brevity and conciseness of the German version can be explained
partly by the high/low context phenomenon. The English copy: "Both colour
copiers ensure the highest overall colour quality available while dramatically
increasing the value of everything you communicate on paper" is adequately
rendered in German as "... setzen mit ihrer bisher unerreichten Farbqualität
neue Maßstäbe". This example also illustrates the varying perspective of the
consumers and their needs. The English advertisement presents an evaluation
of how the product benefits will positively affect the company's
communication with the world outside, and maximise productivity inside,
whereas the German advertisement assumes the consumers will be in a
position to evaluate the benefits for themselves, so no evaluative statements
are required. The English advertisement contains direct affective consumer
appeals: "your disposal", "your productivity", "your company" and reassures
the consumers that the product is "user-friendly", even adding the information
that the paper cassettes have a high capacity, implying that once the cassette
has been filled in the morning, there will be no need to worry about
incompetent, non-technically-minded staff having to refill it during the day.
The German advertisement in contrast focuses on information, including
more details which are not available to the English-speaking consumers, and
highlighting what is perhaps the most innovative feature of the copier, its
capacity for double-sided copying as an additional underlined rubric:
"Weltpremiere beim ..."
This advertisement also reveals a second difference which is important in
translating advertisements, namely the expectation of the target language
consumer as to what advertising should be like. Empirical studies showed that
British and German consumers characterised their own national advertising
differently, thus demonstrating that expectations differed:
The British described their commercials as primarily humorous, entertaining and
emotive but relatively low on information, understandability and credibility. [...]
Germans view their commercials as relatively less humorous, although
entertaining in an emotional way. [...] Germans also regard their commercials
as reasonably informative, (de Mooij 1994:241)
Whilst humorous appeals are common in the United States and Britain, they are
not used often in Germany, where consumers tend to be more serious and do not
respond favorably to humor in advertising. (Belch and Belch 1993:763)
We can predict that humour might have to be added to an advertising
campaign presenting a German product to the English market, and that more
and possibly more precise information might need to be added for English
product being introduced to Germany. In other words, the approach would
generally focus on an emotional/affective appeal for the English market and on
an informational/cognitive appeal for the German market.3
1
Lack of space prevents us from demonstrating this with the advertisment for a "high
involvement" cosmetic product originally included in the conference presentation.
CANON INTRODUCES THE LAST WER KANN BEI FARBCOPIERERN

WORD IN COLOUR COPYING DAS LETZTE WORT HABEN?
Canon präsentiert zwei neue

Canon introduces t w o new colour Colour Laser Copier, die eine
copiers that are now a generation Generation voraus sind: sowohl
ahead. The CLC 700 and CLC der CLC 700 als auch der CLC
800 put a totally new technology 800 setzen mit ihrer bisher
at your disposal. unerreichten Farbqualität neue
Maßstäbe.
Both colour copiers ensure the
highest overall colour quality
available while dramatically
increasing the value of everything
you communicate on paper.
Working to further maximize your

productivity is: *Sie werden auch mit Karton
- the world's first automatic fertig
colour copier that copies on t w o und
sides *schaffen eine
- highspeed: 7 copies per minute Höchstgeschwindigkeit von 7
and an optional sorter Seiten pro Minute in Vollfarbe.
- frontloading cassettes with high *Beide Geräte können am PC,
paper capacity Mac und Netzwerk als Drucker
- thick paper copying: for reports und Scanner verwendet werden.
and presentation covers.
Weltpremiere beim CLC 800:
Both models are connectable to Als erster Farbcopierer der Welt
any PC or Mac and are user- copiert er automatisch
friendly enough for anyone in your doppelseitig. Und das muß man
company to work with. uns erst einmal nachmachen.
To find out more, please contact Wenn Sie mehr über Farbcopierer
your Canon distributor. von Canon wissen wollen, rufen
Sie 0222/68 36 41-533
A GENERATION AHEAD EINE GENERATION VORAUS
Canon Canon
A pleasure to work with kann's
Exhibit 4: CANON
An advertisement for gardening equipment (Exhibit 4) gives an idea of

how this works in practice. It has a picture of an elderly gentleman half lying
on top of an attractive young blonde woman, both entangled in an unruly
garden hose. This advertisement, presenting a German product to the English
market, exploits the British penchant for slapstick humour. At the same time
it capitalises on both positive and negative stereotypes which British people
hold about Germans, namely the high quality of their engineering (de Mooij
1994:154) and their lack of a sense of humour, encapsulating them both in the
metaphor "serious" in the slogan. Obviously, this advertising campaign has
been designed with the British market in mind, and could not easily be adapted
for other markets.
A GARDENA HOSE REEL NEATLY SIDESTEPS THE OH-SO-FUNNY "I'VE

TRIPPED AND FOUND MYSELF IN A COMPROMISING SITUATION"
SITUATION.
Untidy garden hoses have been known to make some people fall about.
But please excuse our German designers if they don't laugh. You see
they've just expended a lot of time and effort perfecting our range of
garden hose reels. They worked with only the finest, most resilient
grades of thermoplastics to ensure the reels were light yet immensely
strong. They gave each one an ergonomically designed carrying handle
and, where necessary, a low centre of gravity to improve stability.
Whilst designing an ingenious device that guides hoses round corners
without snagging, they even found time to calculate the appropriate
crank length for effortless hose rewinding. And the exact angle at which
to set the central hose connector to avoid kinking. (It's 45 degrees if
you're interested). Of course a range of hose reels designed to store
hose with maximum efficiency provides little opportunity for old-
fashioned British humour. Trust the Germans to take all the fun out of
gardening.
GARDENA
II Serious gardening equipment from Germany
Exhibit 4: GARDENA
A comparison of advertisements originating in a third language demonstrates
each country's favoured styles. These two advertisements (Exhibit 5) are based
on a French original. They form one of a series announcing improvements in
the services for business passengers. The important sememe RIGHT occurs
twice in each text. In the English version, containing THE RIGHT TO PRIVACY
and AIR FRANCE INTRODUCES PASSENGER'S RIGHTS, the word is used both
times with the meaning of entitlement. The German copywriter does not
succeed in transmitting this theme; s/he has attempted to make a linguistic

joke in BEI Am FRANCE SIND SIE MIT RECHT FLUGGAST but the idea of the
passenger's general privileges - and thus the overall impact of the campaign
- is lost here (as well as in the other advertisements in this series). Only the
privilege of privacy in the punch line IHR RECHT AUF DISTANZ is adequately
conveyed.
The advertisement is based on a visual of a vampish looking woman, and
the tenor of the text is that the seats (unfortunately) are too far apart for the
traveller to get as close to her as he might wish (sexism!).
AIR FRANCE AIR FRANCE

(picture of vampish lady) (picture of vampish lady)
IHRE CHANCEN STEHEN SCHLECHT,
THE CHANCES OF HER BEING SEATED DASS SIE NEBEN IHNEN SITZT. IHREM
NEXT TO YOU ARE SO SLIM THAT YOU KOMFORT ZU LIEBE HABEN WIR DEN
WON'T REGRET THE EXTRA SPACE ABSTAND ZWISCHEN DEN SITZEN SPÜR-
BETWEEN OUR SEATS BAR VERGRÖSSERT
L'ESPACE EUROPE L'ESPACE EUROPE
We know how hard it is for Geschäftsreisende wollen im Flug-

business travellers to have to zeug Akten studieren, Zeitung
concentrate on their work while lesen oder sich in Ruhe auf eine
waging the eternal battle of the Sitzung vorbereiten. A m liebsten
armrest, so we have rearranged ohne Tuchfühlung zum Nachbarn.
the space between our L'ESPACE Oder zur Nachbarin. Darum haben
EUROPE seats. Where there used wir unsere L'ESPACE EUROPE von
to be rows of three seats, there Grund auf neu gestaltet. Größer,
are now t w o seats separated by a schöner, bequemer und vor allem
little table. Your seat is much mit viel willkommener
wider, more comfortable and the Ablagefläche zwischen den Sitzen.
total space more conducive to a Für viel Ellbogenfreiheit beim
little privacy. Now, when you take Lesen, Essen und Entspannen
a seat in one of our planes, you genau die richtige Distanz. Und
take your seat in space. auch für ein anregendes Gespräch.
(picture of seats with table) (picture of seats with table)
THE RIGHT TO PRIVACY IHR RECHT AUF DISTANZ
AIR FRANCE INTRODUCES BEI AIR FRANCE SIND SIE MIT

PASSENGER'S RIGHTS RECHT FLUGGAST
Exhibit 6: A I R FRANCE
The humour in both versions is based on this situation and it does come
across, although at different points: in the headline "The chances of her being
seated next to you ..." / "Ihre Chancen stehen schlecht", and later in the
German "Oder zur Nachbarin", which makes a linguistic joke impossible in
English.
The German picks up the theme of the visual at the end of the body copy
"Und auch für ein anregendes Gespräch" but the English at this point has the
inept "you take your seat in space".
It is interesting to note that at two points in the adaptation of the
advertisement for the German market we have more precise specification of
activities than in the English version. The English travellers are going to
"concentrate on their work" while the Germans are going to "Akten studieren,
Zeitung lesen oder sich in Ruhe auf eine Sitzung vorbereiten". Moreover the
Germans need space to "lesen, essen und entspannen"; there is no attempt to
inform the English travellers why they need space. The English advertisement
also contains more direct affective consumer appeals expressed by frequent
repetition of "you" and "your" in the body copy; this direct address is
restricted to the headline and slogan in the German advertisement. A close
study of both texts also reveals quite a number of small ineptitudes. This
advertisement clearly sticks too close to the original.
Strategies for translating advertisements
In this paper we have only been able to use a very small sample of the
material we have collected. It is possible, however, to group the approaches
to the problem of translating advertisements into five broad categories.
Don't change advertisement: retain both graphics and text
This strategy can be adopted where the brand name is very strong, and the
product needs very little verbal support. Prime examples of the no-change
strategy are those products which adopt an affective approach, e.g. perfumes,
cigarettes, alcohol, soft drinks, jeans, CDs etc. The main targets for this type
of global advertising are businessmen and young people (de Mooij 1994:200
ff.).
Export advertisements: play on positive stereotypes of the originating culture,
retaining logo, slogan etc. in the original If necessary, have additional copy
in target language.
This approach is used for many of the same products as above. Here the
cultural origin of the product is felt to be an asset, and therefore needs to be
stressed in the advertising. At the same time an additional appeal is addressed
in the target market language copy.
Straight translation
At the level of international marketing this, the most obvious strategy, seems
to be the least frequent and the least preferred. It is unsuitable because it fails
to adjust to the cultural demands of the new market. It is precisely this
technique which is found in the lower level texts (e.g. tourist materials, hotel
brochures etc.) and which cause the howlers so much enjoyed by native
speakers.
Adaptation: keep visuals, change text slightly or significantly
This is the technique which makes necessary tactical adjustments in terms of
addressee needs and expectations, cultural norms, frames of reference.
According to Belch and Belch (1992:757), this is the dominant strategy used
by international advertisers. It is also the type of advertising which is most
interesting to examine in terms of translator training.
Revision: keep visuals, write new text
This strategy is somewhat problematical because the visuals of a campaign are
designed with a specific communication strategy in mind and so the message
cannot deviate substantially from the original concept. It is, however,
important to point out that it is easier to build on an existing original than to
start again from scratch.
But there are products whose appeal is entirely different in different
cultures. French women drink mineral water to retain their slender figures;
German women drink it because it is healthy (de Mooij 1994:218); English
women drink it because it is considered trendy. Thus advertisements may be
angled to stress these different marketing objectives in the different countries,
while the visual might well be a picture of the bottle in all three cases.
Naturally a sixth strategy exists but it cannot generally be identified:
independent local advertising campaigns produce different advertisements with
different visuals and texts for each country. This is not relevant for our
investigation here.
Conclusion
One of the major problems encountered in training translators is that they are
unwilling to leave the safe haven of a "straight translation". Examination of
translated advertisements shows how far professionals are prepared to deviate
from their original in order to achieve success with the target audience.
Moreover since translators of advertisements have to deal in a succinct and
successful way with many of the purely language-oriented difficulties of
translation: puns, jokes, assonance, metaphor, alliteration etc., strategies for
dealing with these components can be derived from such analyses. Comparison
of paired translations is therefore a useful tool for translator training.
References
Alexander, Ralph S. (ed.) 1965. Marketing definitions. Chicago: American Marketing

Association.
American Marketing Association. 1985. Marketing News. March 1,1.
Belch, George E. and Belch, Michael A. 1992. Introduction to advertising and promotion.
Homewood, IL: Irwin.
de Mooij, Marieke. 1994. Advertising worldwide. New York: Prentice Hall.
Klein-Braley, Christine and Franklin, Peter. 1989. "A potent part of Germany"? Some
Remarks on the Quality of Translations of Promotional Material". In: E. Haberfellner
(ed.) Sprache - Wirtschaft - Neue Medien. Reutlingen: International Vereinigung
Sprache und Wirtschaft, III, 43-59.
Leech, Geoffrey. 1966. English in Advertising. Harmondsworth: Penguin.
Ray, Michael L. 1982. Advertising and Communication Management. Englewood Cliffs, N.J.:
Prentice Hall.
Ricks, David A. 1983. Big Business Blunders: Mistakes in Multinational Marketing.
Homewood, IL: Dow Jones-Irwin.
Smith, Veronica and Klein-Braley, Christine. 1985. ... in other words. München: Hueber.
Snell-Hornby, Mary. 1992. "Translation as a cultural shock." In: C. Blank, Claudia (ed.)
Language and Civilization. A Concerted Profusion of Essays and Studies in Honour of
Otto Hietsch. Frankfurt: Peter Lang.
Wells, William, Burnett, John and Moriarty, Sandra. 1992. Advertising - principles and
practice. Englewood Cliffs, N.J.: Prentice Hall.
New advertising markets as target areas for
translation
Zuzana Jettmarova/Maria Piotrowska/Ieva Zauberga
The present study is based on the corpus data comprising press consumer
advertisements and television commercials translated in recent years from and
into Czech, Latvian and Polish. The corpus shows the impact of the respective
target cultures on translation strategies applied to this specific textual category.
Advertisement translation has emerged as a topical problem in many
cultures, and has acquired scope due to the recent switch away from canonised
literature studies to non-canonised texts. In post-communist countries like the
Czech Republic, Latvia or Poland, advertising of consumer products and
services has emerged as a new text type in consequence of the swift transition
from planned economy, which rendered advertising futile, to market economy,
which is advertisement-dependent. A domestic genre of advertising in these
countries was non-existent apart from announcement-like advertisements
between 1945 and 1989 whose function was different in the state-monopolised
non-competitive market. There are no established generic conventions in the
receiving system which would guarantee the new function of advertising (to
promote sales on a competitive market), whereas the communicative and
persuasive strategies of the pre-war conventionalised advertising genre are
generally outdated.
The phenomenon of current translations brought about by the socio-
economic and political changes after 1989, offers the translator and the
translation scholar interesting material for research and contrastive analyses.
Unfortunately, the common ground for investigation seems to be provided by
unsatisfactory performance in advertisement translation. What poses major
obstacles for efficient translation is not only the language but primarily cross-
cultural unawareness.
Cultural stereotyping
The present decade — a period of incredible expansion of the advertising

industry in the East European markets — is a scene of obvious foreign (i.e.
West European and American) impact. As a genre representing very strong
correlation between the text and culture, as consumer-oriented and overtly
persuasive texts recommending not only goods for sale but also attitudes,
advertisements produce and are themselves the result of cultural stereotyping.
Translated advertisements import social values and often unrecognised beliefs,
as well as linguistic patterns. New text type conventions are being created
through the adoption of foreign textual features, partly mediated by literal
translations, which sometimes causes a clash of linguistic and cultural norms.
Adverts are products of the culture from which they arise. In Eastern
Europe they also reflect the western cultures post-communist countries aspire
after. European ideas are becoming attractive via the medium of
advertisements, which on the one hand are culture-specific because they reflect
consumer orientations on the market concerned, and are bound to culture-
specific situations in which the respective goods and services are produced
(e.g. in Poland there are television commercials of washing powder produced
locally, which revive old Polish noblemen's traditions, are humorous and
deeply rooted in the Polish history of the 17th and 18th centuries). On the
other hand, they show a universal character because they reflect the
phenomenon of globalisation of products and services. The power of brand
names spreads internationally and leads to the import of source images to the
target texts, as well as bringing foreign stereotypes into the target language
society and culture (e.g. perfume ads which advertise the product associated
with prestige and luxury).
In the circumstances of internationalisation of production and marketing,
a certain amount of advertisements are written with the view that they will be
exported together with the products they advertise. One of the existing policies
of global scale trusts, for example, is the practice of supplying local agents in
the export countries with background information on goods intended for
marketing. In such cases the linguistic dimension vanishes in translation, as
source and target texts can no longer be perceived as opposites or there is no
source text in the traditional sense at all.
Zuzana Jettmarova/Maria Piotrowska/Ieva Zauberberga 187
Translation strategies in advertising
Because of the fact that various aspects of advertisements are closely bound up
with cultural phenomena, intercultural and not merely intertextual comparisons
have to be made and appropriately considered in translation. Consumer
orientations and cultural stereotypes constitute translation determinants, thus
establishing the pragmatic level of equivalence as the translator's priority. The
advertising message is considered as a whole. Verbal and non-verbal
components complement each other, the former frequently being less important
for the function and efficiency of the advertising discourse.
Major strategies implemented in advertisement translation are three-fold:
1. total transfer = literalness (image and semantic contents preserved, exotic
features of the original highlighted)
2. translation with minimum changes = advertising compromise = partial
adaptation (various degrees of departure from the original, partly adapted
discourses)
3. adapted translation = cultural transplantation = total adaptation (images
and text transformed to appear more alluring to the target audience,
exchange of picture and sound or text for a domestic milieu.)
Literalness and adaptation constitute extreme variants of translational policy,
the continuum in between the two opposites being filled in by various degrees
of departure from the original advert (cf. Hervey & Higgins 1992:28-35). The
advertising compromise uses devices like dubbing, voice-over or subtitling of
the verbal component leaving the picture or sound only transposed, e.g.
television commercials of Wash-and-Go, Wella, Procter and Gamble, Tchibo,
Wrigley Spearmint, etc. during the period from 1990 to 1992. This is a
strategy of partial and overt translations since the non-verbal
semantics/semiotics and the verbal content are perceived as alien to the target
text audience. Cultural transposition of the same commercials advanced in the
years 1993-1995. A further degree of adaptation requires the domestic setting
of the advertisement, while the macrostructure of the discourse and the content
of the verbal message remain unaltered since the original scripts have been
literally translated from the English originals.
Deficiencies of direct transfer
Faithfulness to some extent still prevails in East European advertisement

translations where it fails compared to translation of other text types, e.g.
expressive texts. Direct transfers in advertisement translation frequently result

in the loss of the persuasive force and the change of the source text function
(e.g. from an operative to an informative text). The recipient from a different
cultural context cannot participate in the game, the rules of which have been
worked out for a different, remote recipient. To achieve intended response the
rules have to be revised, i.e. the text rewritten. Otherwise the cheerful
superlatives of Western adverts are perceived as alien, related to a different
world where the target readership feels out of place. As pointed out by Gideon
Toury, a certain amount of deviance from the source text is to be regarded not
only as justifiable, or even acceptable, but as actually preferable to complete
normality, on all levels at once (Toury 1995: 28).The following example into
Latvian can be used to demonstrate this point:
(1) Our limousines give you unsurpassed quality and maximum flexibility.
Musu limuzeni garante jums neparspejamu kvalitati un visaugstako fleksibilitati.
The textual-linguistic structure of the source text has been retained, almost
50% of words are obvious loans (limuzini, garante, kvalitati, fleksibilitati).
Today the acceptance of western mass culture in the East European countries
is unconditional — psychologically because of its novelty and politically
because of its non-Soviet orientation. The practical consequence is tolerance
for its interference. Non-translation or zero translation is often to be traced in
the field of advertising:
(2) DHL Worldwide Express. DHL International Latvia SIA. We keep your promises.
("Rigas Laiks" '95/6:69)
Non-translation has been generally recognised as evidence of linguistic
imperialism and foreign cultural dominance. During the exchange between
cultural traditions the exporting (active) systems are in a power position in the
eyes of the importing (passive) systems; this is especially relevant for the
importation of non-translated discourse, which obliges given populations to
adapt themselves to the idiom and the rules of the visitors. In the present East
European culture situation, however, few are concerned about the rapid growth
of Anglo-American influence and the import of values into native cultures, as
well as the flood of borrowings into the native languages. The domination of
an anglophone culture is rather taken as a welcome switch and defence
mechanism against the possible reinstatement of the former exposures, as the
means of joining the rest of the civilised world. Thus the absorption of
anglophone loan items is accepted, even if in many cases it interferes with the
efficiency of translation. Perhaps in the age of global internationalisation, mass
scale communication and progressive acculturation, the very concept of the

target culture should be reconsidered.
Adaptation as a prerequisite of efficient translation
Advertising which is considered to be a form of persuasion directed at large

numbers of people by means of the media (O'Donnell 1991:105), is
successfully translated if "in an operative text it produces a text-form which
will directly elicit the desired response" (Reiss 1976:109). Since each target
text is always addressed to the recipient-in-situation different from those to
whom the source text is addressed, adaptation of the text emerges as a major
consideration in the process of translation. Dirk Delabastita, in discussing
translation and mass communication, points out that instead of translating mere
semantic and syntactic structures, translators rather translate texts into texts,
and in that process a lot of things may happen which are quite similar to the
manifold operations that occur in film translation and which defy any static
definition: reductions, additions, stylistic or ideological shifts, adaptation of
sociocultural data, changes in the visual presentation of the text (cf. Delabastita
1989:214).
Adjusting the source text to the target culture background involves foreign
culture words which need to be adapted or acculturised. In the promotional
booklet "Country Holidays in Latvia" one can read the following text:
(3) Garlaicibai nebus vietas, jo notiks saulosanas un peldesanas, izbraucieni ar laivu,
velosipediem, zirgu izjades, persanas pirtina, makskeresana, ogosana un senosana,
sporta speles, dejas. ("Country Holidays in Latvia" 1995:1)
The travel agency offers the favourite pastimes of Latvians: sunbathing,
bathing, boating, cycling, horse-riding, steam bath, angling, mushrooming,
berry picking in the woods, sports games and dancing. They also offer
nourishing country food. The English translation omits sunbathing, which for
a foreigner is hardly a plausible attraction in the given climate, sports games,
cycling and steam bath for which the facilities are hardly developed, mushroom
and berry picking which are typically Latvian pastimes. Instead the English
text offers bird watching and making new friends. The sentence about the
nourishing food has been left out all together:
Enjoy yourself boating, swimming, fishing, watching birds and animals, riding,
dancing, making new friends.
The success of the translation largely depends on the translator's awareness of

the necessity of adaptive translation, determined by the way the intended target
language receivers are assumed to react to the texts. Otherwise the target text
will turn out to be an ineffective hybrid containing contradicting stylistic
features with an undefined addressee and blurred intonation.
It seems that of all text types the advertisement is the one that requires the
most free recreation which arises from its manipulative function: it should
convince the reader of the need or benefits of the commodity and persuade him
to act. Sometimes the need is created artificially by the source culture. Even
if we assume that often the aesthetic function of the advertisement supersedes
the manipulative, the importance of belonging to a concrete cultural situation
is not diminished. Many English advertisements for example are strongly
language bound and hence resist the transfer:
(4) To air is human. To Volkswagen is divine. Or at least heavenly, as anyone who
has ruffled their hair in a Cabriolet would agree, (play upon Pope's To err is human;
to forgive, divine, "Auto Riga", August 1992)
(5) For those who aren't at their best at breakfast. The best Breakfast Tea. (play upon
the polysemy of the word best, "Cosmopolitan", May 1994)
(6) Perfume CAROLINA HERRERA. The fragrance that dresses the dream.
(alliteration, "New Woman", March 1993)
(7) We always fly at the right altitude, (implied similarity between 'altitude',
'attitude', British Airways advertisement, "Time" May 1993)
Advertisement translation from the Czech, Latvian and Polish perspective
For obvious reasons, the present review can neither be treated as an exhaustive
report on advertising techniques, nor a detailed study of the linguistic and
social phenomenon of advertising. A question beyond any doubt here is the fact
that it is essential to consider the importance of cultural information coming
from images brought into adverts, cultural implications and allusions created,
stereotypes built into adverts.
Literalness conceived as a translational strategy operating on the semantic
level, has been a global method in the advertisement translation in the Czech
Republic in the formative period of the free economy. It was especially
predominant during the period from 1990 to 1992 and has been declining since
then, although it still covers about 90% of all advertisement translation.
Marketing research, consumer articles in periodicals and letters of television
viewers reveal that the Czech consumer generally tends to prefer the hard-sell
Zuzana Jettmarova/Maria Piotrowska/Ieva Tauberberga 191
advertising format built on argument and factual information. Literal

translation has been approved of as acceptable and reliable norm justified by
an ignorant assumption that literal means closer, that is more faithful to the
original. Current Czech translations are inevitably hybrids, and their imported
element is both enriching for the receiving generic system and at the same time
perceived as irritating by the consumers.
(8) The Fiat Punto. The Answer, (a billboard)
Czech translation, 1st version: Fiat Punto. Odpoved! (Fiat Punto. The Answer.)
Czech translation. 2nd version: Fiat Punto. To je odpoved! (Fiat Punto. It is the
answer.)
In the first, word-for-word translation the conventions of public speech were
violated in copying the block language (i.e. two noun phrases). In the second
version the still literal translation has a copular verb inserted and the structure
is changed into NP + S, which is closer to the target language convention.
In Poland literalness has almost lost its domineering position as a
translational convention. A very dynamic development of the consumer-
oriented activities and growth of adverts frequently originating as translations
has been characterised by a dramatic transition from non-translation to cultural
transplantation in the field of advertising. Poland started by promoting products
in a professional way, taking into consideration factors referring to the
producer, technological advantages, and the merits of a product that made it
superior to other goods on the market. The hard-sell approach was soon
replaced by the soft-sell one, and advertising has been in the process of moving
towards a more western-like promotion based not so much on logical
arguments and objective merits of a given product (in western markets the
amount of similar-quality goods is so great that it eliminates the effectiveness
of factual advertising), but on selling the image of a product. Advertising
trades on desirable connotations and correct impressions. Very often it is the
psychological appeal to individuality or fashion, to recognise brand names, to
identify oneself with people interested in quality of life, ecology, etc. All of
these are beginning to be established as systemic values in Polish society.
Frequently used extraordinary images and texts of adverts are calculated to
create a shocking effect on the buyer, to attract him/her in an unconventional
way (e.g. an advertising sequence in a Polish daily "Gazeta Wyborcza"
promoting cars by way of juxtaposing their names, for example Volkswagen,
Renault, and abstract qualities like strength, reliance, independence represented
by human bodies).
Today Western adverts flood the Latvian market together with Western
goods. To use Pym's metaphor, texts are like sails raised to the wind: not all
texts are transferred in all directions all the time (Pym 1992:136). Winds in
Latvia at present seem to be blowing from the West, bringing in commercial
culture where advertisements are indispensable. Since there are no ready-made
models available, Latvian advertising can be said to be groping its way trying
to imitate Western patterns, which due to differences in cultural setting often
fail to produce identical behavioural reactions. The dominance of the Western
pattern can also be explained by the limited scope of local production. The few
efficiently functioning local companies often produce very attractive
advertisements, especially non-verbal commercials, but the degree of
'aggressivity' of these companies as well as the number of goods to be
marketed fall behind the imported products.
The tone of Latvian advertisements reflects the low self-image and Latvian
(Soviet?) mentality — adverts are more modest in comparison with Western
advertisements due to the lack of confidence about the quality of advertised
objects and reluctance to shoulder the responsibility. There is often an
apologetic tone to be detected:
(9) A walk in the streets of the old town could be quite pleasant at night but you
should better have a reliable companion or postpone all the romantic impressions until
daytime. ("Riga This Week", summer 1993:28)
Another example reflects the long-standing admiration for Western goods,
which for a different readership may sound like an anti-advertisement:
(10) The second highlight is Latvian — Swedish JV "Eurolink Hotel" which is
intended for businessmen. As regards its interior, the mode of rendering services, its
cuisine — they are European ones (provisions are supplied by Swedish party, only
flowers and greens are local). ("Riga This Week", summer 1993:20)
An ad of the insurance company "Rigas Fenikss" is laconic:
(11) We ENSURE according to Western standards, (magazine "Rigas Laiks", 95/6:61)
In a glossy magazine "Latvia. Baltic State" (1995/2:23) the chief engineer of
the leading Latvian cement and roofing slate factory is interviewed. The
journalist tries hard to promote the enterprise and gives the article a lofty title
"Feathers Grow on Broceni (name of the factory) Wings". Accordingly the
final question: What is your noblest objective? And the answer: We shall
continue the efforts to renew the roofs — a typical sample of Latvian modesty.
Due to self-image related problems in the Latvian case the first
requirement seems to be emotive adaptation, i.e. the text needs to be toned
down if translated from English into Latvian or enhanced if translated from
Latvian into English.
Beside some distinctive characteristics of the three advertising markets,

there are also certain common features and considerations applicable in the
three countries. With the society changing, with the system of values and
beliefs also changing, literal translating of advertisements (regarded as a
contributing factor to the internalisation of language, discourse and culture),
tends to be replaced by adaptation. Advertisements are obviously only
establishing their place in the East European cultural scene in countries like the
Czech Republik, Latvia or Poland. They have already become more target-text
oriented, as certain translational norms for the rendering of operative texts
have already emerged. As shown in the preceding discussion, the very first
kind of norms that have a chance to shape any kind of communication and
hence also translation are norms that have been established by political and
economic institutions. Using Toury's terminology, this can be called a
preliminary norm which determines the choice of texts to be translated. Hence
it is the initial norm and the operational norm that are still in the process of
being shaped. The actual decisions made during the translation process are
lacking in consistency, thus leading to uneven performance. At the back of
operational indecisiveness there seems to be lurking the dilemma concerning
the initial norm, i.e. the translator's dilemma whether to do homage to the
original text or to the linguistic and textual norms active in the target system,
which is the prerequisite for an effective advertisement translation. At this
stage of the development of advertising as a genre in its own right, there is as
yet no matching counterpart on the macrostructural level in the receiving
systems.
References
Chesterman, Andrew. Oy Finn Lectura Ab, Finland (eds.) Readings in Translation Theory.
1989, 105-116.
Delabastita, Dirk. 1989. "Translation and Mass Communication". Babel 35 (4), 193- 218.
Duff, Alan. 1989. Translation. Oxford: Oxford University Press.
Dyer, Gillian. 1982. Advertising as Communication. Boston, Massachusetts: Auburn House
Publ.
Hervey, Sandor & Higgins, Ian. 1992. Thinking Translation. A Course in Translation Method.
London: Routledge.
Jänis, Marja & Priiki, Timo. 1994. "User Satisfaction With Translated Tourist Brochures: The
Response of Tourists from the Soviet Union to Russian Translations of Finnish Tourist
Brochures". In: C. Robyns (ed.) Translation and the (Re)production of Culture. Leuven:
The CERA Chair, 49-56.
Leiss, William, Kline, Stephen, Jhally, Sut. 1993. Social Communication in Advertising.
O'Donnell, W. R. & Loreto, Todd. 1991. Variety in Contemporary English. London, New
York: Routledge.
Pym, Anthony. 1992. Translation and Text Transfer. An Essay on the Principles of
Intercultural Communication. Frankfurt/Main, Bern: Peter Lang.
Reiss, Katharina. 1976. "Text Types, Translation Types and Translation Assessment". In: A.
Chesterman, Oy Finn Lectura Ab, Finland (eds.) Readings in Translation Theory. 1989,
105-116.
Toury, Gideon. 1995. Descriptive Translation Studies and Beyond. Amsterdam: John
Benjamins.
Vestergaard, Torben & Schröder, Kim. 1985. The Language of Advertising. Oxford: Blackwell
Publ.
Williamson, Jane. 1981. Decoding Advertising. London: Marion Boyars.
Getting the message across — Simultaneous
interpreting for the media
Ingrid Kurz
The linguistic fragmentation of Europe continues to be a problem for mass

communication. The growing number of Europe-wide television broadcasts by
satellite and the resulting need to make them understandable to an audience
composed of people speaking some thirty different languages are calling for
more translation and interpretation than ever before. It is estimated that some
50,000 program hours are being treated annually by one of the various forms
of language transfer for television in Western Europe — with English being
the biggest single source language (Luyken 1991). For live TV broadcasts,
simultaneous interpretation — initially limited to the coverage of major events
— has also become increasingly important.
Following a description of some of the special features of media
interpreting, the paper will give an overview of the author's experience and
outline the present TV interpretation 'scene' in Austria. It will conclude with
a brief discussion of a recently conducted survey comparing the expectations
of TV professionals and conference participants regarding high-quality
simultaneous interpretation.
Special features of TV interpreting
The early literature on media interpreting includes several "case studies" by

interpreters reporting on their practical experience with live interpretation on
TV programs, such as the Apollo 11 broadcasts (Pinhas 1972; Kurz 1985;
Nishiyama 1988) and the Eurikon experiment (Daly 1985). Subsequent
publications have discussed the use of simultaneous interpretation for live
broadcasts during the Gulf War (Gambier 1994) and for numerous other
programs (Kurz 1986, 1993a; Bros-Brann 1994; Kurz & Pöchhacker 1995;
Pöchhacker in this volume).
Most authors (Daly 1985; Kurz 1990; Bros-Brann 1993) stress the need
for close cooperation with program producers and sound engineeers on
technical matters in order to make sure that interpreters get what they need to
work well: individual volume control, light mono earphones, etc. (cf. also
AIIC Guidelines for Technicians 1986).
Likewise, the special stress factors and challenges of media have been
widely discussed (cf. Skuncke 1983; Kurz 1982; 1990, Daly 1985; Strolz
1992; Bros-Brann 1994; Mayer 1994). A summary of the most salient aspects
will be given below.
Stresses and strains
Live TV interpreting is generally felt to be more stressful than simultaneous

interpreting in other settings, with special stress factors stemming from at least
three major sources:
1. Physical environment: When working for TV, the interpreter may have to
sit in the newsroom or in a separate studio (rather than in a soundproof booth),
subject to all sorts of visual and acoustic distractions and disturbances.
In the majority of cases, the interpreter has no direct view of the
speaker(s); the visual input is received via a monitor.
Whereas in a conference setting participants and speakers usually interact
with each other, communication in the case of radio and TV broadcasts is in
one direction only. The interpreter receives no feedback from the audience,
and the listener or viewer cannot indicate verbally or otherwise that
information has not been understood.
2. Work-related factors: Quite often, TV interpreting has to be done late at
night and/or on short notice, with little opportunity for preparation.
In an 'ordinary' conference setting, the interpreter can 'get used' to a speaker
with a peculiar accent or speaking style. On TV, this is usually not possible,
given the brevity and the stress of a live performance (e.g. a short but crucial
interview during a newscast).
Despite the many technological advances and improvements the sound
quality may be — unavoidably — poor (as e.g. during the Apollo and Gulf
War broadcasts). Together with other occasional technical problems (e.g.
feedback of the interpreter's voice) this adds to the difficulties.
3. Psycho-emotional stress factors: Knowing that he/she is interpreting for an
audience of hundreds of thousands or even millions, the TV interpreter is more
Ingrid Kurz 197
keenly afraid of failure than during 'ordinary' conferences. Newspapers will

not hesitate to report critically.
TV viewers do not understand or appreciate the difficulties the interpreter
may be facing. Being used to the self-generated "authenticity" of TV (Daly
1985), i.e.the fluent presentation of texts by newsreaders and moderators, they
expect the interpreter to live up to the same standards of performance. The TV
interpreter must therefore endeavor to make his/her style and delivery
particularly smooth, even if the speaker he/she is interpreting is not all that
lucid.
Another stress factor for the media interpreter, especially when working
from English, is that the original sound, although reduced to a lower level of
audibility, is never completely "covered up" by the interpreter's voice but
remains audible in the background, thus allowing the audience to check on the
completeness and fidelity of the interpreter's output.
Challenges and chances
In media interpreting, speed is of the essence, i.e. the interpreter's voice must
coincide as much as possible with that of the person being interpreted. In an
interview situation with fast, brief questions, an interpreter lagging too far
behind would make the whole exercise unpalatable and unacceptable (Kurz
1990). Ideally a television interpreter should be able to work at supersonic
speed (Bros-Brann 1994).
Since 'revoicing' on TV replaces only one element of the entire opus —
the spoken text — without affecting the visual component, it is both more and
less than conventional translation. Audiovisual language transfer incorporates
an editorial element (Luyken 1991). Interpretation for the media, too, is a form
of communicative language transfer requiring editorial decisions, content-
related judgments and cultural considerations.
The TV interpreter works for a very heterogeneous audience which is
likely to comprise a wide cross-section of the population, part of which may
have no (or very little) knowledge of the program-originating country and its
culture. According to TV professionals (Mayer 1994), the TV interpreter
should therefore try to coordinate his/her interpretation with the images the
viewers are receiving and should occasionally add commentary to provide the
audience with contextual information to render the speaker's message
meaningful.
In view of all these demands, some authors feel that TV interpretation
requires a "hybrid" or new breed of interpreter (Laine 1985), a new job profile
(Kurz 1990), a new "mind set" compared to the everyday practice of

conference interpretation. In the words of Bros-Brann (1994) "it takes a very
special sort of person to be a conference interpreter in the first place, but (...)
it takes an even more special type of person to work on TV."
TV interpreting in Austria
In Austria, the preferred methods of language transfer for "produced"

programs (foreign-language films, documentaries, prerecorded interviews, etc.)
are dubbing and revoicing, i.e. voice-over, narration or free commentary
(Luyken 1991). For 'live' programs, the Austrian Broadcasting Organization
(ORF) has been using interpretation (mostly in the simultaneous but
occasionally also in the consecutive mode) for more than a quarter of a
century.
The first Austrian TV programs using interpretation were broadcasts of
major, spectacular events attracting a vast audience, such as the Soviet and
U.S. space missions and U.S. election night in 1968.
The author's experience with TV interpreting dates back to these early
broadcasts:
— an all-night live TV satellite transmission of the 1968 U.S.presidential
elections (a remarkable technological achievement at that time, receiving
extensive coverage in the print media), and
— a 28-hour TV broadcast (the longest live transmission in the history of
Austrian TV) of the first moon landing (Apollo 11) in July 1969, which
had practically the entire nation glued to the TV screen.
A summary report of the developments and changes of the TV interpretation
'scene' in Austria will be given below. Considering the fact that so far all
statements regarding media interpreting have been either anecdotal references
or individual case reports, a review and statistical evaluation of some 25 years
of experience is likely to enhance the validity of those statements.
All figures quoted in this article (number of assignments, percentages
indicating breakdown by type of program) reflect the author's TV interpreting
activities from 1970 through 1995. They involve no claim to completeness and
are not synonymous with the total amount of English-German interpretation
during that period. However, they probably yield a fairly representative picture
of the overall volume of and trends in media interpretation in Austria during
the last 25 years or so, since by far the largest share of English-German TV
interpretation was done by the author of the present paper. Since prior to the
Ingrid Kurz 199
nineties, ORF used live interpretation mainly for news and current affairs
programs, a review of the author's assignments is reminiscent of leafing
through a contemporary history book.
For the sake of simplicity, the data for 1970 -94 were compressed into 5-
year periods. The figures for 1995 are shown separately. An overview of the
data is given in Table 1.
Period ø No. of News & Sports Info- Religion

assignmts. Current tainmt.
per year Affairs
1970-74 8.8 93.7% 6.3% - -
1975-79 5.0 89.3% - 10.7% -
1980-84 5.0 83.5% - 16.5% -
1985-89 8.8 67.5% - 25.0% 7.5%
1990-94 21.0 70.5% 14.3% 15.2% -
1995 20.0 50.0% 5.0% 40.0% 5.0%
Table 1. English-German interpreting assignments for ORF (1970-95)

Even a cursory look at these data reveals that:
— It was only some ten years ago that programs other than news and current
affairs broadcasts started to make wider use of live interpretation. Prior
to that, the volume of TV interpretation had remained more or less
stagnant over a period of twenty years (1970-89).
— It was not until the nineties that the volume of live TV interpretation in
Austria increased markedly.
A more detailed quantitative and qualitative analysis yields the following
picture: From 1970-74, space missions (Apollo 12-17, Skylab) and U.S.
politics (Nixon's visit to Moscow in 1972, his resignation and Ford's
inauguration in 1974) continued to be the predominant topics of live TV
programs using English-German interpretation. The average number of
interpreting assignments/year during that period was 8.8. No less than 93.7%
of those broadcasts were news and current affairs programs; a mere 6.3% were
sports programs.
The next five years (1975-79) saw a slight decrease in the overall amount
of TV interpretation (average number of assignments/year: 5). Major topics
were the Soyuz/Apollo flights in 1975 and U.S./international politics (e.g. the
Ford-Sadat meeting in Salzburg in 1975, U.S.presidential elections in 1976,
the SALT II agreement and the Pope's address to the U.N.General Assembly
in 1979). News and current affairs accounted for 89.3%; the remaining 10.3%
were "infotainment" programs (i.e. a wide range of information and
entertainment programs including science and arts programs, cultural
programs, human interest programs, talk shows, interviews, etc.).
The list of the author's TV interpreting assignments between 1980 and
1984 again looks like an excerpt from the chronology section of the World's
Almanach. 1980: failure to free U.S. hostages in Iran; U.S.presidential
elections (Carter/Reagan); 1982: the Pope's visit to Great Britain; 1984: Indira
Gandhi's funeral; U.S.elections (Reagan/Mondale). While the total number of
TV programs with English-German interpretation remained unchanged (an
average 5 assignments/year), the use of interpretation in infotainment programs
(including talk shows) increased to 16.5 %, leaving news and current affairs
programs with a share of 83.5%.
With the exception of the decreasing news value of space flights and a
slight increase in infotainment programs, the overall TV interpretation 'scene'
in Austria did not undergo any dramatic changes during the first fifteen years
under review (1970-84).
It was only in the second half of the eighties that the picture began to
show greater signs of diversification. Those were the years of the
Reagan/Gorbachev summit meetings (1987-88), the election of Bush (1988),
his visits to Warsaw and Budapest and the summit meeting with Gorbachev
(1989) — all outstanding media events requiring interpretation (accounting for
67.8% of the total work volume). Besides, infotainment and religious programs
started to rely on interpreters more often (with shares of 25% and 7.5%, resp.,
in a total average of 8.8 assignments/ year).
It is interesting to speculate whether this was because TV professionals,
having become increasingly familiar with how TV interpreting works, began
to see what it can offer the producer and presenter of a program:
— a considerable saving in time;
— direct communication without the usual delay that is inevitable if the
presenter has to translate everything himself (assuming that he is able to
do that in the first place); and
Ingrid Kurz 201
— the advantage that the presenter does not have to concentrate on a foreign
language and can therefore focus on his/her questions.
The most noticeable changes in the Austrian media interpretation 'scene',
however, came with the beginning of this decade. The nineties brought a
sizeable expansion of the total amount of interpreting for TV. The average
number of assignments in the period 1990-94 was 21; the total for 1995 was
20. They mirror the many decisive political events and changes of the time: the
Bush/Gorbachev summit meeting and the signing of the CSCE Paris Charta in
1990; the Gulf War and trouble in the Soviet Union in 1991; elections in Great
Britain and the U.S. and the crisis in Yugoslavia in 1992; events in
Yugoslavia, the Middle East, Moscow, South Africa, and the U.N.World
Conference on Human Rights in 1993; the war in Bosnia, Austria's accession
to the European Union, Clinton's visit to Berlin, and Nixon's funeral in 1994;
the war in Bosnia and its conclusion with the signing of the Dayton Peace
Agreement, and Rabin's funeral in 1995.
Despite this increase in the absolute number of news programs using
interpretation, their share in the total number of interpreted programs went
down to 70.5% in 1990-94 and 50% in 1995 owing to the more frequent use
of interpretation in other programs, such as infotainment (1990-94: 15.2%;
1995: 40%), sports (14.3% and 5%, resp.), and religion (1995: 5%).
The absolute figure quoted for 1995 is less representative than the figures
for previous years, as — for interviews at least — ORF has increasingly begun
to use voice-matching, i.e. male interpreters for male voices and female
interpreters for female voices.
Certainly, one of the factors contributing to the increased use of
interpreters on ORF's broadcasts was foreign competition. Since the Austrian
public has ready access to broadcasts from other countries, ORF's coverage
needs to be cosmopolitan (cf. Mayer 1994). In this context one might add that,
of course, coverage of the events in eastern Europe has also added to the
number of languages. However, English is frequently used by non-native
speakers and remains by far the most important source language.
The author's TV interpretation schedule for 1993 (Fig. 1) may serve as
an illustration of the wide variety of subjects a media interpreter has to expect.
It lends support to those authors who feel that media interpreting can rightly
be considered an "additional specialization of experienced conference
interpreters" (Daly 1985:203) and that flexibility is probably the next most
important thing after speed (Bros-Brann 1994).
Topics of Special Broadcasts:

U. S. -Iraq relations
Problems in former Yugoslavia
Fifty years after the battle of Stalingrad
Zbigniev Brzezinski on the New World Order
Peace in the Middle East
The situation in Moscow
Nobel Peace Prize (Nelson Mandela and Willem de Klerk)
Round Table discussion on Austria and the European Union
World Conference on Human Rights (5 broadcasts)
President Clinton's inauguration speech
President Clinton's State of the Union address
Vancouver Summit
Prince Philip speaking on the World Wildlife Fund
Interviews:
Haris Silajdzic, Bosnian Prime Minister
Yasser Arafat
Harry Belafonte
UNPROFOR reprepresentative (on Yugoslavia)
President Berisha of Albania
U.N Secretary General Boutros-Ghali
Talk shows
(featuring singer LaToya Jackson and U.S. author Camille Paglia)
Several sports programs
(covering a wide range of sports)
Fig. 1: Live interpretation assignments from English into German for ORF in 1993
Quality expectations: Conference vs. media standards
Empirical evidence that expectations of the quality of media interpreting differ

from those of conference interpreting was obtained in a recent study among 19
Austrian and German TV professionals (cf. Kurz & Pöchhacker 1995,
Pöchhacker in this volume) whose ratings were compared with those given by
a total of 124 conference participants (Kurz 1993b). They were asked to assess
the relative significance of eight criteria introduced by Bühler (1986) {native
accent, pleasant voice, fluency of delivery, logical cohesion of utterance, sense
consistency with original message, completeness of interpretation, correct
grammatical usage, use of correct terminology) for the quality of interpretation
on a four-point scale.
Ingrid Kurz 203
Table 2 illustrates how the expectation profiles of TV professionals and

conference participants compare with each other.
native pleasant fluency of logical sense complete- correct terminology

accent voice delivery cohesion ness grammar
TV people
(N=19) 2.84 3.47 3.32 3.68 3.84 2.53 2.79 3.32
delegates
(N=124) 2.37 2.5 3.1 3.46 3.69 3.2 2.5 3.4
Table 2. Comparative ratings of quality criteria for simultaneous interpreting by TV

professionals and conference participants
There is agreement in both groups that sense consistency and logical cohesion
are the two most important criteria. This finding ties in with the results of a
more recent users expectation survey among over 200 conference participants,
who stated that clarity of expression — for which the two above-mentioned
criteria are essential — was the most important quality feature (cf. Mackintosh
1995).
In a discussion of the merits and demerits of the different language
transfer methods in the media, Luyken (1991:81), when commenting on
simultaneous interpretation ("live voice-over" in his terminology), writes that
"the effect of live voice-over is often monotonous and spasmodic." Clearly,
this is an indication that the media interpreter's performance is indeed being
judged against that of the TV moderator or newsreader and that the standards
which the interpreter's voice and diction are supposed to meet are very high.
This is confirmed by ORF's Chief Editor of News and Current Affairs:
Whoever interprets live programs for us must have the voice and clarity of a
broadcaster to satisfy the approximately two million who are our public. We have
a very knowledgeable and critical audience who will comment unfavorably if the
interpreter's voice and diction are not up to the usual standard of our reporters
and speakers." (Mayer 1994:11)
TV professionals participating in the study under review supported this view
by giving a higher rating to pleasant voice, fluency of delivery, native accent
and correct grammatical usage than conference participants. Their demands
were generally higher than those for 'ordinary' conference interpreting — with
one exception: completeness of interpretation, which for media interpreting is
obviously less of an issue than smooth delivery and clarity.
Conclusions
This paper has tried to throw some light on the quality expectations and
standards governing language use by interpreters on TV. It has provided
statistical and empirical evidence in support of the claims that, apart from
encountering all the difficulties of 'ordinary' conferences (a wide range of
technical subjects, unavailability of texts, different accents, excessive speed,
etc.), the interpreter working in the media is also confronted with special
requirements and restraints.
Recently, empirical research has been extended to two more 'down-to-earth'
questions:
1. Do media interpreters know what they are in for ? (Kurz 1996) and
2. How does the actual output of the media interpreter compare with the
standards it is supposed to meet? (Pöchhacker in this volume)
These are another two small steps designed to gather and analyze 'hard data'
in an effort to learn more about a field of work for interpreters that is here to
stay and obviously has a great potential for the future. Members of the
profession are aware of the fact that interpretation is never an end in itself and
that "the chain of communication does not end in the booth" (Seleskovitch
1986:236). Their aim as cross-cultural communicators must be to satisfy their
audience (Déjean le Féal 1990). An understanding of what quality is in the ears
of their listeners should help them in getting the message across even more
efficiently.
References
AIIC Technical Committee. 1986. Guidelines f or Technicians. Geneva.

Bowen, David and Bowen, Margareta (eds,) 1990. Interpreting — Yesterday, Today, and
Tomorrow (ATA Scholarly Monograph Series IV). Binghamton NY: SUNY.
Bros-Brann, Eliane. 1993. "Simultaneous interpretation and the media: Interpreting live for
television". Report to the AIIC Technical Committee. Geneva.
Bros-Brann, Eliane. 1994. "Interpreting live on television:Some examples taken from French
television". Unpublished report at the UIMP Course "The interpreter as a
Communicator", La Coruna, 26-29 July 1994.
Bühler, Hildegund (ed.) 1985. Translators and their position in society. Proceedings of the X
World Congress of FIT Vienna: Braumüller.
Bühler, Hildegund. 1986. "Linguistic (semantic) and extra-linguistic (pragmatic criteria for the
evaluation of conference interpretation and interpreters". Multilingua 5(4): 231-235.
Daly, Albert. 1985. "Interpreting for International Satellite Television". In: H. Bühler (ed.),
203-209.
Déjean le Féal, Karla. 1990. "Some Thoughts on the Evaluation of Simultaneous
Interpretation". In: D. & M. Bowen (eds.), 154-160.
Ingrid Kurz 205
Gambier, Yves. 1994. "Audio-visual communication: typological detour". In: C. Dollerup and
A.Lindegaard (eds.) Teaching Translation and Interpreting 2. Papers from the Second
Language International Conference, Amsterdam/Philadelphia: John Benjamins, 275-283.
Kurz, Ingrid. 1982. "Interpreting for Mass Media". In: Report on the Georgetown University
Round Table. NRCTI Paper. Washington DC: Georgetown University, 4-6.
Kurz, Ingrid. 1985. "Zur Rolle des Sprachmittlers im Fernsehen". In: H. Bühler (ed.), 213-
215.
Kurz, Ingrid. 1986. "Simultaneous interpretation of a panel discussion with Jeane Kirkpatrick
on Austrian TV". In: K. Kummer (ed.) Building Bridges. Proceedings of the 27th Annual
Conference of the American Translators Association. Medford NJ: Learned Information,
315-319.
Kurz, Ingrid. 1990. "Overcoming Language Barriers in European Television". In: D. and M.
Bowen (eds.), 168-175.
Kurz, Ingrid. 1993a. "The 1992 U.S. presidential elections: Interpreting the 'American
debatathon' for Austrian television". In: C. Picken (ed.) Translation. The vital link.
Proceedings of the XIII World Congress of FIT. London: ITI, vol. 1, 441-445.
Kurz, Ingrid. 1993b. "Conference interpretation: Expectations of different user groups". The
Interpreters' Newsletter 5, 13-21.
Kurz, Ingrid. 1996. "Special features of media interpreting as seen by interpreters and
users". In: Proceedings of the XIV World Congress of FIT, Vol 2, 957-965.
Kurz, Ingrid and Pöchhacker, Franz. 1995. "Quality in TV interpreting". In: Y. Gambier (ed.)
Audiovisual Communication and Language Transfer. Proceedings of the International
Forum Strasbourg. Translatio FIT Newsletter série XIV/3-4, 350-358.
Laine, Marsa. 1985. "New Ways of Communication — a New Breed of Translators". In: H.
Bühler (ed.), 210-212.
Luyken, Georg-Michel. 1991. Overcoming language barriers in television: Dubbing and
subtitling for the European audience. Media Monographs series. The European Institute
for the Media.
Mackintosh, Jennifer. 1995. "Portrait of the ideal' interpreter as seen through delegates'
eyes". AIIC Bulletin XXIII/3, 61-63.
Mayer, Horst Friedrich. 1994. "Live interpreting for television and radio". The Jerome
Quarterly 9/2, 11.
Nishiyama, Sen. 1988. "Simultaneous interpreting in Japan and the role of television. A
personal narration". Meta 33(1), 64-69.
Pinhas, René. 1972. "Les retombées scientifiques des opérations 'Apollo' sur l'interprétation
simultanée". La linguistique 8, 143-147.
Seleskovitch, Danica. 1986. "Comment: Who should assess an interpreter's performance?"
Multilingua 5(4), 236.
Skuncke, Marie-France. 1983. "Nouvelles perspectives de développement de l'interprétation,
interprétation à la télévision et la radio, interprétation à distance". In: A. Kopczynski et
al. (eds.) The mission of the translator today and tomorrow. Proceedings of the IX World
Congress of FIT. Warsaw: Polska Agencja Interpress, 394-395.
Strolz, Birgit. 1992. Theorie und Praxis des Simultandolmetschern. Argumente für einen
kontextuellen Top-down-Ansatz der Verarbeitung und Produktion von Sprache.
Unpublished doctoral thesis. University of Vienna.
"Clinton speaks German": A case study of
live broadcast simultaneous interpreting
Franz Pöchhacker
The use of simultaneous interpreting for live television broadcasts is one of the
more specialized forms of language transfer in the audiovisual media.
Compared to dubbing, subtitling and other major techniques of mediated
intercultural communication, simultaneous interpreting has a narrower scope
of application, being confined, ideally, to "live unscripted material" (Daly
1985:204) such as interviews, discussions or talk-show-type programs produced
in the TV studio or transmitted via satellite. As a rule, the interpretation into
the language of the audience is broadcast as a voice-over, with the original
speaker still audible in the background.
Notwithstanding the particular difficulties and constraints involved in live
broadcast interpreting (Daly 1985; Kurz 1990 and this volume), the quality
standards by which the performance of media interpreters is judged are at least
as stringent as those for "ordinary" conferences (cf. Kurz 1990:169), and in
some respects the level of output quality expected in media interpreting is even
considerably higher.
In a study by Kurz (in this volume) the quality expectations of various
groups of conference participants (Kurz 1989, 1993) were compared with those
of 19 representatives of Austrian and German TV organizations. The respective
ratings for Bühler's (1986) eight "linguistic" quality criteria (native accent,
pleasant voice, fluency of delivery, logical cohesion of utterance, sense consi-
stency with original message, completeness of interpretation, correct gram-
matical usage, and use of correct terminology) indicate that TV professionals
who employ and work with (simultaneous) interpreters in their programs give
a distinctly higher rating to pleasant voice, native accent, fluency of delivery
and correct grammatical usage but attached significantly less importance to the
criterion of completeness. These research findings clearly support the
impression of media interpreters that in live broadcast interpreting their

delivery must not only meet conference interpreting standards but also match
the "professionally artificial" performance standards of broadcasting. This has
been pointed out by authors like Daly (1985) and Kurz (1990) and summarized
more recently by Russo (1995) in the Media Interpreting Workshop of the
Strasbourg Forum on "Audiovisual Communication and Language Transfer":
The TV viewers' and radio listeners' expectations are so high that an interpreter
ought to become a performer rather than just a linguistic/cultural mediator.
Paramount importance is attached to factors such as: voice quality, a cohesive &
coherent language and a lively & self-confident performance, often to the
detriment, if necessary, of the fidelity or completeness of the original message.
(Russo 1995:343)
The interpreter is of course always a performer. The question to be addressed
in this paper is: how and how well can and does the interpreter perform in the
face of the added difficulties and constraints to meet the added demands on
output quality in live broadcast interpreting?
Quality: Ideal vs. real
Bühler's (1986) questionnaire on quality criteria in conference interpreting as

well as the follow-up surveys by Kurz (1989; 1993 and in this volume) among
different user groups referred to an "ideal interpretation". While there are
interesting divergences in the expectation profiles of interpreters, conference
participants and TV professionals, it is clear to see that the level of quality
expected of an optimum interpretation is invariably high: All of the eight
criteria used in the surveys cited above were rated as (more or less) important
rather than irrelevant. In the light of these high expectations for the optimum
quality of simultaneous interpetation in live broadcasts it is all the more
remarkable that a senior TV executive of the Swiss Broadcasting Corporation,
speaking during the Strasbourg Forum, referred to live broadcast interpreting
as a "rather crude" method of language transfer. Obviously, she must have
been thinking of some actual rather than ideal quality of TV interpretations
when she observed,
It is only the unique and irreplaceable medium of television and the way it offers
topical information that can explain why an interested viewer will put up with all
the inconveniences of a simultaneous translation that is constantly out-of-sync
with the pictures, that is full of hesitations and interruptions, and often leaves the
viewer trying to identify which speaker said what. (Mona 1995:6)
Franz Pöchhacker 209
While Mona's (1995) statement refers specifically to the simultaneous inter

pretation of multi-party discussions (cf. also Kurz 1985:214f.) it provides at
least an informed hint at the fact that in media interpreting, real performance
may not always match ideal standards.
Any attempt at investigating "real performance", i.e. the actual "quality"
of the interpreter's output, with reference to some objectifiable, if not object
ive, standards or parameters is of course fraught with daunting methodological
difficulties, and the fact that we are aware of these (cf. e.g. Kalina 1994; Gile
1995) is of limited help in actually overcoming them. The following case study
can therefore be no more than a modest example of product description and
analysis, with no claim to offering comprehensive solutions to problems like
assessing "information content" or "linguistic correctness". In order to ensure
compatibility with the research findings for quality expectations, the corpus
will be analyzed largely in terms of the criteria used in the questionnaire
studies cited above. (For other analytic proposals see e.g. Kopczyński 1980;
Alexieva 1988; Galli 1990; Vik-Tuovinen 1995.) The data for the simultaneous
interpreter's output will be evaluated in relation to both the general quality
requirements for professional interpretation and to the specific expectation
profile for TV interpreting.
Case study: Clinton in Berlin
The case under investigation is the formal address to the citizens of Berlin
delivered by U.S. President Bill Clinton at the Brandenburg Gate on 12 July
1993. The live transmission of the speech was broadcast on Austrian Televi
sion (ORF) with simultaneous "voiceover" interpretation into German by one
of Austria's most reputed and experienced media interpreters.
The speech which Clinton delivered from a carefully crafted manuscript
to an audience of thousands assembled at the Brandenburg Gate followed a
short address by Chancellor Helmut Kohl and lasted just over nine minutes.
The manuscript was not available to the interpreter working in the studio of the
ORF from a monitor.
Accent and voice
It would certainly be quite difficult to give a reasonably objective assessment

of the interpreter's output regarding native accent and pleasant voice. In fact,
we may assume that one would hardly take note of these requirements unless
they remained ostensibly — and audibly — unfulfilled. Finding agreement on
a threshold below which voice quality and diction are to be deemed unsatisfac-
tory would be the arduous task of those in charge of evaluating the results of
screening tests for would-be media interpreters, as are given, for example, by
the ORF. In the present case study, involving a senior ORF media interpreter
with more than two decades of broadcast interpreting experience, accent and
voice are not likely to be an issue. For anyone "holding on to the job" for so
long, client and customer satisfaction with regard to these quality criteria can
clearly be assumed, since, "anything spoken into the microphone will be liste-
ned to by hundreds of thousands of viewers, who will not hesitate to report
critically if the interpreter's delivery and diction are not up to professional
broadcasters' standards." (Kurz and Pöchhacker 1995:351)
Fluency of delivery
The criterion of fluency as introduced by Bühler (1986) is actually a very

complex paralinguistic phenomenon which relates to such interdependent
features as pauses, speaking speed, voiced hesitation, false starts, slips and
repairs. While the relative weight of these factors in shaping the judgment of
fluency of a simultaneous interpretation is not clearly understood, these textual
and paraverbal parameters are at least amenable to quantitative analysis. The
case study data available for consideration concern tempo (speech rate),
pauses, voiced hesitation and false starts. The average number of syllables per
second is 2.01 in the original and 2.65 in the interpretation. To obtain a more
realistic indication of actual speech rates, the total number of syllables uttered
is adjusted for extended pauses (cf. Pöchhacker 1994:13If.). Using a — very
high — pause criterion of > 2 seconds, the average speech rate is 154 syllables
per minute in the original (with a total pause duration of 120.5 sec.) and 202
syllables per minute in the interpretation (with 157.5 seconds of pauses). While
the fact that the interpreter's tempo appears to be one third higher than the
speaker's may to some extent be due to the pause adjustment process
(unadjusted tempo: 121 syll./min. original vs. 159 syll./min interpretation), it
also reflects the lexical and structural characteristics of the German language,
as exemplified in the following passages from the corpus:
(1) thirty-three years since the Wall went up.
dreiunddreissig Jahre ist es her, da die Mauer errichtet wurde.
(2) when those trapped in the East threw stones at the tanks of tyranny.
als diejenigen, die im Osten gefangen waren, die Panzer der Tyrannei mit Steinen
bewarfen.
Since the interpretation contains approx. 20% more syllables and, at the same
time, one third more pause time than the original speech, the interpreter must
clearly be producing very dense speech bursts. This indication of fluency is
confirmed by the actual pausing pattern, which is best investigated here with
respect to the special delivery characteristics of the original. Of the 13
extended pauses (≥2 seconds) in the original, six are in the range of 2-5
seconds (Ø = 3.3 sec.), another six in the range of 11-15 seconds (Ø = 12.2
sec.), and one pause is of no less than 28 seconds in duration. (It occurs when
Clinton's address to the Germans culminates in the pledge "America is on your
side, now and forever", pronounced in German!) With only one exception,
each of these pauses has its more or less exact counterpart (typically + 0.5
sec.) in the interpretation. Only once the interpreter uses 4.5 seconds of the
speaker's 15-second pause to finish up a complex utterance ("scalding words
about race, ethnicity, or religion") after some initial hesitation. In addition to
these 13 speaker-induced pauses, the interpretation contains 12 pauses in the
narrow range of 2-3.5 seconds which have no counterpart in the original. The
majority of these are clearly a reflection of the interpreter's start-up distance
or time lag at the beginning of distinct utterances. Only about a third of these
processing-related "interpreter's pauses" can clearly be identified as hesitations.
Two of these occur in the following passage, which must be seen as almost the
only exception to the rule of smooth delivery in the case under study:
(3) The quiet courage to lift children above the Wall
Der stille Mut, Kinder über die Mauer zu lassen,
so that their grandparents on the other side could see those they loved but could
not touch.
so dass die Grosseltern auf der andern Seite sie wohl sehen aber nicht berühren
konnten.
It is unmistakably the ambiguous expression "lift children above the wall"
which made the interpreter think twice (3 sec.) and then opt for "über die
Mauer zu lassen", which, when it proved incongruent with the rest of the
utterance, engendered yet another (2.5-sec.) brief interruption in the flow of
the interpretation. It must be emphasized that none of the "interpreter's pauses"
occur within syntactic consituents, so that they are much more akin to ordinary
planning pauses in speech production rather than what Shlesinger (1994) refers
to as "interpretational intonation".
The entire 9-minute stretch of interpretation, which, unlike the original
speech, was produced impromptu without any written support, contains not a
single "false start" and only one instance of voiced hesitation (uh) in the
passage about "race, ethnicity, or religion" referred to above.
All things considered, the degree of fluency and smoothness of the
interpretation under study seems to be as close to the ideal as it could possibly
be.
Cohesion and consistency
It is worth noting that the two crucial demands made on simultaneous inter-
pretation by conference participants and TV professionals alike, i.e. sense
consistency with original message and logical cohesion of utterance, are met
in the case under study to the highest degree. One needs to mention that the
original speech contains many passages in which an idea is developed over a
number of parallel utterances. In two instances, the interpreter, unable to
foresee the unfolding of the rhetorical cascade, opted for non-matching clausal
structures, within which the original semantic material acquires a slightly
different though perfectly plausible meaning or sense. In one case, the sentence
"Already the new future is taking shape" was used by the speaker as a stem
from which four complement phrases, each beginning with "in", were to
branch out. The interpreter, who had ended the preceding sentence with a "full
stop" rather than a "colon", went on to use more self-contained structures by
rearranging or adding the appropriate syntactic constituents. In the example
given below, the predicate "gehören dazu" ("are part of it") is added to round
off the sentence:
(4) In the growing economies of Western Europe, the United States, and our
partners.
Die aufblühenden Wirtschaften Westeuropas, der Vereinigten Staaten und unserer
Partner gehören dazu.
In two other instances the slight deviations, if any, from the "sense" of the
original affect even smaller utterance segments. In one case the interpreter
clearly misperceived the words "no wall" in the original as "no one" and
produced the corresponding utterance in German. The resulting "sense
inconsistency" hardly deserves that label, since it makes little difference
whether "no wall" or "no one" "can forever contain the mighty power of
freedom." In the other case, the interpreter had just caught on to the overall
structure of a rhetorical cascade and made the appropriate output adjustments,
most likely with some compromise to listening attention. In the subsequent
utterance the indirect pledge of "American forces who will stay in Europe to
guard freedom's future" is turned into a politically neutral statement of
historical fact about the "amerikanischen Soldaten, die in Europa den Frieden
gewährleistet haben" (the American soldiers who have safeguarded peace in

Europe).
As regards the logical cohesion of the interpreter's utterances, there are
only two or three brief passages which might have resulted in a momentary
raising of eyebrows in a — very attentive — TV audience. One has already
been cited and discussed in example (3) above. Another, very similar case,
results from the ambiguity of the verb "go" in example (5). In the context of
the walled-in population of the former GDR, the interpreter takes the phrase
"can go as far as" to refer to the newly acquired freedom to travel:
(5) Where all our citizens can go as far as their God-given
wo alle unsere Bürger so weit reisen können, wie es ihre
abilities will take them
Fähigkeiten zulassen
When "abilities" complements the meaning of the utterance on a much more
general level, the interpreter chooses to accept the semantic incongruence of
"travel" and "abilities" rather than embark on a large-scale repair and repeti-
tion of the previous clause.
Completeness and correctness
In the list of quality criteria used for user expectation surveys the issue of com
pleteness appears as a separate, seemingly quantitative parameter. In fact,
however, it is closely tied in with the qualitative issue of "sense consistency",
since an omission of part of the original speech is likely to have some impact
on the sense of the message in the interpretation. The real issue implied by the
criterion of completeness is therefore that of message redundancy, i.e. the
question of what can safely be left out without detracting from the information
content of the speech.
It would be foolhardy to assume that an objective method for measuring
information content or a deficit thereof could be readily applied to the
comparative analysis of an original speech and its interpretation. Luckily,
however, the material under study presents few, if any, methodologically
challenging cases. On the contrary: The German interpretation is a remarkably
close rendering of the English original; it is so "faithful" that it contains hardly
any "omissions" which deserve that label. Take this example:
(6) We stand where crude walls of concrete separated mother from child
Wir stehen, wo Betonmauern Mütter von Kindern trennten
The alliterative adjective in "crude walls of concrete" has no counterpart in the

German interpretation, though one could hardly say that this makes "Beton-
mauern" any less crude and heavy.
Similarly, the correctness of both grammatical usage and terminology
leaves little to be desired in the interpretation under study. There is of course
no specialized terminology as such. Nevertheless the interpreter needs — and
clearly proves — to be well-versed in the rhetorical language and style
characterizing and indeed shaping the nature of such formal public addresses.
With only one or two inconsequential exceptions, the syntax of the interpreter's
output is as "correct" as that of the original speaker reading from a manu
script. The same holds true for speech production phenomena like slips of the
tongue. There is only one corrected slip ("WENN DIE NEME DIE NAMEN DER
PILOTEN" and one clear case of an uncorrected slip ("PARTNERSCHIFT" rather
than Partnerschaft), both of which can be attributed to phonetic interference
from the original ("names" and "partnership", resp.) in passages of particularly
demanding output production.
The paraverbal and textual parameters discussed in the preceding
paragraphs are summarized below by contrasting the data for the English
original with those for the German interpretation.
Table 1: Paraverbal and textual parameters (original vs. interpretation)
Parameter English original German interpretation
Total duration: 9 min. 12 sec. 9 min. 2 sec.

Length (in syll.): 1110 1330
Avg. syll./sec.: 2.01 2.65
Number of pauses (≥2 sec.): 13 25
Duration of pauses (total): 120.5 157.5
Tempo (adj.; syll./min.): 154 202
Voiced hesitation (uh): 0 1
False starts: 0 0
Corrected slips: 0 1
Sentence blends: 0 2
Uncorrected slips: 0 1
Conclusion
Irrespective of the particularly difficult and stressful circumstances under which

simultaneous interpreting for live broadcasts is typically carried out, media
professionals expect simultaneous interpreters to perform at a similar or even
higher level than in the conference setting, with particularly stringent demands
on the interpreter's pleasant voice, native accent and fluency of delivery. In a
case study of an authentic live broadcast interpretation these high expectations
were indeed found to be matched by a remarkable degree of smoothness and
fluency as well as accuracy and correctness of the interpreter's output. While
working conditions in this particular case may not have been as adverse as they
can sometimes be, these findings for a particular piece of English-German
simultaneous interpretation on Austrian television demonstrate that simultane-
ous "voiceover" interpreting of a live broadcast speech is possible at such a
level of quality that it implies few, if any, "inconveniences" to the broad-
caster's audience.
References
Alexieva, Bistra. 1988. "Analysis of the simultaneous interpreter's output". In: Paul Nekeman
(ed.) Translation, our future. Proceedings of the Xlth World Congress of FIT.
Maastricht: Euroterm, 484-488.
Bühler, Hildegund. 1986. "Linguistic (semantic) and extra-linguistic (pragmatic) criteria for
the evaluation of conference interpretation and interpreters". Multilingua 5 (4), 231-
235.
Bühler, Hildegund (ed.) 1985. Translators and their Position in Society. Proceedings of the
Xth World Congress of FIT. Wien: Braumüller.
Daly, Albert F. 1985. "Interpreting for international satellite television". In: Bühler (ed.), 203-
209.
Galli, Cristina. 1990. "Simultaneous interpretation in medical conferences: A case study". In:
L. Gran and C. Taylor (eds.) Aspects of Applied and Experimental Research on
Conference Interpretation. Udine: Campanotto, 61-81.
Gile, Daniel. 1995. Regards sur la recherche en interprétation de conference. Lille: Presses
Universitaires de Lille.
Kalina, Sylvia. 1994. "Analyzing interpreters' peformance: Methods and problems". In: Cay
Dollerup and Annette Lindegaard (eds.) Teaching Translation and Interpreting 2.
Insights, Aims, Visions (Benjamins Translation Library 5) Amsterdam/Philadelphia:
Benjamins, 225-232.
Kopcynski, Andrzej. 1980. Conference interpreting: some linguistic and communicative
problems (Seria Filologia Angielska 13) Poznan: A. Mickiewicz University Press.
Kurz, Ingrid. 1985. "Zur Rolle des Sprachmittlers im Fernsehen". In: Bühler (ed.), 213-215.
Kurz, Ingrid. 1989. "Conference Interpreting: User Expectations". In: Deanna L. Hammond
(ed.) Coming of Age. Proceedings of the 30th Annual Conference of the American
Translators Association. Medford NJ: Learned Information, 143-148.
Kurz, Ingrid. 1990. "Overcoming language barriers in European television". In: D. Bowen and
M. Bowen (eds.) Interpreting — Yesterday, Today, and Tomorrow (ATA Scholarly
Monograph Series IV) Binghamton NY: SUNY, 168-175.
Kurz, Ingrid. 1993. "Conference interpretation: Expectations of different user groups". The
Interpreters' Newsletter 5, 13-21.
Kurz, Ingrid and Pöchhacker, Franz. 1995. "Quality in TV interpreting". Translatio. Nouvelles
de la FIT — FIT Newsletter (Nouvelle série) XIV (3/4), 350-358.
Mona, Tiziana. 1995. "The choices and politico-cultural impact of simultaneously-translated
television programmes in Switzerland" [Ms.], publ. in French as: "Les choix et
l'impact politico-culturels de la traduction simultanée d'émissions de télévision en
Suisse". Translatio. Nouvelles de la FIT— FIT Newsletter (Nouvelle série) XIV
(3/4), 329-336.
Pöchhacker, Franz. 1994. Simultandolmetschen als komplexes Handeln. Tübingen: Gunter Narr
(Language in Performance 10).
Russo, Mariachiara. 1995. "Media Interpreting: Variables and Strategies". Translatio. Nou-
velles de la FIT— FIT Newsletter (Nouvelle série) XIV (3/4), 343-349.
Shlesinger, Miriam. 1994. "Intonation in the production and perception of simultaneous
interpretation". In: S. Lambert and B. Moser-Mercer (eds.) Bridging the Gap.
Empirical Research in Simultaneous Interpretation (Benjamins Translation Library 3)
Amsterdam/Philadelphia: John Benjamins, 225-236.
Vik-Tuovinen, Gun-Viol. 1995. "Progress in simultaneous interpreting — an evaluation of the
development of four students". Hermes. Journal of Linguistics 14, 55-64.
Bridging the gap: Verb anticipation in German-
English simultaneous interpreting
Udo Jörg
The process of anticipation is not confined to interpreting. In a more general

sense it could be defined as the prediction of speech units in a given situation.
Presumably, everyone has experienced this phenomenon in normal everyday
conversations (cf. Lederer 1978:330). At times, the listener knows what an
interlocutor is about to say. This may be explained by the fact that this person
and their ideas are very familiar, he or she is very repetitive, certain linguistic
clues indicate how the utterance will develop or a situation leaves only little
room for spontaneous, unpredictable twists. Most people have probably been
in situations where they were inclined to complete slow or hesitant speakers'
unfinished sentences. In other words, they anticipated speech units before their
actual utterance.
A rough and ready way of classifying anticipation would be to differentiate
between linguistic and extralinguistic anticipation (as in Gile 1995:176ff.), i.e.
speech unit prediction based on linguistic or extralinguistic clues respectively.
From a linguistic point of view, anticipation can be attributed to the probabili-
stic nature of speech comprehension (cf. Gile 1995:176, Hörmann 1971:79ff.).
Frequent sources of linguistic anticipation are collocations. Collocations are the
regular co-occurrence of lexical items and are thus a way of establishing
lexical cohesion in a given text (cf. Halliday and Hasan 1976:284). One of the
most common types of collocation is the verb-object collocation: a certain verb
evokes the use of a particular noun as its object, or vice versa (e.g. to pay —
a visit or a bill, depending on the context). A linguistic clue, in this case a
lexical unit, evokes the use of another lexical unit.
Linguistic anticipation skills increase with linguistic proficiency and
sensitivity, and can (and should) be maximised by means of sensitisation and
specific training.1 With extralinguistic anticipation on the other hand, predic-
tions cannot be based on linguistic clues, but the anticipator has to rely on his
or her knowledge of the speaker, the subject matter and the situation.
Anticipation in simultaneous interpreting
The process of anticipation, which takes place in speech and language

comprehension in general, is of particular relevance in simultaneous inter-
preting (SI). It can be defined as the prediction and interpretation of source text
(ST) units before their actual utterance and can be explained as a response to
previously received and processed linguistic and extralinguistic stimuli (based
on a similar definition by Wilss 1978:348).
As a process, SI requires from the interpreter the (near-)simultaneous
performance of various tasks (listening/analysis, memory and speech
production, cf. Gile's effort model 1995:162). In order to perform these tasks
adequately, the interpreter has to allocate sufficient processing capacity to each
of them. The interpreter should aim at minimising the individual efforts so as
to optimise his or her overall performance. The ability to anticipate (or in
Moser's words "predict") "greatly facilitates the interpreter's task" (1978:359)
and may help him or her to save precious processing capacity.
Chernov goes even further and sees anticipation ("probability prediction"),
together with message redundancy, as one of the main prerequisites for SI
(1978:53ff.). He regards the "probability prediction of the verbal and semantic
structure of the oral message in progress as the most essential psycholinguistic
factor explaining the phenomenon of simultaneity in simultaneous interpreting"
(1994:140).
Verb anticipation in German-English simultaneous interpreting
The underlying problem in German-English SI, which often compels

interpreters to resort to verb anticipation, is syntactic divergences between the
two languages. Complex German verb phrases can be split by objects,
complement phrases, participle constructions, relative clauses, etc. and thus do
not correspond to the English subject-verb-object pattern. In German, the
1
A small-scale study carried out by the author with final-year students at Heriot-Watt
University, Edinburgh, showed that linguistic anticipation (above all on the basis of
collocations) had been mastered to a greater extent than extralinguistic anticipation
(1993).
Udo Jörg 219
semantically relevant element of the verb phrase is often in end position (cf.
Wilss 1978:347), and, if the clause is convoluted with many subordinate
clauses or complementing phrases embedded in the complex verb phrase,
waiting for the main verb may overtax the interpreter's short-term memory and
result in a loss of information (cf. Kirchhoff 1976:61). One way out of this
dilemma is the anticipation of the verbal component at the end of the sentence.
Wilss investigated what he referred to as syntactic anticipation (anticipation
triggered off by syntactic cues) in German-English SI and concluded (mainly
on the basis of research work by Mattern 1974)
...that syntactic anticipation normally is something quite different from blind
textual hypothesising. It is rather the result of intelligent textual prediction
triggered by linguistic units (morphemes, lexemes or lexeme combinations)
which, within the framework of specific communication situations, serve as
important cues for the achievement of high-quality SI performance (1978:349).
The phenomenon of verb anticipation in German-English SI is occasionally
mentioned in the literature. Usually, however, it is not given extensive
coverage and the difficulties of the process sometimes tend to be played down
(cf. Lederer 1981:257 or Dalitz 1983:157).
Within the framework of this study, the intention was to shed some light
on the phenomenon of verb anticipation in German-English SI by means of an
empirical investigation.
Experiment
Source Text
As a source text (ST) I chose a speech delivered by the German President

Roman Herzog at a state ceremony in the Concert Hall in Berlin on 8 May
1995, on the occasion of the 50th anniversary of the end of the Second World
War.
The text was considered appropriate because it was a) topical and b) on
a subject which would not require special preparation from the interpreters.
Furthermore, its ideational development was clear and it was well formulated.
The speech seemed difficult enough to be a challenge, yet did not appear
unmanageable.
The original speech was abridged by approximately 30% so that the length
of the ST would not overtax the student interpreters. In a few places, in order
to make the task really testing, its syntax was altered. In this way, admittedly,
an authentic text was doctored, which some people might hold against the
validity of the experiment. However, all subjects completed a questionnaire
and commented on the speech after interpretation. Nobody complained about
lack of coherence or artificiality of the text, which I took as proof that my
editing did not have a noticeable effect on the text as a whole.
The duration of the speech was 17 minutes, 24 seconds. The speed of the
diction amounted to 115.23 words per minute, which, according to Gerver, lies
within the range of optimal input rate (between 95 and 120 words per minute)
for SI (1976:172). However, 10 out of 12 subjects thought the speech was
rapid (comments ranged from "a little fast", "fast at times", "very fast" to "too
fast"), which was probably due to the fact that it was read out and thus, as one
subject rightly put it, "perceived as fast".
The ST was not technical in nature. Mr Herzog referred to the Second
World War, its end and the historical development of West Germany and the
united Germany up to the present. The original speech was given on 8 May
1995, the British students' performances recorded at the beginning of June,
their Austrian counterparts' at the end of June, and the professional inter-
preters' recording dates ranged from end of June to beginning of August. The
speech was thus relatively topical for all subjects, slightly more so for the
student interpreters. This might have compensated for the fact that the
professional interpreters, being older and more knowledgeable, had probably
an advantage from a content point of view in connection with a speech about
the end of the Second World War and the ensuing implications.
The syntax of the ST was quite complex, and some of the ideas put
forward by Mr Herzog were relatively abstract and might have caused
comprehension problems. Vocabulary and register befitted both the speaker and
the occasion of a state ceremony.
Subjects
12 subjects took part in the experiment. 6 were interpreting students and 6

were professional interpreters. 3 of the students had English as a mother
tongue and were studying on the Postgraduate Diploma Course in Interpreting
and Translating at the University of Bradford. They had reached the final stage
of the one-year course and were about to take their final simultaneous
interpreting examinations. When interpreting the Herzog speech, they did not
know that their performances would be analysed within the framework of a
research project. They were intentionally not given a prior briefing so that they
would not adopt a 'slack' attitude and put less effort into the task than they
Udo Jörg 221
would in an examination situation. Likewise, the 3 Austrian interpreting

students were also only told about this project after the recording. They were
attending an advanced SI class at the Department of Interpreting and
Translating at the University of Vienna. Having been a student on both courses
myself, I can say that the level of the class and of the competence of students
are more or less comparable.
Of the 6 professional interpreters, 3 were English and 3 were native-
speakers of German (2 Austrians, 1 German). Their work experience ranged
from 6 to 21 years (average work experience: 14.16 years).
In order to persuade the professionals to participate in the experiment, I
had to tell them in advance that their performances would be recorded and
analysed. They were not, however, briefed about the specific nature of the
project and did not know that anticipation was the topic of the study.
The German mother-tongue professionals, like the student interpreters,
were recorded in university SI booths.
The English professionals, on the other hand, had to perform under the
least natural conditions. In order to minimise the subjects' inconvenience, they
were recorded individually by means of a twin-track recorder, head-sets and
microphone, at their respective homes the UK. One of the interpreters told me
expressly that the unauthentic nature of the interpreting situation had had an
influence on concentration and hence performance.
This argument, however, could be countered by the claim that a
professional interpreter should be able to perform in all places, provided
equipment, space, air, light and sound conditions are adequate. Nonetheless,
the artificiality of the simulated SI situation for the professional English
interpreters should be borne in mind.
While interpreting, the subjects did not have a copy of the ST.
Objectives
This work was intended as a descriptive, explorative and observational/ex-

perimental study. Rather than starting out with translational hypotheses, the
intention was to examine the results of the experiment as a whole and to try
to make inferences on this basis (as described by Gile 1994:50).
The initial objective was to discover whether verb anticipation took place
at all in German-English SI. And if it did, how common was it? Were there
any differences in verb anticipation occurrence between student and professio-
nal interpreters or between source language (SL) mother tongue and target
language (TL) mother tongue interpreters?
222 Translation as Jntercultural Communication
Procedure
The Herzog speech contained 26 sentences and clauses where verb anticipation
was likely to occur. In all these instances, complex German verbal structures
were split by other complex phrases which would have to be memorised unless
the verbal component in end position was anticipated. These 26 sentences and
clauses from the 12 recordings were transcribed.
In order to ascertain whether verb anticipation had taken place, three
categories for the simultaneously interpreted sentences were devised. One
category contained clauses where interpreters had successfully anticipated the
relevant German verbal component in end position (successful anticipation),
the second category comprised clauses where no verb anticipation had occurred
(no anticipation), and the third category consisted of those instances where the
verb had been wrongly anticipated (incorrect anticipation).
Category 1 (successful anticipation) was further divided into two subcategories
(exact anticipation and more general, but still successful, anticipation). The
German clause Millionen...waren zum Opfer gefallen, for example, was
interpreted (with anticipation) by one subject millions...had fallen victim to,
which would constitute a case of exact anticipation, whereas another inter-
preter's version millions...were destroyed by would also be a case of
successful but less exact anticipation. The former rendition would qualify as
exact anticipation, the latter as more general, but still successful anticipation.
I deemed the differentiation between exact and more general anticipation
necessary, as both types occurred frequently in the various interpretations, but
could be commented on differently. Whereas instances of exact anticipation
indicated that the subjects had indeed predicted the verb in an intelligent way,
i.e. by inferring it from linguistic and extralinguistic clues, the case was not
as clear-cut when it came to more general anticipations. At times, subjects
might not have anticipated in the way just described, but simply came up with
a make-shift, stop-gap verb, usually as non-committal and semantically void
as possible, and yet, in doing so, managed to stay in line with the gist of the
sentence and speech. In other words, instances of exact anticipation can be
regarded as evidence of intelligent predictions, whereas more general
anticipations might at times also be the result of damage minimisation
strategies, which are more erratic and much less based on inferences drawn
from linguistic and extralinguistic clues. For this reason it was thought
worthwhile to differentiate between the two types of successful anticipation.
Udo Jörg 223
Results
With 12 subjects and 26 sentences/clauses, the corpus contained a total of 312

anticipation possibilities. The distribution of successful anticipations (exact and
more general), no anticipations and incorrect anticipations are shown in Figure
1.
Figure 1. Distribution of anticipation categories
If we look at the three main categories of successful, no or incorrect

anticipation, we see that the successful anticipation score (156) equals the no
anticipation and incorrect anticipation scores taken together. Anticipation
actually occurred in exactly half of the 312 anticipation possibilities. Again, it
has to be borne in mind that the ST was a written speech with much less
redundancy than an impromptu text. The Herzog speech had been prepared in
advance and was thus more taxing and required a greater effort of com-
prehension than a spontaneous statement. If we regard the anticipation process
as a cognitive act triggered by various clues, it is only logical, in accordance
with Gile's effort model, that anticipation processes can be better carried out
if less effort has to be allocated to the comprehension process. One would
therefore expect anticipation rates for a spontaneously produced ST to be
higher (over 50% in anticipation-likely sentences) than in this corpus with a
written ST.
The cases of successful anticipation can be divided into approximately
40% general anticipation and approximately 60% exact anticipation.
Of all the cases where verb anticipation had been attempted (163 —
exact, more general and incorrect anticipation), only 7 (4.29%) misfired. From
this figure we can draw the conclusion that interpreters mainly anticipated in
cases when they were relatively certain about the verb to come. Otherwise the
failure rate (incorrect anticipations) would have been higher.
The next object of investigation was individual verb anticipation
performance. In Figure 2, anticipation scores (exact and more general) are
broken down according to individual interpreters.
Figure 2. Individual anticipation scores
In Figure 2, two scores lie outside the field of the standard deviation (standard
deviation: 4.05; outsiders: SG3 & PG32). The remaining ten scores are
relatively evenly distributed around the mean value of 13 (corresponds to
anticipation in every other anticipation-likely sentence). A closer look reveals
that the scores of the professional interpreters cluster more closely around the
mean value, which indicates that their anticipation performance is more
consistent and regular than that of the student interpreters.
The fact that there are two extreme values, one on either side beyond
the standard deviation, could be interpreted as evidence that every interpreter
Whenever reference is made to individual performances, the following kind of

abbreviation will be used: a first letter to designate status (S for student or P for
professional), a second letter for mother tongue (E for English or G for German) and
an arbitrarily allocated figure to differentiate between individual interpreters within
the various subgroups.
Udo Jörg 225
has an individual anticipation potential. If it is possible to anticipate 22 verbs

in a total of 26 anticipation-likely sentences (84.62% as achieved by superanti-
cipator PG3), underachieves (like SG3 with a score of 6 anticipations or
27.27% ) could see this as a challenge and work on their anticipation skills.
Figure 2 also reveals that the anticipation scores of interpreters with
English as a mother tongue are more evenly distributed than those of their
German mother tongue counterparts. However, if the average anticipation
score per mother tongue (English: 12.83, German: 13.18) is calculated, the
difference of less than 0.4 anticipations seems to be negligible. SG3 with
his/her poor anticipation score has had an unfavourable effect on the German
mother tongue value. The German mother tongue average without that
particular score would amount to 14.6, corresponding to almost two more
anticipations on average for interpreters working from their mother tongue.
This might indicate that anticipation processes function better when working
from one's mother tongue.3
Anticipation scores broken down according to mother tongue might
shed more light on this question.
Figure 3. Distribution of anticipation categories
a) English mother tongue
This is in line with some of Chernov's findings about probability prediction. In his
experiments, simultaneous interpreters interpreted in 50.89% of cases according to
their predictions when working from their mother tongue, as opposed to 28.68 % when
working into their mother tongue (1978:82).
b) German mother tongue
Figure 3 shows that in spite of the fact that the total number of verb anticipa-
tion instances is almost identical for English and German mother tongue
interpreters, the latter have over 10% more exact anticipation scores than the
former, thus confirming the hypothesis that anticipation skills (in the sense of
intelligent predictions) are probably better developed in the mother tongue.
This statement could also be underpinned by the distribution of incorrect
anticipation scores as shown in Figure 4.
Figure 4. Individual anticipation scores: incorrect anticipation
The scores for incorrect anticipation are in general very low. It is almost non-
existent in native speakers of German (one single incident of incorrect
Udo Jörg 227
anticipation in SGI) and occurs, if only rarely, more often in the English
mother tongue students than in the English mother tongue professionals.
This shows that anticipation certainty seems to be higher when
interpreting from one's mother tongue. When working from a foreign
language, professionals seem to be better at anticipating. The scores for exact
anticipation are also worth examining:
Figure 5. Individual anticipation scores: exact anticipation
Figure 5 demonstrates quite clearly that students hit the absolutely correct verb
less frequently than professionals (only one student scores above-average and
only one professional scores below-average in exact anticipations). From this
we can conclude that, as a rule, anticipation proficiency increases with
experience, even though (as SG2 shows) exceptional performances by non-
experienced interpreters are possible.
In summary, then, the performances of 12 interpreters (students and
professionals), who interpreted a political speech which was read out when
they did not have a copy of it, revealed the following:4
On average, verb anticipation took place in half of the anticipation-
likely sentences. In more spontaneous speeches, anticipation would probably
be more frequent. ■■
Even though there is an individual anticipation potential, it was found
that anticipation performance was more consistent among professional
interpreters than among students.
For further analyses and results relating to verb anticipation in German-English SI and
correlations with quality features such as precision, style, fluency of delivery, etc. cf.
Jörg 1995.
Moreover, verb anticipation proficiency, i.e. the degree of exactitude

in anticipating the verb, tended to increase with interpreting experience, and
various factors suggested that verb anticipation skills were better developed in
native speakers of the SL.
References
Brislin, Richard W. (ed.) 1976. Translation — Applications and Research. New York:
Gardner Press.
Chernov, Ghelly V. 1978. Teoriya i praktika sinkhronnogo perevoda (Theory and practice of
Simultaneous Interpreting). Moscow: Mezhdunarodniye Otnosheniya.
Chernov, Ghelly V. 1994. "Message Redundancy and Message Anticipation in Simultaneous
Interpretation". In: S. Lambert/B. Moser-Mercer (eds.), 139-153.
Dalitz, Günter. 1983. "Deutsche erweiterte Attribute beim Simultandolmetschen". Fremd-
sprachen 27/3, 157-162.
Drescher, Horst W./Scheffzek, Sigrid (eds.) 1976. Theorie und Praxis des Übersetzens und
Dolmetschern: Bern: Peter Lang.
Gerver, David. 1976. "Empirical Studies of Simultaneous Interpretation: A Review and a
Model". In: R.W. Brislin (ed.), 165-207.
Gerver, David/Sinaiko, H. Wallace (eds.) 1978. Language Interpretation and Communication.
New York: Plenum Press.
Gile, Daniel. 1994. "Methodological Aspects of Interpretation and Translation Research". In:
S. Lambert/B. Moser-Mercer (eds.), 39-56.
Gile, Daniel. 1995. Basic Concepts and Models for Interpreter and Translator Training.
Amsterdam/Philadelphia: John Benjamins Publishing Company.
Halliday, M.A.K./Hasan, Ruqaiya. 1976. Cohesion in English. London/New York: Longman.
Hörmann, Hans. 1971. Psycholinguistics —An Introduction to Research and Theory. Berlin:
Springer (translation by H.H. Stern of Psychologie der Sprache, Heidelberg:
Springer, 1967).
Jörg, Udo. 1993. Syntactic and Semantic Anticipation in Simultaneous Interpreting. Heriot-
Watt University Edinburgh: Unpublished Paper.
Jörg, Udo. 1995. Verb Anticipation in German-English Simultaneous Interpreting: an
Empirical Study. University of Bradford: Unpublished MA Dissertation.
Kirchhoff, Helene. 1976. "Das Simultandolmetschen: Interdependenz der Variablen im
Dolmetschprozeß, Dolmetschmodelle und Dolmetschstrategien". In: H. W. Drescher/
S. Scheffzek (eds.), 59-71.
Lambert, Sylvie and Moser-Mercer, Barbara (eds.) 1994. Bridging the Gap — Empirical
Research in Simultaneous Interpretation. Amsterdam: Benjamins.
Lederer, Marianne. 1978. "Simultaneous Interpreting — Units of Meaning and other
Features". In: D. Gerver/ H.W. Sinaiko (eds.), 323-332.
Lederer, Marianne. 1981. La traduction simultanée: expérience et théorie. Paris: Minard.
Mattern, N. 1974. Anticipation in German-English Simultaneous Interpretation. Saarbrücken:
Manuscript.
Moser, Barbara. 1978. "Simultaneous Interpretation: A Hypothetical Model and its Practical
Application". In: D. Gerver/H.W. Sinaiko (eds.), 353-368.
Wilss, Wolfram. 1978. "Syntactic Anticipation in German-English Simultaneous Interpreting".
In: D. Gerver/H.W. Sinaiko (eds.), 343-352.
A thinking-aloud experiment in subtitling
Irena Kovačič
Although subtitlers themselves usually claim that they are not aware of using
any particular strategies in the process of deciding what from the original text
(dialogue) they are going to preserve in the subtitle, what to render in a
reduced form and what to leave out altogether, comparison of subtitles with
original texts reveals consistent regularities (Kovačič 1992). Could scrutiny of
the subtitling process itself reveal more about the mechanisms underlying
subtitlers' selections and decisions while translating? This decision-making
process may be regarded as a typical example of problem-solving behaviour,
and as such eligible for investigation by thinking-aloud experiments. Subtitling,
like any translation, is a case of the same input possibly (usually) leading to
different outputs. Mere observation of differences among the outputs cannot
tell us much about how they came about. Therefore, as Lörscher (1991:48)
puts it, "the assumption seems justified that the quality of the hypotheses on
the processing of information on the brain can be improved by taking into
account data from the production process which the language user has been
asked to externalise while producing the language output". However, it is
necessary to emphasise that such experiments can be regarded only as an
attempt to gain some insight into the process and by no means as a definitive
account of the functioning of the subtitler's brain. Mental processes can be
influenced by many different factors, frequently unobservable, which cannot
be controlled in an experimental situation.
Thinking-aloud experiments
The thinking-aloud method is basically an attempt to externalise internal

(mental) processes. It was first introduced by Gestalt Psychology in the first
half of the 20th century as a method of investigating thought processes,
particularly in the course of problem solving (for a survey of the history and
the major problems of the method see e.g. Börsch 1986 or Lörscher 1991:48-
55).
In this empirical method, experimental subjects are asked to verbalise
anything they are aware of as going on in their minds while they are solving
a problem. Their verbalisations are recorded and protocols are made of the
recorded material. These Thinking-Aloud Protocols (TAPs) are analysed both
for recurring patterns and idiosyncratic features in the hope that the findings
will help to shed light on the mental processes underlying the recorded
verbalisations.
One of the central controversies regarding the TAPs method has been the
issue of what can be verbalised (Ericsson and Simon 1984). The currently
prevailing view is that experimental subjects hold some thoughts in short-term
memory and this information is accessible for verbal reports; part of that
information is moved to long-term memory and may be retrieved
retrospectively (Jääskeläinen and Tirkkonen-Condit 1991:90). If a thought is
not attended to in short-term memory while a task is being performed, it
cannot be verbalised, for a number of reasons. One is automation of cognitive
processes, which is a consequence of repetitive performance of identical or
similar tasks. According to Börsch (1986:207), this is also what happens in the
case of experienced translators.
TAPs research in language-related fields
In language-related fields, TAPs began to appear in the 1980s, mainly in three

domains:
— literary reading, esp. reception theory and reader response criticism,
— second language learning,
— translation as process (Lörscher 1991 — a project started in 1983; House
1988; Tirkkonen-Condit 1991; Jääskeläinen and Tirkkonen-Condit 1991).
In all this research, TAPs were painstakingly analysed to provide insight into
processes underlying various linguistic activities. The procedures used in
individual studies differed in a number of parameters, but the basic premise of
research remained the same: verbalisation as an overt manifestation of
otherwise unobservable mental processes. This premise has been questioned by
a number of critics; for some of the problems and controversies see e.g.
Börsch (1986:200).
Irena Kovačič 231
In the field of translation studies, the most explicit criticism comes from
Toury (e.g. Toury 1991, slightly revised in Toury 1995:234-9). Toury holds
that in the case of verbalised translations, the main problem is "the possible
interference of two modes of translation", claiming that (in experiments
"involving the gradual production of a written translation of a written text")
"the need to verbalize aloud forces the subjects to produce not just mental, but
spoken translation before the required written one; and that there is a real
possibility that spoken and written translation do not involve the exact same
strategies" (Toury 1995:235; italics in the original). I will argue below that this
objection does not hold for TAP experiments in subtitling, where translators
attempt to reproduce a written version of dialogue, which is spoken discourse
by its very nature. Subtitles appear to be that strange marginal hybrid of
spoken-written translation characterised by external symptoms of 'inner
speech', whose existence Toury (1995:236) is willing to recognise, yet
questioning "its relevancy for the establishment of a general 'psycholinguistic'
model of translation".
The experiment
The objective of our experiment was twofold:

— primarily, to find out what types of verbalisation accompany the subtitling
process; i.e. what types of thoughts occurring in the subtitler's mind
during his or her work surface in this experimental procedure;
— in the second place, to see whether there is any significant difference in
automation of subtitling processes correlated to the amount of experience
in subtitling comparable to differences found in other TAP experiments in
translation (see above).
Description of the experiment
Six subtitlers were asked to subtitle in Slovene a passage from the 1986 Miles
Company television adaptation of a Broadway production of Eugene O'Neill's
Long Day's Journey into Night, starring Jack Lemmon, Bethel Leslie, Peter
Gallagher and Kevin Spacey. With its rapid dialogues, abundant overlapping
and American culture-specific concepts, this play seemed particularly
appropriate for the experiment. It could be anticipated that the subtitlers would
have to apply a number of specific subtitling procedures to allow Slovene
viewers access to the meaning of the original text, notably:
— reductions due to space/time restrictions;

— selection of relevant utterances in overlapping sections of dialogue in
terms of discourse coherence;
— explication and modification of culture-specific terms.
In the central stage of the experiment, having parsed the text into subtitling
units by themselves, the subjects were asked to translate the dialogue of the
drama in the form of subtitles. TAP experiments are often restricted to oral
work only. Subtitling, however, is so crucially dependent on the space
dimension (two lines of a maximum of 32 symbols) that asking people to
produce subtitles only orally would create a completely abnormal situation,
since the basic framework for their work would be taken away from them.
Looking for solutions appropriate for restricted space is a fundamental
parameter in subtitling. The experimental subjects were therefore asked to use
computers as usual, verbalising what they were aware of as going on in their
minds, not only directly related to the text, but any thought they could register.
Experimental subjects
In order to test for the degree of automation in subtitling, the experimental

subjects were selected according to their subtitling experience: subjects A and
B (beginners) had less than a year's experience and had subtitled less than 20
hours of material; subjects C and D (moderately experienced) had more than
a year and less than 5 years of experience in subtitling, with between 100 and
120 hours of material subtitled; and subjects E and F (experienced sub titlers)
had been sub titlers for more than 5 years, with over 200 hours of material
subtitled. This classification was not intended in any evaluative way. Its only
purpose was see whether there exist any noticeable differences in the amount
of conscious (verbalised) manipulation of the material that could be related to
the experience in subtitling. The groups were defined arbitrarily and further
experiments with more subjects would be needed to establish more exact cut-
off lines between beginners, moderately experienced and experienced subtitlers.
Discussion of the results
Automation of the subtitling procedure
The TAPs obtained in the experiment confirm that less experienced subtitlers
are a richer source of introspective data (cf. e.g. Lörscher 1991:35 for an
Irena Kovačič 233
argument in favour of using non-translators as experimental subjects; and

Jääskeläinen and Tirkkonen-Condit 1991 for a comparison of novice and
experienced translators). The experienced subtitlers apparently had their mental
processes automated to such an extent that they could not verbalise them.
Subject F (25 years of experience, the only full-time subtitler in the
experiment) subtitled almost directly, 'self-dictating' as she was typing.
Subject E, who was in the structure of the experimental group an additional
intermediate case between the moderately experienced C and D on the one
hand and F on the other (7 years of experience, approximately 200 hours of
subtitled material), did do some thinking-aloud, but also had considerable
problems observing the procedural instructions. Rather, he very quickly
developed a method of his own: he translated a section of the text, usually a
subtitle or a line in a subtitle, including changes and corrections, and after that
offered some retrospective comment on what he had done or how he usually
proceeds in similar cases.
This lack of verbalisation and 'repair' retrospective verbalisation
respectively seem to confirm that in experienced subtitlers the subtitling
routines become so internalised and automated that conscious manipulation of
material only takes place in cases of difficulty.
Types of verbalisations
Verbalisations in protocols could be divided into several major categories

(these categories had not been determined in advance but appeared in the
course of protocol analysis):
1. common translation problems (plot analysis; 'translation equivalents');
2. subtitling-specific problems of subtitle units (how to cut the dialogue into
individual subtitles);
3. related to 2, subtitling-specific 'condensation' problems (how to squeeze
the text within the limited space of a subtitle);
4. problems of spoken-to-written transfer (how to capture the flavour of
spoken discourse in the written form of subtitles);
5. execution-related problems (typing errors, outside noise, etc.)
These categories frequently overlap; e.g. condensation is related to plot
analysis (what may be left out with the least damage to the comprehensibility
of the story) or to written-to-spoken transfer (colloquial language is less packed
than written), and the like.
1. Common translation problems were not within the scope of our research
interest, so they were just recorded but not analysed. Less experienced
subtitlers typically sought help from dictionaries (either traditional or
electronic) much sooner than more experienced ones, who first searched in
their memory or used deduction.
Plot (discourse) analysis is particularly salient in subject E's protocol, to
a large extent due to the fact that his verbalisations were only partly
introspective and much more retrospective. Sometimes he first performs the
analysis and then writes the translation; even more frequently he first
completes a subtitle and then explains why he focused on a certain part of an
utterance.
In other subjects' protocols, discourse analysis is typically verbalised
indirectly (when subject B wonders "Is this important here? Is he an athlete?"
because she needs to decide which part of the text to leave out; or when both
D and E mimic the character's words in exaggerated intonation).
2. Subtitling units are not fixed in advance. Even those subtitlers who
mark off subtitle borders in the dialogue list while first viewing a programme
frequently change the structure later. This may depend on the length of an
utterance or phrase in the target language or on a different analysis of the
dialogue structure. In the former case a section of the text that is indispensable
for the story and is relatively short in the original becomes much longer in
translation and something in the text has to be 'sacrificed'. In the latter case,
analysis in terms of adjacency pairs, dramatic pauses and other elements of
spoken discourse that were not noticed during the first viewing may lead to a
change in the original grid of subtitles.
The protocols show that subtitlers do not verbalise subtitle borders in
advance. Typically, they do not dwell on them except when problems occur,
due to either of the above mentioned reasons. In such cases, experienced
subtitlers hesitate much less and calibrate their subtitles without lengthy
verbalisations, frequently going one or two subtitles back and rearranging
them. Beginners, on the other hand, either put down a translation as they have
originally conceived it and leave it for later editing ("Let's do it like this; we'll
reduce it later. We'll have to shorten this quite a lot."), or try out a number
of alternatives before making up their mind. In general, subtitle arrangement
seems closely associated with discourse structure and it is only within the
framework of larger text units that their optimal structuring becomes evident.
The protocols contain a number of statements of the type "Now let's move that
sentence down", (i.e., the second part of a subtitle is made the initial part of
the next subtitle) or "We should go back and split that one in two," which
seem to imply that it is only from the perspective of subsequent text that the
division of previous subtitles becomes more clearly evident.
Irena Kovačič 235
3. The 'condensation' procedures are usually wrapped up in a cloud of

mystery. When asked about their approach to reducing the text of the original,
subtitlers are very evasive. They say that their choice is led by their feeling of
"what is more important", but how they determine this remains unclear. A
previous product-based study (Kovačič 1992) showed that certain consistent
priorities in this 'importance' can be established in terms of linguistic and
discourse functions. The current process-oriented study suggests the following
conclusions:
— Translation and condensation seem to be two originally separated
processes, but with experience they gradually merge. A beginner will
typically first verbalise unabridged translation and then try to decide how
to change it. The more experienced a subtitler, the more straight-forward
is his or her verbalisation of the final text of a subtitle.
— When a section of text is too long for a subtitle, less experienced subtitlers
are more likely to verbalise a complete translation, then say something
like "This will be too long" or "I have to make this shorter" and 'try out'
various alternatives. More experienced subtitlers either verbalise several
alternatives without any additional comment and put down the one they
find the most appropriate, or simply start typing a version and when they
realise it is going to be too long, they stop briefly and delete or rewrite
part of the text.
With the exception of subject E, no subtitler, either beginner or more
experienced one, specifically verbalises any analysis in terms of functional,
pragmatic, semantic, or syntactic categories (although these categories may be
established through analysis of subtitles as a product). Subject E seems to be
very conscious of the structure of subtitles in terms of discourse organisation
(having the first and the second part of an adjacency pair in the same subtitle;
noticing a move which indicates a shift in conversation topic; focusing on
words which seem metaphorical of the relations among the characters). No
other protocol contains a metalinguistic statements of the type "element B is
less important than element A, so I am going to drop B", or "this part is only
a link to what he says next; no great damage if it isn't there". These
deliberations can only be suspected behind the verbalisation of various versions
which actually differ in these respects. Sometimes they are verbalised
indirectly, as in the already mentioned example of subject B wondering
whether "this is important". If it is true that only those elements of the process
get verbalised which are cognitively controlled (Ericsson and Simon 1984:90),
a general conclusion of the experiment may be that the decisions regarding
reductions of text in subtitles, register selection and the like are made
intuitively, on the basis of the subtitlers' general experience in communication.
The reasons for this can be twofold: (1) the general communication processes
are automated to such a degree that they are no longer accessible to
verbalisation, or (2) lack of systematic awareness of (formal training in)
functional/discursive power of language makes verbalisation impossible.
4. Observations regarding transfer from written to spoken discourse are
twofold: some can probably be also generalised to other languages, while some
are specific of subtitling into Slovene (and possibly other languages with the
so-called flexible word order). The protocols show that subtitlers 'try out' a
number of alternatives, characteristically in search of a balance between
economy of expression and colloquial register. It is well known that the so-
called lexical density of colloquial language is much lower than that of formal
language (Halliday 1985:61-64) and as such an additional problem facing the
subtitler who wants his text to sound natural, i.e. conversational, and yet be
sufficiently short to fit the limited space.
The next problem is the information conveyed by so-called suprasegmental
elements of language, especially intonation and stress. What spoken language
can convey by intonation, has to be signalled in written form by punctuation,
special syntactic patterns or (in a language like Slovene) marked word order.
The protocols contain a number of examples of the subjects trying to find an
optimal version which would also contain the suprasegmental information of
the original. In verbalisations of these options, intonation and emphasis still
play a very important role. The subtitlers pretend to be the characters from the
dialogue, varying the intonation pattern and emphasis before they put down a
subtitle or as they are writing it. The word order they produce is sometimes
very different from the word order typically found in dialogues in books,
which indicates that they are trying to reproduce spoken language, i.e.
something close to what Toury (1995:236) called "a kind of 'spoken-written'
translation".
5. Execution-related verbalisations are of two kinds. The first group
contains what might be called 'self-instructions', typically associated with
typing ("Here's another space." "Let's go on. Page 2.") or using dictionaries
("Let's see what the dictionary says."). Here belong also exclamations of
satisfaction at finding a good solution or expressions of frustration, including
cursing. The second group is related to the subjects' awareness of being in an
experimental situation ("Damn it, Irena, this is tough!" "In a situation like this,
I usually..."). This latter group is a warning signal that no matter how
objective we may try to be in designing and carrying out an experiment of this
kind and in interpreting its data, there will always be some extra component
Irena Kovadië 237
in the protocols which arises from the subjects' awareness of being observed
and analysed.
Conclusion
On the most general level, the initial analysis of the protocols in our
experiment yields results that are consistent with findings of similar research
in other fields of translation: a significant amount of the translation/subtitling
process does not get verbalised; the more experienced an experimental subject,
the less he or she verbalises. What prevails in verbalisations, is 'browsing'
through various straight-forward alternative wordings of a subtitle. In this,
particularly salient is the search for optimal balance between the need to reduce
and the need to sound natural, which coincides with the choice between
registers of written and spoken discourses. Among more analytical thoughts,
efforts to understand plot or discourse relevance of individual utterances is
verbalised, sometimes directly, even more frequently indirectly. Not
verbalised, however, are 'categorisablé' criteria underlying the final selection
among the several options reviewed in the browsing section of verbalisation.
Further experiments, in combination with interviews and product-oriented
analysis (i.e. analysis of subtitles) will be needed; what seems obvious at this
stage is that one of the most interesting aspects of the subtitling process, viz.
what guides subtitlers in their selective translation, is not accessible through
TAPs.
References
Borsch, Sabine. 1986. "Introspective methods in research on interlingual and intercultural

communication". In: J. House and S. Blum-Kulka (eds.), 195-209.
Ericsson, K. Anders and Herbert A. Simon. 1984. ProtocolAnalysis. Verbal Reports as Data.
Cambridge, Mass.: MIT.
Halliday, Michael A. K.. 1985. Spoken and Written Language. Oxford: Oxford University
Press.
House, Juliane. 1988. "Talking to oneself or thinking with others? On using different thinking-
aloud methods in translation". Fremdsprachen Lehren und Lernen 17, 84-98.
House, Juliane and Blum-Kulka, Shoshana (eds.) 1986. Interlingual and Intercultural
Communication. Discourse and Cognition in Translation and Second Language
Acquisition Studies. Tübingen: Gunter Narr.
Jääskeläinen, Riitta and Tirkkonen-Condit, Sonja. 1991. "Automatised Processes in
Professional vs. Non-Professional Translation: A Think-Aloud Protocol Study". In: S.
Tirkkonen-Condit (ed.), 89-109.
Kovačič, Irena. 1992. Jezikoslovni pogled na podnaslovno prevajanje televizijskih oddaj.

(Linguistic Aspects of Subtitling Television Programmes.) PhD thesis. Ljubljana:
University of Ljubljana, Faculty of Arts.
Lörscher, Wolfgang. 1991. Translation Performance, Translation Process, and Translation
Strategies Investigation. Tübingen: Gunter Narr.
Tirkkonen-Kondit, Sonja (ed.) 1991. Empirical Research in Translation and Intercultural
Studies. Tübingen: Gunter Narr.
Toury, Gideon. 1991. "Experimentation in Translation Studies: Achievements, Prospects and
Some Pitfalls". In: S. Tirkkonen-Condit (ed.), 44-66.
Toury, Gideon. 1995. Descriptive Translation Studies and beyond. Amsterdam: Benjamins.
Comprehension processes and translation. A think-
aloud protocol (TAP) study
Paul Kussmaul
For some time now in translation studies the comprehension process has
become a focus of interest. Psycholinguistic notions such as bottom-up and
top-down processes, prototypes, scenes-and-frames, schemas and scripts
have been applied to the translation process (e.g. Vannerem/Snell-Hornby
1986; Snell-Hornby 1988:79ff.; Neubert 1988; Vermeer/Witte 1990; Kuß-
maul 1994:9ff.; Kußmaul 1995). We are turning away from purely
linguistic, system-based models to those which include the language user.
This is surely the right thing to do. There has been little work, however, on
observing how this process actually works when we translate.
On a small scale I have tried to do just this. In this paper I shall report
on TAP-experiments that focus on comprehension. My first hypothesis is: If
we can observe "scenic" notions in the TAPs, then there is a good chance
that subjects will understand the text correctly. A correct understanding will
in turn lead to a good translation. Making use of scenic notions would thus
be part of a successful translation strategy and could be used for training
purposes. If, however, we can observe that in spite of correct
comprehension, the text is not translated in a satisfactory way, the subjects
most likely will have had problems with verbalising what they have
understood. By analysing the TAPs we might be able to find some of the
causes of their problems.
My other hypothesis concerns creativity. Creativity is usually regarded
as a special gift or at least a special state of mind. It has always had strong
connections with art. The processes involved in producing creative products,
so the thinking goes, do not just happen in the normal course of events.
They have a semi-mystical flavour about them. Words like "inspiration" and
"illumination" suggest that these processes are out of our conscious reach.
Stories about creative ideas in dreams support this opinion. If we want to be
creative, special arrangements such as brainstorming sessions have to be
made. I have a feeling that creativity is not all that special. It is an everyday
affair. My observations of comprehension processes, which are basic to the
human mind, and the resulting translations suggest that they involve a large
amount of creativity, and if we make conscious use of these comprehension
processes, we can be creative as translators.
I examined 4 dialogue protocols of students in their 3rd year who were
acquainted with psycholinguistic notions and models. The text to be
translated described the economic improvement in the eastern parts of
Germany. It ran:
Sleek new cars speed along straightened and repaved autobahns. Shiny service
stations come equipped with well-stocked convenience stores and gleaming
self-service restaurants. Enormous supermarkets, furniture stores and shopping
emporiums dot the east German landscape, and giant cranes stand tall against
the sky. Every seat is filled at Dresden's magnificent neoclassical opera house:
comfortable burghers sip French champagne during the intermissions. Even in
grimy Bitterfeld, a mining and chemicals centre notorious for its pollution,
well-dressed women from a nearby retirement home gather for creamy coffee
and gigantic pastries at a Swiss-owned coffee shop. {Newsweek, February 28,
1994, p. 14)
The subjects were given the fictitious translation assignment that they should
translate the text for the News Department of the German Government
under the general heading "How Germany is seen abroad". When analysing
the protocol I examined how the following phrases of the text were
understood and translated:
1. Shiny service stations ... gleaming self-service restaurants
2. well-stocked convenience stores
3. comfortable burghers
4. gigantic pastries.
These items, I found, were especially suited to create a scene in Fillmore's
sense in the reader. Given 4 text items and 4 protocols there were 16
solutions altogether. 8 of them were good, 4 were satisfactory, and 4 were
not quite bad, but could have been improved. How were the 8 good
translations achieved? 6 of them were prepared by the subjects visualising a
scene, for 2 of them no scene could be observed, which does not mean that
it was not there; it only means that it was not verbalised by the subjects.
Let us look at some of the examples in the TAPs were scenes appeared.
Paul Kussmaul 241
Successful processes
"well-stocked convenience stores "
When translating this phrase some of the subjects visualised scenes which
led to good translations. I shall pick out and discuss two possibilities of
creating a scene (for the term see Fillmore 1976, 1977). A scene can
originate in the translator's own personal experience and knowledge stored
in his or her long-term memory. It can also originate in the translator's
experience and interpretation of the given text stored in his or her short-
term memory. Here is an example of the first type. The subjects were not
quite happy with their translation "gutausgestattete Geschäfte" for "well-
stocked convenience stores":
B: convenience stores? Man könnt' höchstens noch anhängen: Dinge des
täglichen Bedarfs.
A: Aber das ist ja nicht unbedingt nur so, also in Raststätten, diese
Geschäfte, z.B. an den Tankstellen, die haben ja auch Sachen, die du
nicht für den alltäglichen Bedarf brauchst, irgendwelche besondere
Geschenke, Landkarten.
B: aber auch in Kühltheken, Joghurt, Getränke und Zeitungen, alles
mögliche, ich weiß nicht wie man das passender ausdrücken könnte, oder
wir lassen's dann einfach.
A: Man kann's umschreiben: mit gut ausgestatteten Geschäften, in denen
man alles finden kann, was das Herz begehrt.
B: Ja, oder so.
A: So könnte man's umschreiben.
B: Ja, ich glaub, das wär doch, das ist gut (Text wird wiederholt)
(Lachen)
A: Oi, oi. Jetzt sind wir fertig.
This protocol passage is well suited to show the connection between
comprehension and creative processes. When visualising the scene the
subjects here make use of divergent or lateral thinking as opposed to
convergent or vertical thinking (for the terms see Guilford 1975; de Bono
1970). If the subjects had used vertical thinking, they might have proceeded
in the way a dictionary does and tried to find a definition for well-stocked
based on a logical sequence of semantic features, such as: "having a
sufficient supply of goods for future use". The list of scenic details the
subjects mention in our protocol, however, cannot be reached by convergent
thinking. There is no direct logical link between the notion well-stocked and
words like maps or yoghurt or newspapers. Such words can only be arrived
at by thinking "horizontally" by associations. Nevertheless, this is not a

random list. The items are linked together by the motorway-store scenario (I
use the term in the sense of larger conceptual structures, cf. Bobrow/Brown
1975:23), which for instance, at least in Germany, excludes things like
fresh meat, vegetables, clothing, furniture, or gardening tools. As a matter
of fact, there is great affinity between the guiding principles of the search
for these details and creativity tests, where subjects for instance are asked to
write down things that are white and edible. By the use of two conditions
the selection of items is limited. In the same way, there are two conditions
for the search of our subjects. They name things that (a) can be sold and
that (b) can be found in motorway stores.
In this very vivid and detailed description of the shop scenario we can
observe another aspect of creativity, namely fluency of thinking (for the
term see Guilford 1975). As has been observed in creativity research, our
thought processes not only have to be divergent but also fluent, that is, our
ideas should occur in a relatively short space of time. This can be observed
in the present case. The subjects here jointly and speedily list a large
number of articles that can be purchased in these stores (Geschenke,
Landkarten, [Lebensmittel in] Kühltheken, Joghurt, Getränke und
Zeitungen), and, with their thinking processes speeded up, this immediately
leads to the solution: "in denen man alles finden kann, was das Herz
begehrt". The translation is also marked by divergent thinking. Instead of
focussing on the goods on the shelves, the subjects focus on the customers,
their wishes and needs, which is an alternative and perhaps even more
successful way of dealing with the same thing, namely expressing the notion
of affluence which pervades the whole text about eastern Germany.
"shiny service stations ... gleaming self-service restaurants "
The second type of scene, which results from the interpretation of the text
in question, was observed in another protocol when the subjects were
discussing the translation of "shiny service stations ... gleaming self-service
restaurants":
B: Es geht jetzt eigentlich eher um den Wohlstand in den neuen Ländern,

wenn ich das richtig sehe.
A: Vielleicht könnte man das "shiny" und "gleaming" irgendwie

zusammennehmen (Pause)
B: Hmmmm. Ich weiß nicht, aber
Paul Kussmaul 243
A: Ich meine, das Hauptaugenmerk ist ja da drauf, daß alles brandneu und
wunderbar und alles schön und
B: Hm (zustimmend)
A: Und tja, Traum des Westens ist
For the final translation "brandneu" is taken up, and the whole phrase then
runs: "brandneue Autobahnraststätten mit gut sortiertem Warenangebot und
blitzsauberen Selbstbedienungsrestaurants". This translation again is rather
"removed" from the phrase in the source text, but it exactly renders its
meaning. Here again we may also say that it has been brought about by
processes of creative thinking. When subject A interprets the meaning of
"shiny" and "gleaming" she displays obvious features of fluency and lateral
thinking. In a short space of time she mentions quite a number of
characteristics of the situation in the new German Bundesländer which are
not connected with shiny and gleaming on a structural semantic level:
"brandneu, wunderbar, schön, Traum des Westens".
I would like to point out another phenomenon. There is an extremely
close relationship between the comprehension and the reverbalisation phase.
"Brandneu", one of the words used in the interpretation phase, as we saw,
actually turns up in the translation. The traditional notion that in the
translation process we can distinguish two separate phases should, I think,
be replaced by a model that leaves room for overlapping of the phases.
Indeed, in my teaching I have observed, that there is not only overlapping
but identity, that is, comprehension and translation is often the same thing.
Then a good translation is nothing else but the verbalisation of one's
comprehension processes (cf. Kussmaul 1995). This seems to be true at
least for the cases where the world knowledge and the cultural expectations
of source-text readers and target-text readers are fairly similar. When they
differ, translators may have to adapt their comprehension of the source text
to the needs and expectations of the target readers. The text I chose was
"easy" in this respect. It was a text about Germany, and German readers
would not have found it difficult to comprehend the cultural background it
referred to. I chose an "easy" text for my first experiment. Future
experiments will have to show what happens when texts do not deal with
topics that can be comprehended by making use of scenic knowledge stored
in one's memory. Moreover, results may differ when one translates into the
foreign language, because then the verbalisations of the comprehension
process (of a text written in the translator's mother tongue) will most likely
take place in the mother tongue. Identity between comprehension and
translation would then impossible.
Partly successful processes
"gigantic pastries "
In one of the protocols there is a good example of repairing a mistake

through scenic imagination. The subjects at first thought that "pastries" were
Pasteten (pies), probably because the words look almost alike. But then they
realised that in a German café pies are not a typical meal:
A: Aber irgendwie in einem deutschen Cafe kommt man doch nicht
zusammen, um mächtige Pasteten zu essen. Ich weiß nicht, ob wir mit
diesen pastries da nicht auf dem falschen Weg sind. Kaffee und Kuchen
ist traditionell deutsch, aber doch nicht Kaffee und Pasteten. Da kann
was nicht stimmen.
The subjects recognized that the scenic detail initially suggested by the
frame "pastry" did not fit in with the general scene of a German café the
subjects had stored in their memory. They then replaced Pasteten by
Kuchenstücke.
It is interesting to see how the imaginative visualisation of the café
scenario triggers off excellent words and phrases in some of the protocols,
but, strangely, they are sometimes not part of the final translation. For
instance, two subjects, when discussing the translation of "gigantic
pastries", mentioned quite a number of vivid scenic details:
A: Ja, aber ich meine, gigantic, ich meine, das hört sich auch nach riesig
an, soll ja auch so gesagt werden, daß die regelrecht dahingehen und sich
echt vollfressen.
B: Iii, ja
A: Nicht dezent so ein Keks so verdrücken, sondern riesige Stücke...
And when they put other details of the café scenario (creamy coffee) into
their proper perspective they came up with quite an original idea:
A: Ja, ich find auch, sollte schon ein bißchen superlativmäßig, und riesige
Kuchenstücke zu genießen. Jetzt ist nur, so, mit diesem guten Kaffee,
dieser creamy coffee, das gefallt mir noch nicht so. Weißt du, das hört
sich nicht richtig an, das hört sich nicht deutsch an: um dort guten Kaffee
und riesige Kuchenstücke zu genießen. "Guter Kaffee" (lange Pause) und
wenn wir einfach sagen: um dort Kaffe zu trinken und riesige
Kuchenstücke zu vertilgen, ja vertilgen (Lachen)
They seemed to realise that being specific about the coffee is not really
important for the scene that is created here, and after this the way was
paved for a brilliant solution, "riesige Kuchenstücke vertilgen" (devour huge
Paul Kussmaul 245
pieces of cake), and the laughter of the subjects can be taken as a sign that
they evaluated their version positively.
In another protocol a similar observation can be made. In their
discussion of gigantic pastries the subjects are imagining a vivid scene:
A: Oder treffen sich einfach zu Kaffee und Kuchen; ich mein...
B: Das finde ich zu schwach, weil das gigantic pastries... ich denke, das ist
so, daß die dann Kuchen haben, wo dann auch Zuckerguß drauf ist und
vielleicht eine Fettglasur, so Schichtkuchen, so schwarzwälderkirsch-
ähnlich, so so
A: Mhm, genau, ja doch, mhm...
(3 Sekunden Pause)
(...)
B: Schweizer Kaffeehaus zum Kaffeetrinken und Kuchenschlemmen. Die
schlemmen da schon wenn sie so...
After having discussed the translation of creamy coffee the subjects come up
with additional details of the cake-eating scene:
B: ... die Kuchen finde ich schon wichtig, nö...die Torten, riesige Torten...
A: Buttercremetorte, Sahnetorte
In both cases we can observe some features of creative thinking. As I
mentioned earlier, fluent and divergent (or lateral) thinking are probably the
most important features of creativity. Fluency manifests itself here in the
fact that the subjects, especially the pair in the second protocol, produce
quite a number of scenic details in a relatively short space of time, and
divergent thinking can be seen in the fact that the subjects do not try to find
a formal equivalent (i.e. an adjective) for "gigantic" but express the notion
inherent in the word by alternative, sometimes additional, linguistic means
such as verbal phrases (regelrecht dahingehen und sich echt vollfressen,
nicht dezent so ein Keks so verdrücken, riesige Kuchenstücke vertilgen,
Kuchenschlemmen) or nouns and adjectives having to do with the type of
cake (Fettglasur, so Schichtkuchen, so schwarzwälderkirschähnlich, Torten,
riesige Torten, Buttercremetorte, Sahnetorte).
In both cases, however, the subjects finally did not make use of the
ideas that had come to their minds ("Kuchenstücke vertilgen",
"Kuchenschlemmen") when they had imagined the scene. The first pair
eventually opted for: "Um dort Kaffee zu trinken und riesige Kuchenstücke
zu genießen", and the second pair used the rather colourless phrase "treffen
sich zu einem guten Stück Torte in einem Schweizer Café".
What might be the reason for this type of translational behaviour?
Obviously, the comprehension process worked very well, possibly because
it was dominated by the recognition of the macrostructural meaning of the
text (the improving economic situation and greater affluence in the former
GDR). What happened then?
The first pair, when asked why they did not keep vertilgen, said they
thought the word sounded too "crude". Apparently, they did not dare to
make use of what they had verbalised in the comprehension phase. This
seems to be a rather typical kind of behaviour. The subjects seem to lack
self-confidence, one of the basic features of a professional translator, and
maybe they also thought that translating and comprehending were two
separate phases, whereas, as we can see in this instance and as we saw a
little earlier, they are often one and the same phase. The second pair,
although having taken into account the macrostructure of the text in their
comprehension phase, nevertheless, when entering into the translation phase
were occasionally fixed on microstructures. For instance, they extensively
discussed if "creamy coffee" should be translated by Cappuccino or by
Kaffee mit Sahne, but they did not mention the decisive criteria in their
decision making process, namely that the ladies are now able to indulge in
some luxury. And when they translated "coffee shop" they looked up the
word in a dictionary and discussed such irrelevant details as the size of the
house before finally deciding for Schweizer Café, which they could have
done right from the start. This is not an isolated case but seems to be rather
typical. What we can observe here is that microstructures become
predominant and are no longer related to macrostructures. In other words,
the scenic visualisations are blotted out, and features which should have
actually been part of the scene cannot be visualised within the general
scenario. Consequently, the subjects tend to slip back into a rather
unprofessional way of behaviour. They try to understand and look up
isolated words, when they could have easily understood these words by
relying on their own comprehension processes (for similar observations cf.
Hönig 1993:86ff.; 1995:59ff.). This may be interpreted as a sign of lacking
self-confidence, which again prevents the subjects from combining the
comprehension with the translation phase and recognising that very often the
verbalisation of their comprehension is actually the best translation.
My hypothesis that imagining a scene will lead to good translations is
to some extent confirmed by the experiments. In addition, the processes
involved in scenic comprehension and subsequent translation have a great
affinity to creative processes, which means that being creative is not really a
special gift but part of the normal mental make-up of any person. But there
is a snag. People often do not realise that they have just been creative. Our
Paul Kussmaul 247
subjects are a case in point. Sometimes they did not notice that they had
actually found a good translation. There may be several reasons for this. It
may very well be that they do not accept the verbalisations of their
comprehension as proper translations. My teaching experience suggests that
this is indeed so, but some more empirical studies would be needed to
confirm this impression. Moreover, the subjects in their decision-making
sometimes seem to be fixated on microstructures when they should actually
have considered the macrostructure of texts and of corresponding scenes.
For teaching purposes I would suggest that for comprehension of texts
we practise imagining scenes and point out that verbalising comprehension
can be translation. This means that we do not draw a dividing line between
text analysis and translation. Of course, the functions of source and target
text have to be clarified before we begin to translate, but at least as far as
meaning is concerned, text-analytical models that involve "steps" are
dangerous. For translators, text analysis is of pedagogical use only if
analysis and translation are closely correlated. If we fail to do this, we will
impede professional and indeed creative behaviour.
References
Bobrow, Robert J. & Brown, John Seely 1975. "Systematic Understanding: Synthesis,
Analysis, and Contingent Knowledge in Specialized Understanding Systems". In: D.
G. Bobrow/ A. Collins (eds.) Representation and Understanding. Studies in Cognitive
Science. New York: Academic Press, 103-130.
de Bono, Edward. 1970. Lateral Thinking. A Textbook of creativity. London: Ward Lock
Educational.
Fillmore, Charles J. 1976. "Frame Semantics and the Nature of Language". In: J. Harnard
et al. (eds.) Origins and Evolution of Language and Speech. Annals of the New York
Academy of Sciences. Vol. 280. New York, 20-32.
Fillmore, Charles J. 1977. "Scenes-and-Frames Semantics". In: A. Zampolli (eds.):
Linguistic Structures Processing. Amsterdam: N. Holland, 55-88.
Guilford, Joy Peter. 1975. "Creativity: A Quarter Century of Progress". In: I.A.
Taylor/J.W. Getzels (eds.) Perspectives in Creativity. Chicago: Aldine, 37-59.
Honig, Hans G. 1993. "Vom Selbst-Bewußtsein des Übersetzers". In: J. Holz-Mänttäri/ C.
Nord (eds.) Traducere Navem. Festschrift für Katharina Reiss zum 70. Geburtstag.
Tampere (studia translatologica ser. A, vol. 3), 77-90.
Honig, Hans G. 1995. Konstruktives Übersetzen. Tübingen: Stauffenburg .
Kußmaul, Paul. 1994. " Semantic Models and Translating". Target 6/1 (1994), 1-13.
Kußmaul, Paul. 1995. Training the Translator. Amsterdam/ Philadelphia: Benjamins.
Neubert, Albrecht. 1988. "Top-down-Prozeduren beim translatorischen Informationstrans-
fer". In: G. Jäger/A. Neubert (eds.) Semantik, Kognition und Äquivalenz. Leipzig:
VEB Verlag Enzyclopädie, 18-30.
Snell-Hornby, Mary. 1988. Translation Studies. An Integrated Approach. Amsterdam:
Benjamins.
Vannerem, Mia & Snell-Hornby, Mary. 1986. "Die Szene hinter dem Text: "scenes-and-
frames semantics" in der Übersetzung". In: M. Snell-Hornby (ed.) Übersetzungswis
senschaft. Eine Neuorientierung. Zur Integrierung von Theorie und Praxis. Tübingen:
Francke, 184-205.
Vermeer, Hans J. & Witte, Heidrun. 1990. Mögen Sie Zistrosen? Scenes & frames &
channels im translatorischen Handeln. Heidelberg: Groos.
Übersetzen als transkultureller Verstehens- und
Produktionsprozeß
Der Titel Übersetzen als transkultureller Verstehens- und Produktionsprozeß

weist Übersetzen zum einen als Verstehens- und zum anderen als Produktions-
prozeß aus. Die Trennung dieser beiden Prozesse ist jedoch rein methodologi-
scher Art, in der Praxis greifen sie stets ineinander und sind untrennbar
verbunden. Zunächst wird also ein Modell des Textverstehens vorgestellt, das
— wie alle neueren Theorien des Textverstehens und -produzierens — dem
kognitiven Paradigma verpflichtet ist. Es konzipiert Verstehen als neuronales
Geschehen und Resultat kognitiver Prozesse (Zur kognitiven Wende vgl.
Rickheit/ Strohner 1993:9). Auf der Basis dieses Modells, das die theoretischen
Grundlagen und die methodischen Verfahren einer handlungsorientierten
Analyse von Verstehensprozessen deutlich macht, wird danach anhand
konkreter Beispiele der Zusammenhang zwischen sprachlichen und kognitiven
Aktivitäten nicht nur in der Verstehensphase, sondern auch in der Formulie-
rungsphase einer Übersetzung aufgezeigt.
Neuronales Geschehen und kognitive Prozesse
Nach kognitions- wie neurowissenschaftlichen Erkenntnissen besteht eine

Parallelität zwischen Gehirnvorgängen und Verstehensprozessen. Die
Gehirntätigkeit besteht aus Neuronenströmen, wobei Milliarden von Neuronen
über Synapsen hinweg elektrische Signale voneinander erhalten, sofern diese
Signale stark genug sind, diese Kontakte herzustellen. Es entstehen Signal-
oder Erregungsspuren, die also in einer Verkettung von Neuronen bestehen und
Engramme genannt werden. Kommt ein Eindruck, eine Wahrnehmung etc. von
außen, die bereits bekannten ähneln, so finden sie einen teilweise gebahnten
Weg vor und benutzen alte Erregungsspuren, bereits bestehende Engramme.
Neuronale Netze erlernen Verbindungen zwischen den Knotenpunkten, den
Synapsen, aktivieren bereits häufig benutzte, stellen über eine Veränderung der
Synapsenstärken, also Stromstärken, neue Verbindungen her. Hauptmerkmal
des Gehirns als neuronales Netz vorgestellt ist also seine Gedächtnis- und
Lernfähigkeit. Lernen geschieht daher nicht über eine Veränderung der
Neuronen sondern über die Veränderung von Synapsenstärken (vgl. Schade
1992:38f.). Neuronale Netze sind also konnektionistisch organisiert, da jegliche
mentale Aktivität aus einer Vielzahl unabhängiger parallel ablaufender
Vorgänge besteht, die sich immer wieder neu kombinieren. Diese Ver-
netzungen stellen Beziehungen her zwischen sprachlichem und konzeptuellem
Wissen. Sie stellen kognitive Strukturen dar, die als Wissenseinheiten nicht
isoliert und ungeordnet, sondern in systematischen Zusammenhängen im
Langzeitgedächtnis abgespeichert werden. Solche systematischen Zusammen-
hänge und typischen Konstellationen, die aus interkonzeptuellen Beziehungen
bestehen und natürlich kulturspezifisch sind, werden scenes genannt bzw.
frames, wenn es sich um die Repräsentation stereotypen sprachgebundenen
Wissens handelt (Zur Kulturspezifik von scenes und frames vgl. Vermeer
1992). Solche Wissensstrukturen repräsentieren komplexe soziale Erfahrungen
und Situationen mit den dazugehörigen Gegenständen, Personen, Handlungen,
Erwartungen, Absichten etc.
Verstehen in der Kognitionswissenschaft
Verstehen beruht somit auf einer vernetzten Integrationsleistung, wobei wir

sprachliche und außersprachliche Wissensbestände in unser Bewußtsein
eingliedern. Es ist das Resultat kognitiver Prozesse, meist von Inferenzen, die
aus mentalen Schlußfolgerungen und Problemlösungsoperationen bestehen (vgl.
Schwarz 1992:29).1 Diese Inferenzen verbinden die Textinhalte mit dem
Wissen über sprachliches Handeln, dem Interaktionswissen (Vgl. Heinemann/
Viehweger 1991:96-108) sowie Erfahrungs- und Weltwissen, um einen
kohärenten und in sich stimmigen Textsinn zu erhalten. Textverstehen wird als
strategisch-konstruktiver Prozeß konzipiert (Näheres zu diesem Prozeß:
Kupsch-Losereit 1995a; 1995b). Im Verstehensprozeß werden also mit
1
Der pure Vergleich von eigenem und fremdkulturellen Sprach- und Weltwissen stellt
keine kognitive Operation an sich dar, so noch Müller (1986:34,37). Diese Vergleichs-
handlung geht der Inferenz voraus, ist also nicht das entscheidende Paradigma des Ver-
stehensprozesses, auch wenn Fremdes erst einmal über bereits Bekanntes wahr- und
aufgenommen wird.
Sigrid Kupsch-Losereit 251
Wörtern, Sätzen und Texten verbundene Konzepte und scenes aktiviert, d.h.
das Referenzpotential einer sprachlichen Äußerung, die mit ihr verbundenen
Situationen, Affekte, sozialen und interaktioneilen Muster etc. Die mentale
Leistung des Translators besteht darin, eine aktuell-spontane Beziehung
zwischen Ausgangstext (AT), Handlungszusammenhängen (d.h. der kom-
munikativen und sozialen Funktion dieses AT) sowie seinem kulturellem
Wissen herzustellen. Aus dem Dargestellten ergibt sich, und neuere Forschung
belegt es, daß sein Verstehen und damit seine kognitiven Strategien abhängig
sind von den Erwartungen (wie z.B. Textsorte, Textintention) und den
Absichten und Zielsetzungen, mit denen der AT gelesen wird.2 Der Translator
liest den Text also von vornherein aus der Perspektive seiner Kultur, aus der
Perspektive des fremdsprachigen und fremdkulturellen Lesers, eben sub specie
translationis. Das Verstehen des Translators bezieht die kommunikative
Funktion von Texten, den sozialen Sinnzusammenhang textvermittelten und
textvermittelnden Handelns mit ein.
Integrativ-produktiver Verstehens- und Produktionsprozeß
Der Translator zieht zugleich die Bewußtseins- und Handlungsdimensionen

seiner Leser in Betracht, antizipiert folglich die Verstehensbedingungen und
-möglichkeiten seiner Leser. Die komplexen Voraussetzungssituationen, in die
Textrezeption und Textproduktion eingebettet sind, können daher nicht aus der
verstehensorientierten Analyse ausgeblendet werden. Ohne Beachtung der
kulturspezifischen Präsuppositionen, des sprachlichen und kulturellen Wissens
sowie der Emotionen des ZT-Lesers und dessen Erwartungen könnte — wie
gezeigt — kein integratives Verstehen erfolgen. Der Translator ermöglicht die
für den ZT-Leser notwendigen Inferenzprozesse, indem er einer veränderten
gesellschaftlichen Wirklichkeit, unterschiedlichen Wirklichkeitsmodellen,
praktischem Regelwissen, kulturellen und sozialen Konventionen und den damit
verbundenen Wissensvorräten Rechnung trägt. Indem er die kulturelle und
sprachliche Differenz sowie nichttextualisierte Verstehensvoraussetzungen
beachtet, kann er falsche Inferenzen hemmen, kommunikationsgefährdende
Texte reparieren oder verhindern. Wurde im Verstehensmodell schon auf die
Wichtigkeit von außersprachlichen Wissensbeständen und den Grad der
Aktivierung der relationalen Vernetzung begrifflichen Wissens hingewiesen,
2
Zwaan (1993:15-18 und 157) belegt die Abhängigkeit kognitiver Strategien und Kontroll-
mechanismen von Texttyp und Texterwartung. Bei Zwaan auch weiterführende Literatur.
so betone ich jetzt im Übersetzungsprozeß die Annahmen bezüglich der
kognitiven Verstehensprozesse beim Leser. Die wichtigsten Annahmen sind
situativer, kommunikativer und interaktionistischer Art.3 Der Translator
formuliert also auf der Grundlage text- wie wissensbezogener Informationen
einen Text, der durch zwei Kraftfelder bestimmt wird. Zum einen durch
translatorische Zielsetzungen, Hypothesen und Relevanzentscheidungen und
zum anderen den durch Normen, Konventionen, Zwänge etc. bestimmten
Diskurs zweier Sprachgemeinschaften.
Das vorgestellte integrativ-konstruktive Modell des Textverstehens und der
Textverständlichkeit erlaubt Aussagen über den Integrationsprozeß: Beim
Übersetzen wird die fremdkulturelle textuelle Information in die eigenen
Wissens- und Handlungsstrukturen sowie die eigene Textwelt integriert auf
Grund des spezifischen historischen Standortes, der Erwartungen, Ein-
stellungen, Intentionen und Interessen des Translators. In einem interaktiven
Prozeß, der ebenfalls auf translatorischen Annahmen und Zielsetzungen (s.o.)
beruht, formuliert der Translator eine Makrostrategie für seine Übersetzung.4
Diese Makrostrategie legt 1. die situativen Handlungsbedingungen (Medium,
Kommunikationspartner, Bezugsrahmen, Vorgaben für Preise etc.) sowie die
Mitteilungs- und Wirkungsabsicht und 2. die Art und Abfolge textsorten-
spezifischer Handlungen (Handlungsstruktur, Organisationsstruktur, sprachkul-
turspezifische Muster etc.) fest. Makrostrategische Überlegungen sind somit
grundlegend für die Übersetzung als zweckgerichteter, situationsbedingter und
auf Verständigung angelegter interlingualer Kommunikation. Alle über-
setzerischen Operationen basieren auf der Einsicht, daß auch Verständlichkeit
das Resultat kognitiver Prozesse ist. Der ZT soll ja eine Integration der vom
AT ermöglichten Handlungszusammenhänge ermöglichen, muß also eventuell
implizit im AT vorhandenes Kulturwissen im ZT explizit machen oder sonstige
Informationen des AT — sei es syntaktisch, semantisch-logisch oder hand-
lungsorientiert — in den ZT inferieren. Die Übersetzung wird nur dann ein
echter transkultureller Kommunikationsvorgang sein, wenn sie in einem neuen
situativ-praktischen Kontext, einem sozial-interaktiven Kontext einer neuen
Diskursgemeinschaft und einem neuen semantisch-kognitiven Kontext der
Erklärung und Interpretation verstanden wird. Eingangs wurde bereits betont,
daß der Verstehens- und Produktionsvorgang nicht getrennt voneinander zu
3
Vgl. dazu ausführlicher Kupsch-Losereit 1996. Allgemein zu Annahmen hinsichtlich
kognitiver Prozesse der Sprachrezeption Rickheit/ Strohner (1993:75f.).
4
Auch Honig (1995:62, 69 und 89) betont und fordert die Steuerung des Übersetzungs-
prozesses durch eine Makrostrategie, ohne jedoch makrostategische Parameter zu
spezifizieren.
betrachten sind, und daß die Verstehenskompetenz ein Teil der übersetzeri-
schen Kompetenz ist. Darüberhinaus wurde festgestellt, daß dem Übersetzungs-
prozeß bestimmte typische Verstehensprozesse vorausgehen, und folglich
Verstehen beim Lesen eines Textes etwas anderes ist als Verstehen um zu
übersetzen (Beispiele dafür in Dancette 1995:207). Während ein Franzose den
Text einwandfrei versteht, wird ein Deutscher den Text nicht verstehen, weil
sprachlich und kognitiv andere Prozesse ablaufen. An Hand des leicht
gekürzten französischen Textes C'est qui le Monsieur sur la croix? (s. Anhang)
sollen einige dieser typischen Verstehensprozesse aufgezeigt werden, die sich
im Unterricht ergaben und dann in eine Produktion des ZT mündeten.
Wir formulierten im Unterricht zunächst die situativen und soziokulturel-
len Handlungsbedingungen (d.s. die Präsuppositionen des ZT-Lesers) sowie die
Mitteilungs- und Wirkungsabsicht. Für welchen Leserkreis, in welcher
Situation, zu welchem Zweck wird dieser Text übersetzt? Für deutsche,
frankreichkundlich interessierte Bildungsbürger von heute, die aus Zeitungs-
berichten bzw. -kommentaren aktuelle Informationen über das Verhältnis von
Schule, Kirche und Staat in Frankreich erfahren wollen.
Sprachlich-konzeptuelle Verstehensschwierigkeiten und -Strategien
Zunächst ergaben sich sprachliche Verstehensschwierigkeiten. In bezug auf

unseren Text sind zu nennen: Abkürzungen, Namen, Titel, Probleme
unterschiedlicher Kollokationen in AS und ZS (besonders bei Adjektiven),
Homophonieprobleme, Probleme der Hyper- bzw. Hyponymie und der
Antonymie, sprachspezifische Konventionen bzw. Gebrauchsnorm, unter-
schiedliche Konzeptualisierung sprachlicher Äußerungen, die Relation der
Konverse.
Abkürzungen:
Z. 5: CE 2 steht für Cours élementaire 2 und entspricht der '3. Grund-
schulklasse' (s.u.).
Titel:
Z. 18: Athalie bleibt erhalten, da alle deutschen Racine-Übersetzungen den
Titel ebenfalls beibehalten, er also bibliographisch identifiziert werden kann;
und Z. 19: Booz endormi bleibt in Klammer hinter der deutschen Version 'Der
schlafende Boas' stehen.5
Adjektiv-Nomen- bzw. Nomen-Adjektiv-Verbindungen:
Diese Verbindungen lösen bei französisch-deutschen Übersetzungen sofort
übersetzerische Verstehensstrategien aus, da die Bedeutung französischer
Adjektive sich aus der jeweiligen syntaktisch-semantischen Stellung/Beziehung
zum Nomen, neuronal gesehen also aus der Vernetzung zu anderen Knoten als
Repräsentation einer Wortbedeutung bzw. eines Konzepts ergibt und diese
interkonzeptuellen Beziehungen in AT und ZT unterschiedlich, weil sprachkul-
turspezifisch (und individuell) sind. Einige Beispiele mögen dies verdeutlichen:
Z. 20/21: institutions confessionnelles wurde mit 'kirchliche bzw. konfessionell
gebundene Institutionen' übersetzt. Eine Reihe von Adjektiv-Nomen-
Verbindungen ist gekoppelt mit der Homophonieproblematik und war im
Unterricht besonders ergiebig. Es handelt sich um das Adjektiv religieux, se.
In den unterschiedlichen Kollokationen muß es jeweils anders verstanden
werden:
Z. 2: histoire religieuse 'biblische Geschichte, die Bibel'
Z. 4, 40: culture religieuse 'religiöse Bildung, religiöse Lebenswei-
se'
Z. 23: enseignement religieux 'Religionsunterricht'
Z. 25/26: phénomène religieux 'Phänomen des Glaubens'
Z. 28: traditions religieuses 'christlich-religiöse Traditionen, Gebräu-
che'
Z. 46/47: cultures et civilisations 'Religionen und Glaubensformen/
religieuses religiöse Lebensformen, -weisen'
Z. 24: Das Adjektiv musulmanes wird durch das ideologieneutrale 'musli-
misch' wiedergegeben.
Unterschiedliche Konzeptualisierung sprachlicher Äußerungen:
Wie bereits beschrieben ist die Repräsentation einer Wortbedeutung das
Ergebnis eines im AT und ZT unterschiedlichen Beziehungsgeflechts. Steht am
Eingang das Wort dinde (Z. 8) bzw. dt. 'Truthahn', so steht am Ausgang für
den Deutschen nicht automatisch die Bedeutung 'Weihnachten'. Etwas
schwieriger wird es in unserem Text mit laïcité (Z. 2), la notion de laïcité (Z.
29) und in einem konkreten Sinn la laïque (Z. 26) als Ellipse für école laïque.
5
Im romantischen Hugo-Gedicht träumt der alte, unverheiratete Boas (Bibel: Rut 2-4; Mt
1,5) das Unmögliche: Rut legt sich zu seinen Füßen, sein Traum erfüllt sich, sie wird
seine Frau.
Mit diesen Termini konnten die Studierenden begrifflich nichts anfangen und
erkannten daher laïcité und la laïque zwar als Kennzeichnung, quasi als
Thema, aber nicht — was hier in diesem Text besonders interessant ist — als
Aufforderung, das entsprechende kulturelle Wissen oder gar Identifikations-
wissen zu aktivieren. Dies gelang erst, nachdem wir — das betrifft das o.a.
Problem der Hyponymie und Antonymie — einen anderen Terminus geklärt
hatten, die école publique (Z. 2). Der école publique steht im Französischen
antonymisch die école privée gegenüber, ja, sie wird quasi implizit mitgedacht.
Der Gegensatz im Französischen ist aber nicht 'öffentlich, staatlich — privat'
wie im Deutschen, sondern 'laizistisch staatlich — konfessionell gebunden!
Um Verstehen zu ermöglichen, lautet die Strategie des Translators daher:
'Staatliche Schule' bzw. 'die von kirchlichen Einflüssen freie staatliche
Schule' für la laïque bzw. école publique vs. 'Konfessionsschule' für le privé
(Z. 27) bzw. école privée. Und laïcité meint den Grundsatz der 'welt-
anschaulichen Neutralität des Staats', 'die strenge Trennung von Staat und
Kirche'. Die Studierenden schlugen daher als Übersetzung für den Untertitel
Laïcité oblige, l'école publique...vor: 'Und so weiß die laizistisch staatliche
Schule — der strengen Trennung von Staat und Kirche verpflichtet —
nicht,...' oder: '...schließlich sind sie [die Lehrer] strikt dem bekenntnis-
neutralen/laizistischen Schu-/Bildungswesen verpflichtet...'. Nachdem die
Radikalität des laizistischen Denkens begrifflich erfaßt war, wurden dann aus
den syndicats d'enseignants (Z. 39) nicht nur 'Lehrergewerkschaften', sondern
'radikal-laizistische Lehrergewerkschaften' und sogar 'antiklerikale/
atheistische Lehrergewerkschaften'. Um also keine Verständnis- und damit
Kommunikationsprobleme entstehen zu lassen — hier durch eigenkulturell
geprägte Wahrnehmungs- und Deutungsmuster in der Übersetzung, z. B.
öffentliche und private Schulen —, und richtige Inferenzen zu ermöglichen,
wurden in der Übersetzung Verstehenshilfen formuliert. Das AT- und ZT-
Leser gemeinsame Vorverständnis bzw. die Herstellung dieses Vorverständnis-
ses war für eine erfolgreiche Referenz unerläßlich.
Sprachspezifische Gebrauchsnorm liegt in folgenden Fällen vor:
Z. 16: 8 Français sur 10 '80% aller Franzosen'
Z. 5/6: avant et après 'Zeiteinteilung vor und nach Christus/
Jésus-Christ Christi Geburt'
Die französischen Deklarativa in Zeitungstexten haben keine semantisch
distinktive Funktion, sie sind als Verba Dicendi rein stilistische Varianten (Zu
fr. Deklarativa und ihrer deutschen Übersetzung vgl. Kupsch-Losereit 1991:93-
96). Die französischen Verben s'étonne, raconte, dit, affirme, ajoute, insiste,
commente, nuance etc. sind folglich der ZT-Norm anzupassen und können fast
ausschließlich mit 'sagen, mitteilen' übersetzt werden, oder auch eventuell
noch mit 'berichten, versichern' oder: 'so...', 'nach den Worten...'.
Die Strategien des semiprofessionellen Translators zielten in den genannten
Fällen darauf ab, den erwähnten Verständnisschwierigkeiten vorzubeugen und
der Textfunktion und -semantik entsprechend zielkulturelle sprachliche Mittel
vorzuschlagen. Meistens jedoch konnten die Studierenden den Text nicht
verstehen, weil kognitiv-kontextuell notwendige Prozesse nicht ablaufen
konnten.
Kontextuelle Verstehensschwierigkeiten und Verstehensstrategien
Kontextuelle Verstehensschwierigkeiten ergeben sich als Folge mangelnder

außersprachlicher oder pragmatischer (interaktioneller) Kenntnisse.
Außersprachliche Kenntnisse:
Sehr schnell stellt sich die Frage, wieviel an unbekanntem Hintergrundwissen
durch Bekanntes verstehbar gemacht werden muß und mit welchen Mitteln.
Wieviel an Erklärung, Kommentar oder Paraphrase ist für den deutschen Leser
notwendig?
Z. 5: en classe de CE2 (die 3. der 5 Grundschulklassen) 'in der 3. Klasse'
Z. 47: Die sixième, Seconde werden entsprechend umgerechnet: 'die 6. und
10. Klasse'.
Institutionen wie Arrondissement (Z. 7 und 35) bleiben als bekannt vor-
ausgesetzt stehen. Dagegen wird die Education nationale (Z. 38) paraphrasiert
als 'Ministerium für Erziehung und Unterricht'. Und der Schultyp collège (Z.
35), den alle Kinder von der 6. bis zur 9. Klasse besuchen, der also in
Deutschland Hauptschule, Realschule und Gymnasium umfaßt, wird wohl am
besten mit 'Sekundarschule' übersetzt. Schwieriger gestaltet sich das
Verständnis des Satzes am Ende des ersten Absatzes (Z. 9/10). Um la galette
des Rois zu verstehen, wurde das notwendige textrelevante Weltwissen quasi
inferiert, etwa 'daß der üblicherweise zum 6. Januar gebackene Kuchen mit
dem katholischen Fest Epiphanias zusammenhängt, an dem die Ankunft der Hl.
Dreikönige in Bethlehem gefeiert wird'.6
6
Hier liegt eine Anspielung auf den am 6. Januar in Frankreich üblichen Brauch vor, tirer
les Rois, bei dem derjenige, der beim Verzehr des Dreikönigkuchens auf die eingebacke-
ne Bohne oder das eingebackene Figürchen stößt, König wird.
Pragmatische und interaktionelle Kenntnisse:

Zu den durch pragmatische und interaktionelle Information des AT ausgelösten
Verstehensschwierigkeiten gehören v. a. der implizierte Rezipientenkreis, die
Bildungs- und Wissensvoraussetzungen, der Informationsstand, die Fach-
sprachlichkeit des Textes, Register (Argot, Schülersprache, Ellipsen) und
Tenor. Natürlich fallen in unserem Text die umgangssprachlichen, elliptischen
Wendungen der Schülersprache auf. Meine Studierenden entschieden sich für
folgende Verstehenshilfen: les profs de lettres (Zeile 17), les instits de la
laïque (Z. 26), les curés du privé (Z. 26) und aller au caté (Z. 23) wieder-
zugeben mit 'Französischpauker', 'Pauker, Lehrer der konfessionslosen
(radikal-laizistischen) Volksschulen', Triester, Pfaffen der Konfessions-
schulen' und 'außerschulischer Reliunterricht' (Besuch normalerweise vom 8.-
12. Lebensjahr).7
Aus allen Beispielen ergibt sich, daß Textverstehen eine strategische
Rekonstruktion eines mentalen Modells ist, das auf bottom-up und top-down
Prozessen beruht. Der Translator versteht den AT auf der Basis seiner
Erwartungen und Ziele, seiner Lebens- und Erfahrungswelt und versucht, den
Sinn des AT in einem anderen Welt- und Lebenszusammenhang fortzuspinnen
und begreifbar zu machen, fremdkulturspezifische Sprech- und Interaktions-
weise einer eigenkulturspezifischen Verstän???gung zuzuführen. Ich legte am
Anfang dar, daß sprachliche wie enzyklopàuische Wissensbestände nicht
ungeordnet sondern in interkonzeptuellen Beziehungen, als scenes gespeichert
werden. Im Verstehensprozeß entwickelt der Translator also eine scene von
dem, was im AT steht. Gibt es die auch im Deutschen? Und wie drücke ich
sie aus?8 Im Unterricht machte ich die Erfahrung, daß ohne eine solche
bewußte Vorstellung, solch eine scene, entweder eine wörtliche Übersetzung
gemacht wurde, oder die eigenkulturelle Vorstellung auch für das Fremdkultu-
relle, quasi als Projektion, übernommen wurde. Besonders lehrreich dafür war
7
Die wahrlich neutral-höfliche, und gerade dadurch provozierende Frage nach einer
männlichen Person in der Titelüberschrift: Cèst qui, le monsieur sur la croix? kommt
in ihrem distanzierten Tenor mit 'Wer ist der Mann oder: der Herr dort am Kreuz?'
nicht so ganz zum Tragen, da 'der Mann' im Tenor nicht ganz stimmt, und 'der Herr'
ja bereits auch 'der Heiland,' fr. le Seigneur, bedeuten kann.
8
Zur Notwendigkeit der scece-Erstellung für das Übersetzen vgl. Kußmaul (1995:67,
94ff.). Eine schwierigere Frage schließt sich an: Selbst wenn im Deutschen dieselbe
scene verbalisiert werden kann, so ist die Bewertung dieser scene häufig eine andere. So
ist z. B. das Ansehen und die Macht der französischen Gewerkschaften für Erziehung und
Wissenschaft wesentlich größer und stärker als etwa die von GEW oder ÖTV in
Deutschland.
das Beispiel mit der galette des Rois, das — wörtlich und unverständlich im
Deutschen — zum 'Blätterteigkuchen der Hl. Dreikönige' wurde bzw. — als
Projektion — zum 'Sternsingen'. Wie wichtig das Erstellen einer scene für das
Übersetzen ist, soll folgendes Beispiel zeigen. Der erste Satz des ersten
Kapitels (Primo giorno. Prima) in Umberto Ecos // nome della rosa lautet: Era
una bella mattina di fine novembre. Nella notte aveva nevicato unpoco, ma il
terreno era coperto di un vélo fresco non più alto di tre dita. Die deutsche
Übersetzung lautet: 'Es war ein klarer spätherbstlicher Morgen gegen Ende
November. In der Nacht hatte es ein wenig geschneit, und so bedeckte ein
frischer weißer Schleier, kaum mehr als zwei Finger hoch, den Boden.' Ich
finde diese Übersetzung genial, gibt es doch bei uns keine schönen November-
morgen und schon gar nicht gegen Ende des Monats. Die scene, die wir haben,
sieht eher nach trüben kalten Regen- und Nebelmorgen aus. Um also den
Eindruck aus der Textvorlage zu übermitteln, übersetzte Kroeber 'klarer
spätherbstlicher Morgen', ergänzt den velo fresco zu 'frischer weißer Schleier'
und aus den non più alto di tre dita, was logisch gesehen weniger als drei
Fingerbreiten sind, wurden in der deutschen scene 'kaum mehr als zwei Finger
hoch'. Wie gelungen diese Übersetzung ist, sieht man besonders auch im
Vergleich zur französischen Fassung: C'était une belle matinée de la fin
novembre. Dans la nuit, il avait neigé un peu, mais le terrain était recouvert
d'un voile frais pas plus haut que trois doigts.
In allen genannten Fällen wird der AT unter translationsspezifischen
Annahmen mit Hilfe sprachlicher und kognitiver Problemlösungsstrategien des
Translators in Verstehenskategorien integriert. Anschließend imaginiert der
Translator einen ZT, der ein Verständnis ermöglicht. Textverständnis bezieht
sich dabei auf die "Textwelten", d.i. das kulturspezifische Handeln und Wissen
über die betreffenden Wirklichkeiten und Welten, die durch die Texte explizit
oder implizit aktualisiert werden. Alle kognitiven und übersetzerischen
Strategien des Translators zielen also darauf ab, ein zielkulturell kohärentes
Textverständnis zu ermöglichen.
Literatur
Dancette, Jeanne. 1995. Parcours de traduction: Etudes expérimentales du processus de

compréhension. Lille: Presses Universitaires de Lille.
Eco, Umberto. 1980. Il nome della rosa. Milano: Bompiani (Deutsche Übersetzung von
Burkhart Kroeber. 1982. Der Name der Rose. München: Hanser. Französische
Übersetzung von Jean-Noël Schifano 1982. Le nom de la rose. Paris: Grasset.
Heinemann, Wolfgang und Viehweger, Dieter. 1991. Textlinguistik. Eine Einführung.
Tübingen: Niemeyer.
Hönig, Hans. 1995. Konstruktives Übersetzen. Tübingen: Stauffenburg.
Kupsch-Losereit, Sigrid. 1991. "Die Relevanz von kommunikationstheoretischen Modellen für
Übersetzungstheorie und übersetzerische Praxis." TextConText 6 (2/3), 77-100.
Kupsch-Losereit, Sigrid. 1995a. "Die Modellierung von Verstehensprozessen und die
Konsequenzen für den Übersetzungsunterricht." TextConText 10 (3), 179-196.
Kupsch-Losereit, Sigrid. 1995b. "Übersetzen als transkultureller Verstehens- und Kom-
munikationsvorgang: andere Kulturen, andere Äußerungen. " In: N. Salnikow (ed.), 1-15.
Kupsch-Losereit, Sigrid. 1996. "Kognitive Verstehensprozesse beim Übersetzen." In: A. Lauer
et al. (ed.), 217-228.
Kußmaul, Paul. 1995. Training the Translator. Amsterdam: Benjamins.
Lauer, Angelika et al. (eds.) 1996. Übersetzungswissenschaft im Umbruch. Festschrift für
Wolfram Wilss zum 70. Geburtstag. Tübingen: Narr.
Müller, Bernd-Dietrich. 1986. "Interkulturelle Verstehensstrategien — ein Vergleich und
Empathie." In: G. Neuner (ed.), 33-84.
Neuner, Gerhard (ed.) 1986. Kulturkontraste im DaF-Unterricht. München: Hueber (Studium
Deutsch als Fremdsprache-Sprachdidaktik 5).
Rickheit, Gert/Strohner, Hans. 1993. Grundlagen der kognitiven Sprachverarbeitung. Modelle,
Methoden, Ergebnisse. Tübingen/Basel: Francke (UTB 1735).
Salevsky, Heidemarie (ed.) 1992. Wissenschaftliche Grundlagen der Sprachmittlung. Berliner
Beiträge zur Übersetzungswissenschafi. Frankfurt: Lang.
Salnikow, Nicolai (ed.) 1995. Sprachtransfer — Kulturtransfer. Text, Kontext und Translation.
Frankfurt: Lang.
Schade, Ulrich. 1992. Konnektionismus: Zur Modellierung der Sprachproduktion. Opladen:
Westdeutscher Verlag.
Schwarz, Monika. 1992. Einführung in die kognitive Linguistik. Tübingen: Francke (UTB
1636).
Snell-Hornby, Mary, Pöchhacker, Franz und Kaindl, Klaus (eds.) 1994. Translation Studies.
An Interdiscipline. Amsterdam/Philadelphia: Benjamins.
Vermeer, Hans J. 1992. "Eine kurze Skizze der scenes-&-frames-Semantik für Translatoren."
In: H. Salevsky (ed.), 75-83.
Zwaan, Rolf A. 1993. Aspects of literary comprehension. Amsterdam: Benjamins.
Anhang
C'est qui, le monsieur sur la croix?
Les enfants connaissent très mal l'histoire religieuse. Laïcité oblige, l'école publique ne sait
pas trop comment la leur enseigner.
Le constat des enseignants est unanime: en matière de culture religieuse, les enfant sont
ignares. «En classe de CE 2, lorsque nous étudions la division du temps entre avant et après
Jésus-Christ, sur 25 gamins, deux ou trois ont entendu parler de Jésus», s'étonne Jacques
Trief, directeur d'une école primaire dans le XIIIe arrondissement de Paris. Noël? Une
réjouissance païenne, avec ses sapins, ses guirlandes, sa dinde et ses cadeaux. Qui sait encore
que la galette des Rois a un 10 rapport avec l'Epiphanie, fête catholique célébrant l'arrivée
des Rois mages à Bethléem ?
Plus tard dans le cursus scolaire, les Lycéens connaissent mieux la mythologie grecque que les
personnages de la Bible ou du christianisme. «Lorsque des élèves ayant choisi l'option arts
plastiques ont dû commenter un tableau représentant saint Sébastien percé de flèches, certains
ont affirmé qu 'il s'agissait d'un Indien victime de la conquête de l'Ouest par les Américains»,
raconte le proviseur d'un grand lycée parisien, à peine remis de sa surprise. Les symboles du
catholicisme - religion dont se réclament encore 8 Français sur 10 - sont, peu à peu, devenus
impénétrables. Les profs de lettres éprouvent de plus en plus de difficultés pour expliquer
«Athalie», de Racine, ou certains poèmes de Victor Hugo pétris de références bibliques, comme
«Booz endormi».
Car - c'est une évidence - la société française se sécularise, tandis que les institutions
confessionnelles perdent de leur influence. Ainsi, parmi ceux qui se disent catholiques, les
pratiquants réguliers sont, désormais, à peine 10 %. Et les enfants sont de moins en moins
nombreux à aller au «caté». Le recul de la pratique et de l'enseignement religieux s'observe
de la même manière au sein des familles juives et musulmanes.
Quant à l'école publique, elle a du mal - laïcité oblige - à aborder sereinement le phénomène
religieux. Pendant des années, en effet, les «instits de la laïque» ontferraillé contre les «curés
du privé», en s'efforçant d'apporter les lumières de la modernité à des enfants qu 'ils jugeaient
trop marqués par leurs traditions religieuses et régionales.
Sur cette guerre des idées franco-française s'est édifiée la notion de laïcité, que l'on ne connaît
ni en Belgique ni en Allemagne, où des enseignements religieux sont dispensés, de manière
facultative, dans les écoles publiques.
APPRENDRE LA TOLÉRANCE
Rien d'étonnant donc à ce que les enseignants français demeurent encore parfois traumatisés
par ce conflit idéologique. Pour certains d'entre eux, laïcité égale silence. [...] Dans un
collège du IXe arrondissement de Paris, Marie-France Galland, enseignante en histoire-géo
graphie, encourage, si la situation se présente, des discussions entre élèves sur leurs croyances
respectives. [...]
Ces dernières années, les esprits ont grandement évolué au sein de l'Education nationale et
même parmi les syndicats d'enseignants. Au diable l'athéisme militant et l'eradication de la
culture religieuse! Le problème de l'enseignement de l'histoire des religions a été posé, pour
la première fois officiellement, lors d'un congrès à Besançon en 1991. Aujourd'hui, la
puissante Ligue française de l'enseignement et de l'éducation permanente - qui regroupe 34
000 associations laïques - est favorable à ce principe. «Pourquoi ne pas porter sur la Bible et
le Coran un regard humaniste?» interroge Anne-Marie Franchi, membre de la Ligue. [...]
En tout cas, sous l'impulsion de François Bayrou et sous la houlette du philosophe Luc Ferry,
président du Conseil national des programmes, l'accent sera mis désormais sur les cultures et
les civilisations religieuses, en particulier en sixième et en seconde. Ainsi, les nouveaux
programmes d'histoire de seconde, qui seront appliqués en 1996, comprennent la naissance
et la diffusion du christianisme, l'Islam et l'orthodoxie, l'humanisme et la Renaissance, la
Révolution francaise. [...]
Marie-Laure de Léotard (L'Express, 22 décembre 1994)
Von Scheuklappen, Mikroskopen und Fernrohren:
Der Umgang mit Wissen in der Entwicklung der
Übersetzungskompetenz
Hanna Risku
Wenn wir das Übersetzen-Lernen als Wissenserwerb und den Übersetzungs-

unterricht als Wissensvermittlung beschreiben, liegt der Schluß nahe, die
Kompetenz von Übersetzenden hänge bloß von der Quantität ihres Wissens ab.
Anhand von kognitionswissenschaftlichen Überlegungen und praktischen
Beispielen soll jedoch dargelegt werden, daß bei Übersetzenden für den
Unterschied zwischen Novizen und Experten längst nicht nur die Wissens-
menge verantwortlich zeichnet. Expertinnen und Experten gehen mit vorhande-
nem Wissen vielmehr ganz anders um und sind so in der Lage, jeweils für die
konkrete Situation adäquate Strategien zu erschließen. Für die Darstellung
dieser Kompetenzerweiterung reicht es nicht aus, nur Begriffe zu präsentieren,
durch die die Entwicklungsprinzipien selbst beschrieben werden können, es
müssen zunächst auch die Begriffe Kompetenz und Wissen definiert werden.
Übersetzungskompetenz: Ausgangs- und Zielpunkte
Übersetzungskompetenz — was ist damit gemeint? Wie sehen die Ausgangs-

und Zielpunkte der Entwicklung der Übersetzungskompetenz aus? Nehmen wir
als Ausgangspunkt die Anfänger etwa in Übersetzungsbüros, an Koordinations-
stellen internationaler Zusammenarbeit und auch an translationswissen-
schaftlichen Universitätsinstituten. Sie besitzen oft hohe kommunikative
Kompetenz innerhalb ihrer Arbeitskulturen (z.B. Bikulturelle, Sprachfachleute
und Personen mit langer Auslandserfahrung). Zudem haben sie manchmal
sogar ausgezeichnete Kenntnisse in dem Fachbereich, innerhalb dessen
interkulturell kooperiert werden soll (z.B. Wissenschaftler, Juristen oder
Ingenieure). Dennoch liefern diese Novizen anfangs inkohärente Über-
Setzungen, sie erkennen nicht die Möglichkeit und die Notwendigkeit, sich mit
anderen Übersetzenden, Fachexperten oder mit dem Auftraggeber in geeigneter
Weise abzusprechen oder scheitern bei Kooperationsprojekten an einer
einheitlichen, für alle Partner einsichtigen und verträglichen Linie, die das
Zusammenprallen der auch ihnen bekannten unterschiedlichen kulturellen
Erwartungen verhindern würde.
Der anvisierte Zielpunkt, die übersetzerische Kompetenz, ist erst dann
erreicht, wenn funktionsadäquate Texte geliefert werden können (wenn also die
Übersetzung in der Zielkommunikationssituation verwendet werden kann),
wenn komplexe Besprechungssituationen und Kooperationsprogramme
beherrscht und initiiert werden und wenn die Handelnden in der Lage sind, zu
erkennen, wann eine Aufgabe besser anderen übermittelt werden sollte.
Denken wir aber zurück: Warum konnten unsere Anfänger dies nicht gleich?
Wenn sie schon vor der Ausbildung die Arbeitskulturen inklusive -sprachen
beherrschten und Fach und Betrieb gut kannten, was kann ihnen dann noch an
Kompetenz gefehlt haben?
Diese Kompetenz nenne ich nach Holz-Mänttäri (1984) die translatorische.
Um eine Begriffsbeschreibung zu versuchen: Zur Übersetzungskompetenz
gehört einerseits die allgemeine translatorische Kompetenz, d.h. die Problemlö-
se- oder auch Handlungskompetenz, die es einem erlaubt, komplexe inter-
kulturelle Kommunikationssituationen auf Bestellung zu meistern. Zusätzlich
verlangt gerade der translatorische Bereich des Übersetzens die Kompetenz,
(für die Übersetzenden) potentiell korrigierbare Texte zu gestalten, wie bereits
Kade (1968:35) Übersetzen vom Dolmetschen unterschieden hat. Bereits durch
diese pragmatische Beschreibung wird klar, daß translatorische Kompetenzen
1. nicht angeboren sein können (auch nicht in einem Chomskyschen,
generativen Sinne),
2. nicht einfach durch das Annehmen von Berufsbezeichnungen und andere
soziale Mechanismen erworben werden und
3. auch nicht mit einfachem "Recht haben" gleichgesetzt werden können (in
einem sehr vereinfachten Sinne "zu wissen"), ist doch die Kooperation und
das Erkennen eigener Grenzen gerade ein integraler Teil dieser Kom-
petenz.
Diese pragmatische Beschreibung erklärt aber erst die Funktion, nicht jedoch
die Struktur dieses Begriffes. Wie hat sich die Kompetenz entwickelt?
Hanna Risku 263
Translatorisches Wissen
Der Aspekt der Entwicklung übersetzerischer Kompetenz, der hier behandelt

werden soll, ist der des translatorischen Wissens. Wie bei den Begriffen
"Übersetzen" und "Kompetenz" wäre es hier völlig unzureichend, eine
gemeinsame alltagssprachliche Vorstellung zu übernehmen oder von einer
solchen auszugehen. Denn: das immer wieder bemühte Bild von Wissen als im
Gedächtnis gespeicherte, statische Symbole und Regeln, die aus der Umwelt
aufgenommen worden sind, hat sich als überaus problematisch erwiesen. Über
das Warum und Wieso des daraus entstandenen Umschwungs in den Ko-
gnitionswissenschaften gibt es bereits bibliothekenweise Literatur; wir müssen
uns hier darauf beschränken, einige grundlegende Erkenntnisse kurz zu-
sammenzufassen, über die umfassender Konsens besteht und die besondere
Plausibilität für die Übersetzungspraxis erkennen lassen. Notwendig ist also
zuerst die Erklärung einiger kognitionswissenschaftlicher Begriffe, ohne die
jede Aussage über kognitive Vorgänge beim Übersetzen-Lernen jeglicher
Grundlage entbehren würde.
Unsere Umwelt genauso wie das Anwenden von Symbolen (etwa
sprachlichen Symbolen) und Befolgen von Regeln ist zwar für das menschliche
Denken von größter Bedeutung, die Wahrnehmung der Umwelt gestaltet sich
aber nicht als eine Aufnahme von objektiv gegebenen Merkmalen, sondern sie
erhält erst ihre Eigenschaften in einem mehrstufigen Interpretations- und
Konstruktionsprozeß, der von den Voraussetzungen des Wahrnehmenden
abhängt. Auch die sinnvolle Verwendung von Symbolen (das ist ja ein
wichtiges Ziel des Übersetzen-Lernens) kann erst gewährleistet werden, wenn
Symbole mit ganzen Netzen von Erinnerungen, Situationen und ihren Zielen,
Erfolgen und Mißerfolgen verbunden werden können. Symbole sind also immer
in die assoziativen Vorgänge eingebettet und in ihnen verankert — Symbol-
manipulation ist nur ein sicht- und hörbarer Überbau eines für das bloße Auge
verborgenen, sinngebenden Prozesses. Erst diese selbstorganisatorische Basis
"unterhalb" der Symbole (deshalb "subsymbolisch" genannt) macht ver-
ständlich, warum Intuition und Emotion Merkmale aller kognitiven Vorgänge
darstellen und wie etwa die Sprache Sinn und Bedeutung erfährt, d.h. wie sie
in menschlichen Handlungen eine Rolle spielt. Dieser als Konnektionismus
(PDP) bezeichnete Ansatz in der Modellierung der Gehirnprozesse (z.B.
Rumelhart. & McClelland 1986; Peschl 1990; Dorffner 1991) bietet eine
ernstzunehmende Alternative zu rein symbolmanipulativen Modellen, obgleich
seine Entwicklung erst im Gange ist. In der Übersetzungswissenschaft kann es
insbesondere helfen, den Übersetzungsprozeß realistisch zu modellieren und
seine Entwicklung darzustellen (vgl. Risku 1994; 1995). Ist es aber nicht eine
Themenverfehlung, wenn zuerst von der Entwicklung des zum Übersetzen
notwendigen Wissens und jetzt von der Entwicklung des Übersetzungsprozesses
gesprochen wird? Ist nicht Wissen etwas Statisches, während Prozesse je nach
Ziel und Situation unterschiedlich ausfallen können? Ganz im Gegenteil.
Gerade weil das Wissen in realen Handlungssituationen für die Lösung
begreifbarer Probleme von den Handelnden selbst konstruiert werden muß, hat
es immer eine erfahrungsbedingte Basis — es wird als Strategie für die
Bewältigung bestimmter Situationen vom Problemlöser entwickelt. Eine
mögliche Definition des Wissens wäre "erfahrungsabhängige, relativ stabile
Neuronenaktivierungsmuster, die je nach aktuellem Gesamtzustand (d.h. je
nach Problemdefinition, je nach Situationsauffassung) aktiviert werden".
Wissen als eine potentielle Handlungsfähigkeit wird also für die Bildung der
momentanen Interpretation, der Repräsentation, in stets modifizierter Form
eingesetzt.
Wir haben damit ziemlich eindeutig etwa das Einpauken von Vokabel-
friedhöfen und statischen Übersetzungsregeln als ein unzureichendes Entwick-
lungsprinzip enttarnt. Das folgende Beispiel mag dies weiter veranschaulichen:
Ein österreichischer Verein veranstaltet eine Enquete (österreichisch für
"Arbeitstagung") über internationale Erfahrungen mit Punkteführerschein und
Negativdatei (ähnlich der bundesdeutschen "Flensburger Kartei"). Die
nachträglich zu veröffentlichende Publikation soll für den österreichischen
Gebrauch in deutscher Sprache erscheinen, wozu auch eine schriftliche Vorlage
des britischen Beitrags übersetzt werden soll. Der Ausgangstext zeichnet sich
durch klare Form und Argumentation aus, den Konventionen eines wissen-
schaftlichen Artikels entsprechend mit Quellenangaben und Anhang. Im
gesamten bietet er gutes Ausgangsmaterial für die Übersetzung, die ebenso
zum größten Teil durch Klarheit und gute Argumentation besticht. Sehen wir
uns aber die Literaturliste des Zieltextes an:
1. Verkehrsministerium/Innenministerium: Berichtder interministeriellen Arbeits-
gruppe über das Straßenverkehrsrecht (London, Kanzlei Ihrer Majestät, 1981)
2. Verkehrsministerium/Innenministerium: Prüfbericht über das Straßen-
verkehrsrecht (London, Kanzlei Ihrer Majestät, 1988)
3. Informationen aus den jährlichen Veröffentlichungen des Innenministeriums
"Mit Kraftfahrzeugen in England und Wales begangene Verkehrsdelikte" (der
letzte vorliegende Bericht ist jener für 1993). Da die Zahlen jährlichen
Schwankungen unterliegen, wurden sie aufgerundet.
4. Parlament des Bundesstaates Victoria: Ausschuß für Verkehrssicherheit: Unter-
suchung des Strafpunktesystems (November 1994)
Hanna Risku 265
...dabei gibt es alle angeführten Berichte nur auf englisch. Interessierte

Leserinnen und Leser werden in Bibliotheken unter diesen Angaben vergeblich
suchen. Bei der sonst hohen Qualität des deutschen Textes ist es nebenbei auch
merkwürdig, daß dort von "Unfällen mit Körperverletzung" die Rede ist —
gemeint ist wohl der übliche Terminus "Unfälle mit Personenschaden". Dieser
terminologische Lapsus wird jedoch die Kommunikation nicht weiter
verhindern. Schwerwiegender ist, daß im Zieltext auf "Anhang A" hingewiesen
wird, der sich nirgendwo findet. Offensichtlich ist damit das abschließende
Kapitel 6 gemeint — ohne Hinweis darauf, daß dieses als Anhang zu verstehen
ist! Daß dies auch beim Ausgangstext der Fall ist, vermindert nicht die
Verwirrung der Lesenden.
Wenn wir nun daraus schliessen, daß kompetentere Übersetzende eben die
richtigen Symbole (hier: "Unfälle mit Personenschaden") und Regeln (hier:
"Literaturangaben sollen nicht übersetzt werden", "Fehler der Autoren dürfen
korrigiert werden") gekannt hätten, machen wir es uns zu einfach. Denn beim
nächsten Übersetzungsauftrag gelten diese "Regeln" womöglich nicht mehr.
Außerdem wird ein Teil der Literaturliste sehr wohl übersetzt: Der Text "der
letzte vorliegende Bericht..." bei Quelle Nr. 3 gehört nicht zu bibliographi-
schen Daten, sondern stellt einen Hinweis des Autors dar. Und auswendig zu
wissen, wie die vielen verschiedenen Unfalltypen benannt werden, kann oft
nicht einmal von Verkehrsfachleuten verlangt werden.
Konsequenzen für die Didaktik
In der Entwicklung der Übersetzungskompetenz scheint es nicht um die aus-

schließliche Kumulation von Daten und Fakten zu gehen, sondern um den
Umgang mit dem vorhandenen Wissen. Was heißt das für Praxis und Didaktik?
Wenn etwa das Vorwissen und die eigenen Erfahrungen so wichtig sind, klänge
es sinnvoll, vorhandene Gewohnheiten durch ständige Übung zu festigen,
auszufeilen und zur perfekten Automatik zu trainieren. Das Motto "Übersetzen
lernt man durch Übersetzen" schiene also bestätigt. Bei genauerer Betrachtung
ist aber hier nichts anderes als die normative Kraft des Faktischen bestätigt,
weshalb dieser Ansatz die ideale Scheuklappenmethode bezeichnet: Die
empirisch belegte, für den Laien typische Konzentration auf augenscheinliche
Textelemente und unreflektierte Wiederholung gelernter Verhaltenselemente
(vgl. Jääskeläinen & Tirkkonen-Condit 1991:93; Séguinot 1991) wird
begünstigt, wie auch die strengen Regeln, mit denen die komplexe Aufgabe in
handhabbare Stücke zerteilt wird (etwa "jeden Satz vollständig übersetzen",
"Fachtermini mit Fachtermini ersetzen", (vgl. dazu Hönig 1993:83; 1995:61).

In Ermangelung solcher Regeln werden diese bereitwillig oder auch unbewußt
im Lernumfeld gesucht. Wenn das nicht-translatorische Vorwissen dominiert
(wie bei Sachfachleuten), wird die Ermöglichung interkultureller Kom-
munikation eventuell gar nicht als eigenes Problem erkannt, sondern nach
gänzlich übersetzungsfremden Prinzipien gehandelt und z.B. die Besprechung
von Fachaussagen in den Mittelpunkt gestellt (s. dazu Tirkkonen-Condit
1992:438).
Das Stichwort "Nutzung vorhandener Strategien" bedeutet nicht Automati-
sierung und Reglementierung, sondern Verwendung von Vorwissen als Basis
für die Konstruktion immer handlungsadäquateren Wissens. Typisch für kom-
petentes Handeln ist nämlich nicht die Fähigkeit, das eigene Handeln in
Regelform erklären zu können, sondern die ständige Überprüfung und
Infragestellung des momentanen Handelns {Reflexion), wobei die Fähigkeit zur
Verbalisierung nicht ausschlaggebend ist (Berry & Broadbent 1987). Nur durch
ein genügendes Maß an Chaos kann eine neue, flexible Ordnung entstehen. Um
aus der Erfahrung also auch eine Kompetenzerhöhung und -erweiterung zu
erlangen, empfiehlt sich die "vergleichende Mikroskopenmethode": Übungen
in Form von Fallstudien, wobei die komplexen Anforderungen des Übersetzens
in strategische Etappen bzw. Teilziele aufgeteilt werden (Konkretisierung) und
wo gleichzeitig Ursachen und Folgen verschiedener Handlungsvarianten
genauer betrachtet werden (Flexibilisierung). Dabei relativieren sich Patent-
rezepte, während die Tiefe und Komplexität des Problems sowie die Möglich-
keit mehrerer gleich adäquater Lösungen erkannt wird. So unglaublich dies
auch klingt, die Bildung von abstrakten Begriffen (Abstraktion) in kon-
nektionistischen Modellen ist ein "Nebenprodukt" dieser Reflexion bei der
Lösung konkreter Probleme (s. McClelland et al. 1986:32). Allgemeine
Begriffe werden also genauso wenig wie andere einfach aus der Umwelt
aufgenommen — auch sie müssen mit Hilfe des Vorwissens von den
Handelnden selbst aufgebaut werden und in sozialen und problemlöserischen
Versuchen auf ihre Kohärenz überprüft werden.
Ob es sich nun um Wissen über Kommunikation, Übersetzen oder dessen
Ziele und Maßnahmen handelt, es übernimmt immer weniger die Funktion von
punktuellen Mechanismen und immer mehr die von flexiblen Konstruktions-
werkzeugen.
Hanna Risku 267
Rolle der Verbalisierung
Die vorhin beschriebene gleichzeitige Reflexion, Flexibilisierung, Kon

kretisierung und Abstraktion ist eine Folge der im Laufe der Kompetenz-
entwicklung zunehmenden Rekursivität kognitiver Prozesse. Die mehrmalige
Erarbeitung eines exemplarischen, authentischen Übersetzungsauftrags für
verschiedene Zielsituationen scheint diesem Umstand didaktisch in vieler
Hinsicht gerecht zu werden. Eines haben wir jedoch bis jetzt "nicht einmal
ignoriert": die Verbalisierung oder sonstige Symbolisierung des Wissens. Dies
allerdings aus gutem Grunde: Die eigenständige Bildung von Begriffsnetzen als
Handlungsstrategien ist insofern primär, als sie die Grundlage für jegliche
Sprachproduktion und -rezeption bildet. Die sprachliche, zeichnerische oder
sonstige Externalisierung von Wissen hat aber sehr wohl große Bedeutung in
der Kompetenzentwicklung: Nach empirischen Schreibstudien etwa geht die
Kompetenzerhöhung mit der Zunahme an Externalisierung einher; versierte
Schreiber planen etwa durch Zeichnungen und Listen (Molitor-Lübbert
1989:292). Dies läßt sich dadurch erklären, daß die Sprache genauso wie
theoretische Modelle ein mächtiges Organisations- und Erweiterungswerkzeug
bietet (s. Oeser & Seitelberger 21995:188, Peschl 1994). Den Versuch, auch
etwas darüber sagen zu können, was man weiß, möchte ich also die Fern-
rohrmethode nennen: Durch Kommunikation mit sich selbst und mit anderen
kann die immer begrenzte Eigenerfahrung in Relation zu kollektiverem,
tradiertem und abstrahiertem Wissen gestellt werden. Das Vom-Übersetzen-
Reden ist also nicht wichtig, um etwa den Studierenden eine wohlformulierte
Theorie aufpfropfen zu können, sondern um die selbstorganisatorisch
"gewachsenen" Begriffe weiter zu einem kohärenten Begriffssystem zu
vernetzen und ihren Geltungsbereich bemessen zu können. Außerdem: solange
wir das translatorische Wissen nicht kommunizierbarer machen und auch
darüber kommunizieren, dürfen wir nicht erwarten, daß Auftrag- und
Arbeitgeber, Leser und andere Beteiligte unsere speziellen Anliegen verstehen.
Um die bisher behandelten didaktischen Konsequenzen zusammenzufassen:
Die Fixierung laienhafter und/oder tätigkeitsfremder Handlungsweisen durch
mechanische Wiederholung ("Scheuklappen") ist möglichst zu vermeiden.
Variable Fallstudien ("Mikroskope") sollten dagegen eingesetzt werden —
nicht nur zur Veranschaulichung der Komplexität und Konstruktivität des Über-
setzens, sondern auch zur selbstorganisatorischen Bildung allgemeiner
Begriffsnetze. Dies sollte durch Externalisierung und Kommunikation
("Fernrohre") ergänzt werden, um das Wissen zu stabilisieren, seine Kohärenz
zu verbessern und seinen Geltungsbereich zu erfahren.
Konsequenzen für die Forschung
Ein tieferer Einblick in kognitive Entwicklungsprinzipien, wie sie hier

angeschnitten wurden, könnte uns effektives Lehren und Lernen erleichtern.
Ein Grund dafür, daß die Ergebnisse von Untersuchungen über Laien- und
Expertenverhalten nur in einzelnen Studien (z.B. Jääskeläinen & Tirkkonen-
Condit 1991) in Beziehung gesetzt worden sind, könnten die methodischen
Unterschiede sein: Nicht ausschließlich, aber vor allem das Laienübersetzen ist
mit der zu recht umstrittenen empirischen Methode des Lauten Denkens
untersucht worden (z.B. Krings 1986, Lörscher 1991, Smith 1994), die ja
geeignet ist, sprachliche, also symbolische Prozesse in den Vordergrund zu
rücken. Jegliche sinngebende und nichtsymbolische kognitive Prozesse bleiben
ihr jedoch verborgen, wodurch die Brauchbarkeit dieser Methode für den
konkreten Untersuchungsgegenstand ernsthaft in Frage zu stellen ist.
Kompetentes Übersetzen dagegen ist in Gesamtbetrachtungen modelliert
worden, die etwa soziale Faktoren (z.B. Holz-Mänttäri 1984) betonen. Eine
kritische Kombination dieser Methoden schiene einen Versuch wert — dabei
kann mit der Methode des Lauten Denkens sehr wohl erfaßt werden, wie wir
über das Übersetzen sprechen, jedoch nicht, wie wir übersetzen. Für welche
Methode wir uns auch individuell entscheiden wollen, es wäre meines
Erachtens von größter Bedeutung, daß dies auf der Grundlage einer kohärenten
kognitiven Theorie geschieht.
Bibliographie
Berry, D. C. & Broadbent, D. E. 1987. "The combination of explicit and implicit learning
processes in task control". Psychological Research 49, 7-15.
Dorffner, Georg. 1991. Konnektionismus; von neuronalen Netzwerken zu einer "natürlichen"
Kl. Stuttgart: Teubner.
Holz-Mänttäri, Justa. 1984. Translatorisches Handeln. Theorie und Methode. Helsinki: Suo-
malainen Tiedeakatemia (Annales Academiae Scientiarum Fennicae B 226).
Hönig, Hans G. 1995. Konstruktives Übersetzen. Tübingen: Stauffenburg (Studien zur
Translation 1).
Hönig, Hans G. 1993. "Vom Selbst-Bewußtsein des Übersetzers". In: J. Holz-Mänttäri/C.
Nord (eds.) Traducere Navem. Festschrift für Katharina Reiß zum 70. Geburtstag.
Tampere: Tampereen yliopisto, 77-90.
Jääskeläinen, Riitta & Tirkkonen-Condit, Sonja. 1991. "Automatised Processes in Professional
vs. Non-Professional Translation: A Think-Aloud Protocol Study". In: S. Tirkkonen-
Condit (ed.) Empirical Research in Translation and Intercultural Studies. Selected Papers
of the TRANSIF Seminar, Savonlinna 1988. Tübingen: Narr, 89-109.
Kade, Otto. 1968. Zufall und Gesetzmässigkeit in der Übersetzung. Leipzig (Beihefte zur
Zeitschrift Fremdsprachen I).
Hanna Risku 269
Krings, Hans P. 1986. Was in den Köpfen von Übersetzern vorgeht. Eine empirische Unter
suchung zur Struktur des Übersetzungsprozesses anfortgeschrittenen Französischlernern.
Tübingen: Narr.
Lörscher, Wolfgang. 1991. Translation Performance, Translation Process, and Translation
Strategies. A Psycholinguistic Investigation. Tübingen: Narr (Language in Performance
4).
McClelland, James L.; Rumelhart, David E. & Hinton, Geoffrey E. 1986. "The Appeal of
Parallel Distributed Processing". In: D. E. Rumelhart & J. L. McClelland (eds.), 3-44.
Molitor-Lübbert, Sylvie. 1989. "Schreiben und Kognition". In: G. Antos & H. P. Krings
(eds.) Textproduktion: ein interdisziplinärer Forschungsüberblick. Tübingen: Niemeyer,
278-296.
Oeser, Erhard & Seitelberger, Franz. 21995. Gehirn, Bewußtsein und Erkenntnis. 2.,
überarbeitete und erweiterte Auflage. Darmstadt: Wissenschaftliche Buchgesellschaft
(Dimensionen der modernen Biologie 2).
Peschl, Markus F. 1994. Repräsentation und Konstruktion: Kognitions- und neuroinformatische
Konzepte als Grundlage einer naturalisierten Epistemologie und Wissenschaftstheorie.
Braunschweig: Vieweg.
Peschl, Markus F. 1990. Cognitive Modelling. Ein Beitrag zur Cognitive Science aus der
Perspektive des Konstruktivismus und des Konnektionismus. Wiesbaden: Deutscher
Universitäts-Verlag.
Risku, Hanna. 1994. "Aktive Expertenroutine oder reaktive Verhaltensautomatik? —
Überlegungen zum Begriff der Übersetzungsfertigkeit bei Wilss". TextConText 3/4, 237-
253.
Risku, Hanna. 1995. Translatorische Kompetenz. Kognitive Grundlegung des Übersetzens als
Expertentätigkeit. Wien: unveröffentl. Diss.
Rumelhart, David E. & McClelland, James L. (eds.) 1986. Parallel Distributed Processing.
Explorations in the Microstructure of Cognition. Volume 1: Foundations. Cambridge:
MIT.
Séguinot, Candace. 1991: "A Study of Student Translation Strategies". In: S. Tirkkonen-
Condit (ed.) Empirical Research in Translation and Intercultural Studies. Tübingen: Narr
(Language in Performance 5), 79-88.
Smith, Veronica. 1994. Thinking in a Foreign Language. An Investigation into Essay Writing
and Translation by L2 Learners. Tübingen: Narr.
Tirkkonen-Condit, Sonja. 1992. "The Interaction of World Knowledge and Linguistic
Knowledge in the Processes of Translation. A Think-Aloud Protocol Study". In: B.
Lewandowska-Tomaszczyk & M. Thelen (eds.) Translation and Meaning, Part 2.
Proceedings of the Lodz Session of the 1990 Maastricht-Lodz Duo Kolloquium on
'Translation and Meaning', Held in Lodz, Poland, 20-22 September 1990. Rijks-
hogeschool Maastricht, Faculty of Translation and Interpreting, 433-440.
Ein kohärentes Translat — was ist das?
Die Kulturspezifik der Texterwartungen
Renate Resch
Für die Textlinguistik, Textwissenschaft und Übersetzungswissenschaft

gleichermaßen sind die Eigenschaften jener Texte, die erfolgreiches Kom-
munizieren ermöglichen, von großem Interesse. Dabei gilt "Kohärenz" als jene
Texteigenschaft, die den inneren Textzusammenhalt herstellt, die Text-Kontext-
Beziehung regelt und so die Einheitlichkeit garantiert und Verständlichkeit des
Textes möglich macht.
In der funktionalen Translationswissenschaft, die das Funktionieren des
Translats als Text in der Zielsituation in den Vordergrund stellt, ist der
Kohärenzbegriff zentral. Für eine Orientierung im Übersetzungsprozeß gilt,
daß die "intratextuelle Kohärenz", also die "Kohärenz-für-den-Rezipienten",
wichtiger sei als die "intertextuelle", also "der skoposadäquate Zusammenhang
zwischen Ausgangs- und Zieltext" (Reiß/Vermeer 1984:119). Die intratextuelle
Kohärenz wird dann auch für die Bewertung von Translaten herangezogen:
"Geglückt ist eine Kommunikation dann, wenn sie vom Rezipienten als
hinreichend kohärent mit seiner Situation interpretiert wird" (1984:112). Diese
"bewußt sehr allgemein gehaltene Regel" wie auch der Hinweis, daß die
Anwendung von Kohärenzregeln kulturspezifisch sei, bleiben jedoch vage und
sollen im weiteren näher betrachtet werden.
In ihren Anfängen versteht die Textlinguistik Kohärenz als ein Phänomen,
das sich an der Textoberfläche manifestiert. Diese ursprünglich rein semantisch
ausgerichtete Beschreibung von grammatischen und syntaktischen Textober-
flächenmerkmalen definiert Kohärenz als das Ergebnis an der Textoberfläche
analysierbarer, logischer Verbindungen zwischen den Sätzen eines Textes. Ein
Blick auf den Text als Ganzes macht jedoch deutlich, daß die einzelnen
Textoberflächenmerkmale nur im Kontext des Gesamttextes ihre kohärenz-
stiftende Wirkung entfalten können, also nur in Relation zu den pragmatischen
Aspekten der Kommunikation. Dem Situationszusammenhang kommt somit
eine entscheidende Rolle zu: Kohärente Texte sind nicht per se kohärent,
sondern erst in Bezug auf ihre Funktion und Verwendungssituation. Die
moderne Textwissenschaft (vgl. Heinemann /Viehweger 1991) wie auch die
Translationswissenschaft (vgl. Snell-Hornby 1986) orientieren sich am Text-in-
Situation und verstehen sich als Interdisziplinen. Zunehmend mißt man den
Rezipierenden die entscheidende Rolle bei der Kohärenzstiftung zu. Kohärenz
wird also nicht mehr als eine Texteigenschaft, sondern als das Ergebnis
kognitiver Prozesse der Textverwenderinnen gesehen. Bezogen auf die
Funktionalität des Textes und seinen Erfolg in der Kommunikationsituation
wird Verständlichkeit und Interpretierbarkeit als ausschlaggebende Kategorie
gesetzt.
Auch schriftliche Kommunikation ist also in großem Maße dialogisch, und
zwar in dem Sinn, als die Kohärenzbildung eigentlich erst im Rezeptionsprozeß
erfolgt: Ein Text kann nur in dem Maß funktionieren, in dem es gelingt, im
Text ein ausgewogenens Verhältnis zwischen den Mitteilungsbedürfnissen der
Textproduzierenden und den Lesebedürfnissen der Textrezipierenden herzustel-
len. Der Text steuert diese Interaktion der Kommunikationsteilnehmerinnen in
Form eines "author-reader-pact", wobei das, was durch Rezeption als
Interaktion entsteht, "something different from the respective contributions of
each" darstellt (Nystrand 1986:40). Das, was die Leserinnen in den Text
einbringen, damit er für sie verständlich werden kann, ist daher der ent-
scheidende Faktor in der Kohärenzbildung; die Strukturen im Text, die die
Leserinnen motivieren, den Autor-Leser-Pakt anzunehmen, sind daher von
entscheidendem Interesse für die Kohärenzforschung.
Im Falle interkultureller Kommunikation wird die Situation noch
komplexer, muß doch die Autor-Leser-Kooperation auch über die Kultur-
grenzen hinweg funktionieren. Für die Translationswissenschaft ist es daher
von besonderem Interesse zu eruieren, wie sich in unterschiedlichen Kulturen
diese Kooperation auf Textebene manifestiert, wie also in unterschiedlichen
Sprachen und Kulturen Textkohärenz zustande kommt. Erkenntnisse dieser Art
können dabei helfen zu definieren, welche Merkmale ein intratextuell
kohärentes Translat auszeichnen, und Grundlagen für das Zustandekommen
intertextueller Kohärenz zwischen den Ausgangs- und Zieltexten legen.
Bezogen auf die konkrete Textform scheinen dabei vorallem drei Textaspekte
eine Rolle zu spielen: der Text als Repräsentant einer bestimmten Textsorte,
die Textdynamik, die die Organisation bekannter und neuer Information im
Text steuert und der interpersonale Faktor, der die Beziehung zwischen
Textproduzierenden und Rezipierenden reguliert.
Renate Resch 273
Kulturspezifische Texterwartungshaltungen: Textsorte, Textdynamik und

interpersonaler Faktor
Textsorte
Entsprechend den Erkenntnissen der modernen Texttheorie funktionieren Texte

in der Kommunikation als Handlungsträger. Soll der Text kohärent und
verständlich sein und der Autor-Leser-Pakt funktionieren, muß den Kom-
munizierenden also klar sein, welche Art von Handlung mit Hilfe des Textes
ausgeführt werden soll. Dazu ziehen die Kommunikationspartnerinnen
Rückschlüsse aus der gegebenen Situation und bilden mit Hilfe ihrer bisherigen
Texterfahrung Erwartungshaltungen in Bezug auf den betreffenden Text.
Dieses Textwissen kann als ein Wissen um die in der entsprechenden Kultur
üblichen Textsortenkonventionen betrachtet werden, die eine Einbettung und
Interpretation des Textes als Teil einer Kommunikationshandlung in einer
Situation und Kultur erst ermöglichen. Diese Texterwartungen in Bezug auf die
inhaltliche und formale Textorganisation, die die Kommunizierenden in den
Text einbringen, steuern und beeinflussen die Kohärenzbildung und damit das
Textverstehen maßgeblich.
Bei interkultureller Kommunikation verläuft dieser Prozeß naturgemäß
komplexer, denn Textsortenwissen als Teil des Weltwissens ist von der
kulturellen Erfahrung des einzelnen geprägt. In der Tat belegen Ergebnisse der
Leseforschung, daß bei interkultureller Kommunikation der Leseprozeß
maßgeblich von den Texterwartungshaltungen der Ausgangskultur beeinflußt
ist: Bei Experimenten nahmen Probanden, die mit Texten aus einem anderen
kulturellen Kontext konfrontiert worden waren und deren Inhalt referieren
sollten, bei der Wiedergabe eklatante Umdeutungen und Auslassungen vor,
während sie sich bei der Wiedergabe muttersprachlicher Texte sich "besser
erinnerten" und unbewußt Ergänzungen, die dem kulturellen und situativen
Kontext entsprachen, vornahmen, (vgl. z.B. Carrell 1987). Weiters legen
Ergebnisse der Schreibforschung nahe, daß die kulturellen Unterschiede eine
entschiedende Rolle dabei spielen, ob ein Text als kohärent und verständlich
aufgefaßt wird oder nicht. Clyne (1991) stellte bei seinen Analysen des
akademischen Aufsatzes im englischen und deutschen Sprachraum maßgebliche
Unterschiede im Textaufbau und Textgestaltung fest, die durch kultur-
spezifische Einstellungen und Werthaltungen gegenüber dieser Textsorte zu
begründen sind: Generell, so schließt Clyne, liege die Verantwortung dafür,
daß ein Text verständlich ist, im deutschen wissenschaftlichen Diskurs bei den
Leserinnen, im englischen bei den Textproduzierenden.
Geht man davon aus, daß im Autor-Leser-Pakt, der Kohärenzbildung erst
ermöglicht, die Textsorte als Orientierung eine zentrale Rolle spielt, geben die
Ergebnisse der interkulturellen Lese- und Schreibforschung zu denken: Die
kulturspezifischen Texterwartungen und Werthaltungen, die die Rezipierenden
bestimmten Textsorten und Gestaltungsmerkmalen entgegenbringen, beein-
flussen ihr Textverstehen und bedingen ihr Urteil über deren Kohärenz und
somit ihren Erfolg. Für den Kontext Translation ist es daher von großer
Wichtigkeit, die Einstellungen gegenüber Textsorten in den unterschiedlichen
Kulturen zu reflektieren.
Textdynamik
Die Erkenntnisse der Prager Schule der Funktionalen Satzperspektive machen

deutlich, wie durch die Verteilung von als bekannt voraussgesetzter bzw. neuer
Information im Text der Informationsfluß — die Textdynamik — gesteuert
wird. Diese unterschiedliche Gewichtung nach Thema (bekannter Information)
und Rhema (neuer Information) trägt wesentlich dazu bei, daß der Text
bedeutungsvoll werden kann. Die Textdynamik ist ein weiterer wichtiger
Aspekt des Autor-Leser-Paktes: Es obliegt den Textproduzierenden ab-
zuschätzen, was für die Rezipierenden als bekannt vorausgesetzt werden kann,
und es ist an den den Textverwenderinnen, die als bekannt gesetzten
Informationen im Text im Rezeptionsprozeß zu ergänzen. Dies erfolgt
entsprechend ihres Vorwissens aus dem gemeinsamen kulturellen Kontext wie
auch aus den im Laufe des Textes bereits eingeführten Informationen. Im Falle
interkultureller Kommunikation ist dieser Vorgang naturgemäß durch die
gegebene Kulturdifferenz komplexer. In der Translationswissenschaft sind in
diesem Zusammenhang immer wieder Kulturspezifika im Ausgangstext, die in
der Zielkultur unbekannt sind und mit Hilfe kompensatorischer Strategien
verständlich gemacht werden müssen, diskutiert worden. Die Kulturspezifik der
Textdynamik manifestiert sich jedoch auch auf sehr viel grundsätzlicher Ebene,
wie deutlich werden wird.
Zentral für den Zusammenhalt und die Nachvollziehbarkeit jedes Textes
ist die Themenprogression. Diese Wiederaufnahme von schon Genanntem bzw.
das Nennen von allgemein Bekanntem erleichtert den Rezipierenden die
Verständnisbildung; neue Informationen werden zum Thema in Beziehung
gesetzt, wodurch der Textfortschritt ermöglicht wird. Wie die Textdynamik in
einem bestimmten Textexemplar fortschreitet, ist intrakulturell von der
Kommunikationssitutation und dem Kommunikationsziel abhängig. Variabel
sind dabei die Informationsdichte bzw. Redundanz, die für eine bestimmte
Renate Resch 275
Kommunikationssitution angebracht erscheint; weiters textdynamische

Fokusierungsstrategeien, wie z.B. die Umkehrung der üblichen Thema-Rhema-
Abfolge, die eine Kontrast- bzw. Hervorhebungswirkung erzielen. Allgemein
läßt sich feststellen, daß eine oftmalige Wiederholung des Themas, eine
konsequente Verknüpfung des Neuen mit schon Bekanntem und eine dem
Kommunikationsziel entsprechende Fokusierung einen Text verständlicher,
leichter zugänglich und somit kohärenter machen.
Wie sehr sich jedoch die Toleranzgrenzen bei den Rezipierenden von
Kultur zu Kultur zu unterscheiden, und das auch bei vergleichbaren Kom-
munikationssituationen und Kommunikationszielen, belegen die Ergebnisse
kulturkontrastiver Untersuchungen ein und derselben Textsorte. Bei dem
Vergleich englisch- und deutschsprachiger wissenschaftlicher Beiträge für
Fachzeitschriften einer bestimmten Fachrichtung (vgl. Clyne 1991) zeigten sich
eklatante Unterschiede z.B. in der Einbettung neuer Information in den Text,
die in den deutschen Beiträgen viel weniger konsequent erfolgte als in den
englischen. Verantwortungsvolles Übersetzen verlangt also auch eine
Berücksichtigung dieses Aspektes des Autor-Leser-Paktes. Ein Außerachtlassen
der kulturspezifischen Ausprägungen der Textdynamik lassen Translate u.U.
unzusammenhängend und damit inkohärent erschienen. Da diese kultur-
spezifischen Unterschiede sogar bei einer Textsorte mit einem hohen Grad an
Standardisierung und Internationalisierung festgestellt wurden, kann die
Relevanz dieser Überlegungen für die Translation nicht genug betont werden.
Interpersonaler Faktor
Ein Aspekt des Autor-Leser-Paktes, dem in kulturkontrastiven Studien bisher

wenig Beachtung geschenkt wurde, ist der interpersonale Faktor schriftlicher
Kommunikation. Im Rahmen der englischen Registerlinguistik wurden jene
Textmerkmale untersucht, die, wenn sie gehäuft auftreten, entweder den
Leserinnen oder den Autorinnen mehr Präsenz im Text einräumen — auf diese
Weise positioniert der Text die Kommunikationspartner in ihrer Beziehung
zueinander. Je nachdem, ob gehäuft Merkmale der Leserpräsenz oder der
Autorpräsenz auftreten, macht diese Rollenverteilung Texte mehr oder weniger
interaktiv. Das Maß an Interaktivität ist von der Textfunktion abhängig und,
wie zu sehen sein wird, auch kulturspezifisch.
Für Überlegungen dieser Art relevante Textmerkmale umfassen grammati-
sche Phänomene, wie Satzmodus, Person, Adverbien und Deixis, die mit dem
Gesamttext und seiner Funktion in Beziehung gesetzt werden. Textmerkmale,
die Leserpräsenz und also ein hohes Maß an Interaktivität signalisieren,
umfassen ein vermehrtes Auftreten von Personalpronomen in der 2.Person,

direkte Bezüge zur Rezeptionsituation, Imperative und Fragesätze, bzw.
Textmerkmale mit diesen illokutiven Bedeutungen. Autorpräsenz wird im Text
durch vermehrtes Auftreten von Sätzen mit unpersönlichen Subjekt, Passivkon-
struktionenen, explizit oder implizit zum Ausdruck gebrachte Wertungen und
Bezüge zur Situation der Textproduktion signalisiert (Smith 1985).
So manifestiert sich im Text der interpersonale Aspekt, eine weitere
wesentliche Komponente des Autor-Leser-Paktes: Die Rollenzuweisung an die
Rezipierenden und ihre Bereitschaft diese anzunehmen, tragen wesentlich dazu
bei, daß der Text als sinnhaft und kohärent eingeschätzt wird. Intrakulturell
variiert diese Beziehungsebene der Kommunikationsteilnehmerinnen je nach
Textfunktion. So gilt z.B. für das Englische, daß appellative Texte das höchste
Maß an Interaktivität aufweisen (Smith 1985). Daß diese Forschungsergebnisse
nicht unhinterfragt auch aufs Deutsche angewendet werden können, wird die
Beispieldiskussion zeigen — auch hier manifestieren sich interkulturelle
Unterschiede, die für die Translation relevant erscheinen.
Beispieldiskussion1
Die Grundlage der Beispieldiskussion bildet ein authentischer Übersetzungs-

auftrag, nämlich die Übersetzung einer deutschsprachigen Werbebroschüre
eines österreichischen Hotels ins Englische. Die untersuchten Materialien (siehe
Anhang) umfassen den Ausgangstext (AT) und zwei Übersetzungen, eine von
einer professionellen Übersetzerin, die den zu bearbeitendenText in ihre
Muttersprache übersetzte (Ül) und eine von einer Studentin, die im Rahmen
eines Übersetzungsseminars am Institut für Übersetzer- und Dolmetscher-
ausbildung der Universität Wien den Text aus ihrer Muttersprache ins
Englische übersetzte (Ü2). Auf der Basis der bisherigen Ausführungen werden
der AT und die beiden Übersetzungen diskutiert.
Textsorte — Ebene der Texthandlung
Annahmen über den Handlungszusammenhang in der Kommunikationssituation

und sich daraus ergebende Erwartungen über die inhaltliche Organisation des
Textes beeinflussen die Kohärenzbildung. Werbetexte agieren in der Kom-
1
Die Texte wurden mir von Dr. Kaiser-Cooke zur Verfügung gestellt, wofür ich mich
herzlich bedanke.
Renate Resch 277
munikation zwischen Anbieterin und Konsumentin als Handlungsträger für die

Handlung "zum Kauf überreden" und weisen eine typische Inhaltsstruktur auf:
Den ersten Themenschwerpunkt bildet das "Problem", das sehr oft erst im
Text und explizit für die angestrebte Zielgruppe wenn nicht geschaffen, so
doch zumindest als solches formuliert wird und ein Bedürfnis in den Ziellese-
rinnen hervorrufen soll. Diesem folgt die "Lösung" in Form eines Produkt-
angebotes. Bezogen auf den Beispieltext kann angemerkt werden, daß sowohl
Ausgangs- als auch Zieltext in Gesellschaften mit sog. freier Marktwirtschaft
wirken sollen, in denen Werbung eine so große Rolle spielt, daß diese typische
Inhaltsstruktur als Kulturwissen vorausgestzt werden kann; eine Anpassung der
inhaltlichen Gesamttextorganisation ist daher nicht notwendig. Allerdings sind
auch die einzelnen Textabschnitte in Bezug auf die Texthandlung zu unter-
suchen: das "Kurzum" im ersten Absatz des AT z.B. erfüllt nicht so sehr die
Funktion zusammenzufassen, sondern fungiert vielmehr als rhetorische Figur
des Fokusierens, was in Ü2 nicht erkannt wurde.
Textdynamik — Ebene der Informationsverteilung
Die in der Übersetzungsliteratur oft behandelte Frage der kulturspezifischen

Begriffe wird im AT am Begriff "wellness" aktuell, ironischerweise an einem
Wort, das den Eindruck erweckt, es wäre der Sprache der Zielkultur
entnommen. Die notwendige Kompensierung dieses Informationsdefizits wird
in Ül durch den Zusatz "as we call it" erreicht.
Betrachtet man die eigentliche Textdynamik, also die Abfolge von Thema
und Rhema, weisen die beiden übersetzten Texte eklatante Unterschiede auf.
In Ü2 wird eine wiederholte Rhematisierung, also die Voranstellung neuer
Information, besonders im 3. Absatz sehr deutlich: Durch die Abfolge von
"Merkur recreation" — "a health service" — "the philosophy" — "risk
factors" zu Satzbeginn, deren rhematisierende Wirkung durch die Artikelset-
zung ("a health service") noch verstärkt wird, verliert der Text den Zu-
sammenhalt. Demgegenüber garantiert die einheitiche Thematisierung in Ül
("Merkur Recreation" — "it" — "its philosophy" — "the service") die
Textlogik und Kohärenz. Ein Blick auf den AT macht klar, daß Ü2 sich sehr
stark an der Textdynamik des AT orientiert, wo die Rhematisierung zwar nicht
ideal aber immerhin nicht kohärenzstörend als Fokusierungmittel eingesetzt
wird. Offensichtlich gibt die Autor-Leser-Kooperation im Deutschen den Text-
produzierenden mehr Freiheit in der Gestaltung der Textdynamik. Die
kulturspezifischen Normen und Texterwartungshaltungen verlangen also hier
eine Anpassung an die zielsprachlichen Textnormen, wie sie in Ül erfolgt ist.
Interpersonaler Faktor — Beziehungsebene
Vergleicht man AT und die Übersetzungen in ihrer Verwendung von

Personalpronomen der 2.Person, Imperativen, Bezugnahme auf die Rezipien-
tlnnen — also jenen Textmerkmalen, die ein hohes Maß an Interaktivität im
Text sigalisieren — stellt man erhebliche Unterschiede fest: dem sachlichen
also autorzentrierten Ton im deutschen AT (unpersönliches Subjekt, keine
direkte Bezugnahme auf die Leserinnen) stehen im Englischen sehr interaktiv
angelegte Texte gegenüber. Die Rolle der Kommunizierenden in der Werbe-
kommunikation im Englischen und Deutschen ist tatsächlich grundlegend
verschieden. Im Englischen ist der Autor-Leser-Pakt durch eine große Nähe
gekennzeichnet und die Kommunikation eher informell, während im Deutschen
die Konventionen der Fachkommunikation nachgeahmt werden, die durch
Textmerkmale, die Distanz und sog. Objektivität signalisieren, charakterisiert
ist. Ü2 weist jedoch auch hier wieder eine Orientierung am AT auf: "man's
wellbeing", "his wellness" im ersten Absatz lassen die britischen LeserInnnen
unangesprochen und unterminieren so die persuasive Wirkung, die auf der
Wortebene angestrebt wird. Ganz ähnlich wirkt der auf Wortebene angstrebten
Dramatisierung des "Problems" der Über- und Unterbeanspruchung bestimmter
Körperzonen im Alltag (vgl. AT) die unpersönliche, teilweise passivisch
formulierte Wendung "soothe the overtaxed parts of the body and revitalise
those that have been neglected" entgegen. In Ül hingegen wird die Inter-
aktivität vorallem durch syntaktische Mittel in allen Textabschnitten durch-
gehalten, was sowohl den inneren Textzusammenhalt fördert als auch den
Autor-Leser-Pakt, wie er sich im Englischen manifestiert, stärkt.
Natürlich wären anhand der Beispieltexte noch weitere Aspekte der
Kohärenzbildung im Zusammenhang mit dem kulturspezifischen Autor-Leser-
Pakt zu diskutieren, so vor allem Aspekte der stilistischen Kohärenz, was aber
hier aus Platzgründen nicht geleistet werden kann. Aspekte der Handlungs-
ebene, der Informationsverteilung und der interpersonale Faktor — jeweils auf
den Gesamttext bezogen und vor dem Hintergrund des kulturspezifisch
geprägten Autor-Leser-Paktes betrachtet — scheinen jedoch erste Rückschlüsse
auf die Frage der intra- und intertextuellen Kohärenz bei Translation
zuzulassen.
Renate Resch 279
Schlußfolgerungen
Entsprechend den Erkenntnissen der modernen Textwissenschaft und der

interkulturellen Lese- und Schreibforschung kann Kohärenz nur in Interaktion
zwischen dem Text und den Rezipierenden zustande kommen, genauer gesagt,
bestimmten Textmerkmalen und den kulturspezifisch geprägten Texterfahrun-
gen der Textverwenderinnen, in einem Autor-Leser-Pakt, der sich im Text
manifestiert.
Grundsätzlich scheint die im Text geleistete Integration von zumindest drei
Textaspekten eine große Rolle im Kohärenzbildungsprozeß zu spielen:
Zuallererst muß auf Grund des Kontextes für die Rezipierenden die Möglich-
keit gegeben sein, den Text als Handlungsträger zu erkennen, weiters muß für
sie die im Text gegebene Informationsverteilung und damit die Bewertung
dieser als schon bekannt (thematisch), neu (rhematisch) bzw. für das Kom-
munikationsziel besonders relevant (fokusiert) nachvollziehbar sein; schließlich
müssen die Leserinnen die ihnen zugeteilte Rolle im Kommunikationsprozeß
annehmen können / wollen (interpersonaler Faktor). Erst ein Zusammenwirken
dieser drei Ebenen ermöglicht, daß Rezipierende einen Text als kohärent,
sinnhaft erleben können. Da diese Textmerkmale kulturspezifisch geprägt sind,
erfordert eine erfolgreiche Integration der kohärenzstiftenden Textebenen im
Text vor allem im Falle interkultureller Kommunikation großes Text- und
Weltwissen in den betreffenden Kulturen. Translation macht also eine textuelle
Neuorganisation der zu transferierenden Inhalte nach Handlungs-, Informa-
tions- und Beziehungsebene notwendig — Übersetzen als professionelle
Tätigkeit ist also immer ein Neu texten.
Kohärenzbildung in Texten als Ergebnis eines Autor-Leser-Paktes zu
betrachten, ermöglicht es, kohärenz intratextuell zu beschreiben und einige
Aspekte der Kulturspezifik der Kohärenzregeln näher zu betrachten. Eingedenk
der unendlich vielen Möglichkeiten, die Sprache bietet, sinnhaft und also
kohärent zu sein, und eingedenk der empirisch unfaßbaren und nur theoretisch
beschreibbaren Texterfahrung, die ein Individuum als Mitglied einer Kultur in
den Text einbringt, muß schließlich wohl auch akzeptiert werden, daß
"Kohärenz" immer nur in Annäherung erreicht werden kann.
Bibliographie
Benson, J.D./Greaves, W.S. (eds.) 1985. Systemic Perspectives on Discourse. N.J.: Ablex.
Carrell, Patricia. 1987. "Content and Formal Schemata in ESL Reading". TESOL Quarterly
21 (3), 461-481.
Clyne, Michael. 1991. "The Sociocultural Dimension: The Dilemma for the German Scholar".
In: H. Schröder (ed.), 49-68.
Heinemann, Wolfgang/Viehweger, Dieter. 1991. Textlinguistik. Tübingen: Niemeyer.
Nystrand, Martin. 1986. The Structure of Written Communication. Orlando: Academic Press.
Reiß, Katharina/Vermeer, Hans J. 1984. Grundlegung einer allgemeinen Translationstheorie.
Schröder, Hartmut. 1991. Subject-oriented Texts. Berlin: de Gryter.
Smith, E.L. Jr. 1985. "Functional Types of Scientific Prose". In: J.D. Benson/W.S. Greaves
(eds.), 241-257.
Snell-Hornby, Mary. 1986. Übersetzungswissenschaft — Eine Neuorientierung. Zur
Integrierung von Theorie und Praxis. Tübingen: Francke (UTB 1415).
Anhang
(AT)
WELLNESS GUIDE
Ein Service der Merkur Recreation
Bad Tatzmansdorf.
Als Urlaub oder zwischendurch.

Körper, Geist & Seele haben immer Saison.
Der Alltag ist hart genug. Hohe berufliche Belastung und einseitige Lebensweise führen häufig
zu innerer Unausgeglichenheit. Körperliche, geistige und seelische Dissonanzen können die
Folge sein. Der Mensch ist durch sich selbst in seinem Wohlbefinden gestört. Kurzum: Seine
Wellness ist beeinträchtigt.
Urlaub sollte der Erholung und Entspannung dienen. Und noch besser: die überbeanspruchten
Zonen beruhigen und die unterbeanspruchten vitalisieren, sodaß eine Harmonisierung und
Stärkung des Gesamtorganismus eintritt. Um dieses Ziel zu erreichen, dafür ist einmal Urlaub
pro Jahr zuwenig. Denn das Jahr hat zwölf Monate.
Die Merkur Recreation zeigt den Weg. Ein Gesundheitsservice, der schwerpunktartig durch
das ganze Jahr begleitet. Dessen Philosophie ist die ganzheitliche Betrachtung des Menschen
und die Förderung seiner Gesundheit in Partnerschaft zwischen Natur und Medizin.
Risikofaktoren werden frühzeitig erkannt, Leistungskraft und Vitalität von Grund auf erneuert.
Das 5-Sterne-Hotel Steigenberger in Bad Tatzmansdorf und die Merkur Recreation bereiten
einen schwungvollen Einstieg in das neue Lebensgefuhl: 6 gute Gründe für einen Urlaub oder
ganz einfach ein paar Tage zwischendurch, zusammengefaßt in den Wellness-Tips 1-6 — auf
den folgenden Seiten. Alle Termine und Spezialangebote finden sie übersichtlich zusammen-
gefaßt im Umschlag.
(Ü 1)
Take a few days off
any time of the year .
Feel free to feel good
Life is hard enough. Stress on the job and a one-sided life style often lead to mental and
emotional imbalance and disturb the harmony between body, mind and soul, destroying our
well being — or "wellness", as we prefer to say.
Holidays should primarily be a source of repose and recreation, or- better still- should relax
those parts of you that have been under undue stress, and revitalise those that have not had
enough to do, strengthening body and soul and restoring the harmony between them. And one
holiday every twelve months just isn't enough.
Renate Resch 281
Merkur Recreation shows you what to do:
it offers a health service that focuses on different aspects of health and harmony throughout
the year. Its philosophy is a holistic approach to the human being and the promotion of his/her
health through a partnership between nature and the art of medicine. The service helps to detect
risk factors before it is too late, and to restore your vitality and performance.
The five-star Hotel Steigenberger at Bad Tatzmannsdorf, and the Merkur Recreation service
it offers, show you how to achieve this new way of being. We offer six good reasons why you
should take a holiday or at least indulge in a few days "in between" — six wellness tips, which
you can find on the subsequent pages. An easy-to-read summary of dates and special offers can
be found on the cover.
(Ü 2)
Take a holiday or a few days off any time you want
Body, mind & soul are never off season.
Everyday life is hard enough. Great stress at work and a one-sided way of living often lead
to internal imbalance affecting the physical, mental and emotional equilibrium. Man's well-
being is impaired by man himself. In short: his wellness is impaired.
Holidays should help you recover and relax- and even better: soothe the overtaxed parts of the
body and revitalise those which have been neglected, in order to harmonize and strengthen
your whole organism. One holiday in twelve months is not enough to reach this goal.
Merkur Recreation shows you the way: a health service accompanying you the whole year
through focusing on different aspects of wellness. The philosophy behind it is a holistic
approach to man and the promotion of health in harmony between nature and medicine. Risk
factors are detected at an early stage and your performance and vitality are thoroughly
restored.
The 5-star Hotel Steigenberger in Bad Tatzmansdorf and Merkur Recreation help you get into
this new feeling for life easily: we give you six good reasons for a holiday or simply a few
days in-between summed up in the Wellness Tips 1-6 on the following pages. Please find all
dates and special offers in a short summary on the cover.
Murder in the laboratory — Termhood and the
culture gap
There has been a lot of talk in the last few years about the role of cultural
knowledge in translating, about translating being a trans-cultural activity and
involving some form of cultural transfer. At the same time, the question has
arisen as to what sort of knowledge translators require in order to translate
efficiently, what their 'knowledge bases' should or do consist of and how
training institutions can help novices to acquire a critical mass of such
knowledge. The tendency has been to divide up this knowledge into categories
such as linguistic, cultural and subject-area knowledge, which is certainly
convenient from the point of view of delimiting a specific area of study, but
unfortunately somewhat clouds the issue of how all these 'different types' of
knowledge, often regarded as a static collection of data, interact to form active
know-how — i.e. specifically translatorial expertise.
I shall try to illustrate, on the basis of one or two examples, not only that,
for the purposes of translating, linguistic, cultural and subject-area knowledge
are one and the same thing, but that this so-called knowledge-base (knowing-
that) is in fact know-how, a process which is activated and adapted for each
new translation task, i.e. text. In other words, translationally relevant
knowledge cannot simply be stored as data or pre-specified rules; knowledge
is always task-specific — in our case, text-specific. It is therefore ad-hoc and
necessarily incomplete. Each new translation requires the application of
different aspects of 'knowledge about the world', so that we can never claim,
or indeed need to have a 'complete' collection of domain-specific knowledge
— whether in our minds or in a computerised database.
LSP and term recognition
Many translation teaching institutions, and indeed many translators, regard

texts which represent communication between experts of a specific domain of
knowledge (LSP texts; Fachtexte) as somehow different from other, 'general
language' texts. They are seen as more difficult because of unfamiliarity with
the subject in question, but on the other hand, so the assumption often goes
'once the terminology has been found', all you have to do is more or less slot
it into place. What exactly do we mean by 'term' or 'Fachausdruck'? Terms
are generally taken to be the unique, subject-related use of nouns (and
sometimes verbs). They are "Any conventional symbol for a concept which
consists of articulated sounds or of their written representation. A term may
be a word or a phrase" (ISO/R 1087 definition, in: Picht/Draskau 1985:96).
Further, the characteristics of the term which distinguish it from the non-term
are precision and the fact that it belongs to a system of terms, which are the
linguistic manifestation of a system of concepts, "That the system of terms is
the linguistic representation of a system of concepts may exert an important
influence on the formation of terms" (Picht/Draskau 1985:97).
However, automatic term recognition is only a fiction in the minds of
those who develop so-called translation support systems and not a fact in the
life of the real flesh and blood translator. The studies by Peter Schmitt (1986;
1990) have shown how little translators and especially students can rely on
precision and 'its place in a system' to identify terms and even less on one-to-
one term equivalence, even in very restricted domains. Two examples from the
following text excerpt (from a paper for an international congress) will be used
to discuss why this is so and why the translatorial know-how required for LSP
texts is, in essence, the same as that required for other types of text.
(1) Cochlear prostaglandins under Streptomycin influence
- Tissue preparation:
Die Versuchstiere wurden nach entsprechender Behandlungsdauer mit Pentobarbi-
tal (100mg/Kg) getötet. Nach der Tötung wurden die Tiere thorakotomiert und
ein Katheter über den linken Ventrikel in die Aorta ascendens vorgeschoben. Die
untere Hohlvene wurde eröffnet und 100 ml einer isotonen heparinisierten 4° C
warmen Kochsalzlösung infundiert, um eine Auswaschung des Blutes aus dem
Innenohr zu gewährleisten. Nach Abtrennung des Kopfes wurde der Schädel
entlang der Sagittalebene gespalten und die beiden Bullae entfernt. Die
anschließende Präparation der lateralen Schneckenwand erfolgte unter stereo-
mikrskopischer Sicht in Tris-HCl-Puffer.
According to purely denotational criteria, the word töten in this text ("den Tod
von jemanden, etwas herbeiführen, verursachen, verschulden", Duden) is co-
Michèle Kaiser-Cooke 285
extensive with to kill ("deprive of life or vitality, put to death, cause death of",
Concise Oxford Dictionary). However, familiarity with the connotations of
kill, knowing that it does indeed mean more than simply 'cause to die' and
implies wilful malice, will indicate that this is 'not the right answer'. The so-
called "objective" meaning of the word has a cultural overlay which makes it
inappropriate in a clinical emotionally neutral laboratory setting. For this
domain of experience, the properties of the concept behind the English word
to sacrifice are weighted in such a way that the emotional connotations are
suppressed to evoke a more neutral causation of death.
Nominalisation of to sacrifice, on the other hand, as a possible translation
for Tötung, would reactivate the situational associations of slaughter, victims,
giving up a valued or desired thing, and thus call up emotional elements
similar to those of kill. Even in a highly restricted subject field and one and the
same text, a concept's level of abstraction brings factors into play which are
essentially of a cultural nature and blur the line between specialist and
"general" knowledge.
'Technical' terms and general language words
The crux of the problem is in fact the separation of LSP from general language
(see Hoffmann 1987 and 1988; Kalverkämper 1978 and 1987; Jumpelt
1961:130), which becomes acute when considering the notion of LSP
phraseology; can, for example, verbs be terms? Do changes of preposition,
ellipsis, domain-specific collocation create terms? If so, when do they stop
belonging to general language and start acquiring LSP characteristics (see Picht
1988)? Why should any of this matter? The traditional argument is that terms
represent specific concepts in a given subject field, which remain constant
interlingually, even though the term changes. Thus it is important to recognise
a term in order to identify the concept, which has only one equivalent in the
target language. Those elements of language which are not terms (implication
— which do not represent specific concepts) are allegedly less precise, less
domain-specific and therefore do not matter so much — they are simply the
cement which holds the terminology together.
The notion of LSP is based on the assumption of an essential difference
in communication between specialists in certain subject fields and others,
namely, that the terminology (i.e. concepts) used by them are more precise,
stable and, in particular, identifiable. The fixation of conventional terminology
theory on nouns as terms (cf. Wüster 1979; Picht/Draskau 1985) and the
insistence that concepts "belong to concept systems" shows that the main
feature of 'termhood' is in fact its susceptibility to precise description.
Technical terms are so manifestly not the only words or groups of words
representing concepts that it is almost banal to state it. Language in its entirety
functions with concepts — those in 'everyday' language are apparently less
readily defined (cf. Wierzbicka 1985) and less easily 'assigned to a system'.
But the system is obviously there, otherwise language would not work. Terms
are supposed to be more precise than general language words, but, just as in
general language, the concepts represented by these terms are malleable, and
terms can vary in meaning depending on their context, even within one and the
same subject field and the same text (cf. Spitzbardt 1972; Picht 1988:192;
Trillhaase 1972). The difference between LSP texts and general language texts
is that translators usually have at least minimum knowledge of the concepts
underlying the GL texts, and have to acquire this minimum — and sometimes
even more — for more specialised areas of knowledge. The problems are
essentially the same — creating cohesion and coherence by making the
concepts which underly the text mutually accessible in such a way as to fulfill
the expectations of the readers.
The study of LSP phraseology, as well as terminology, has to do with
investigating "the conceptual relations between LSP (language) elements which
combine to produce a valid and linguistically correct statement" (die begriff-
lichen Beziehungen — sowie deren Veränderungen- zwischen fachsprachlichen
Elementen ..., die zu einer fachlich gültigen und sprachlich korrekten Aussage
zusammengefügt werden können) (Picht 1988:193). If the LSP element were
left out of this statement, it would be applicable to all texts. The difference is
one of subject field, target group, function and situation, not of essential
characteristics.
The idea of 'mapping terms' from one text to another (ST to TT), is
based on the assumption of concept invariance in at least those domains of
experience which conventionally fall under the heading of Fachwissen or
subject-specific expertise, which is also the philosophical basis of traditional
terminology theory. This is in turn based on the assumption that so-called
'scientific' knowledge is 'objective' and interculturally invariant (for a
discussion of the culturally determined construction of all aspects of reality,
including scientific knowledge, see especially Costazza 1993 and Wallner 1990
and 1993). In these domains, once the concept has been identified and
delimited, translation is seen as simply a question of label-swapping. It is
interesting to note that, when it comes to LSP texts, translation scholars seem
to have little difficulty in accepting the notion of complete logical equivalence
(i.e. that concepts can be completely co-extensive), whereas equivalence has

otherwise become something of a four-letter word. At the same time, with one
or two exceptions, Quine's theory of ontologicai relativity and indeterminacy
of translation (1960; 1969) are non-starters for the translating community since,
in blatant refutation of Quine's thesis, translation is of course not only possible
but often even successful. Theory-wise, this is somewhat schizophrenic: we
must either accept the notion of equivalence and reject relativity as a
fundamental principle, or reject equivalence and accept some form of
relativity.
Concepts and cultural knowledge
The sort of concepts which are relevant in translating are, to quote the
psychologist Frank Keil (1989:148), 'the subset of concepts which have lexical
labels! Whether we all actually conceptualise in the same way is not as
relevant for our purposes as the question of how we communicate about what
we conceptualise. On the basis of theories put forward by cognitive linguistics
(cf. Lakoff 1982,1986,1987; Langacker 1988; Talmy 1986), in psychology (cf.
Quinn/Holland 1987) and philosophy (Plotkin 1994; Grossmann 1992; Munz
1993), I would like to draw attention to the notion that the labels we attach to
our concepts conventionalise how we generally see things, i.e. that at a given
moment the linguistic label specifies which part of the content of a concept is
to be articulated for the purposes of social communication. Linguistically
articulated concepts (which are the ones that primarily concern us) are
malleable, ideational constructs with fuzzy edges and a variable distribution of
properties or attributes — i.e. they do not simply consist of necessary and
sufficient properties with a stable distribution (cf. Rosch 1973, 1976, 1978;
Rosch/Mervis 1975; Tabakowska 1993; Zelinsky-Wibbelt 1988) For a given
text, the cotext and context will specify which properties — which include both
connotational and denotational elements — are highlighted and which are
suppressed. The textual and contextual indicators for concept specification vary
both from language to language and from text type to text type.
How does all this tie in with culture and translatorial expertise? The labels
affixed to concepts represent those elements of experience of the world which
the speech or cultural community in question regards as worth communicating
about — in other words, they embody culturally relevant knowledge (cf. also
Beaugrande 1994). This knowledge is always domain-specific in that it is
relevant to a specific domain of experience — whether potty-training or brain
surgery. When we translate, we have to know which 'knowledge of the world'

the respective speech community regards as culturally relevant (language and
target-group specific). In other words, which textual conventions we use to
activate which conceptual content — how to form a coherence pattern using the
appropriate relationships (appropriate to the target group in question) between
the appropriate concept properties (cf. also Reiß 1984). This form of
conceptual shift — establishing new concept relations by activating different
properties, weighting them according to target culture conventions of
conceptualisation — must happen for every type of text if the translation task
is to be successfully completed. Concept invariance or formal equivalence will
be the exception rather than the rule — and then only at isolated nodes within
the system, which in turn have to be connected and integrated into the system
as a whole. The translator's know-how consists in the ability to accomplish this
process. In other words, s/he must identify the hidden structures and implicit
presuppositions (in the original sense of the term) in the source text — which
are not on the text surface — and use these as a basis for constructing a new
coherence pattern, which will have different conventions of what is considered
relevant knowledge, different implicit presuppositions etc. Translating always
involves identifying and articulating culturally relevant aspects of knowledge,
no matter which domain of knowledge or experience we are dealing with.
Conclusions
Linguistically articulated 'real-world' knowledge is culturally significant know-

ledge of the world. Cultural knowledge in the context of translating constitutes
the interface between language and domain of experience — or subject area.
There is essentially no difference between 'subject-specific knowledge'
(no 'objective', culturally neutral knowledge) and other types of 'real-world'
knowledge for translation purposes.
The notion of specifically translatorial expertise is not compatible with the
suggestion that translators must also possess 'subject-area expertise'. Our
subject area expertise is knowing how to translate — and not just 'knowledge
of 2 languages plus a specific subject-area databank;
We have to look for different criteria for identifying terms, which take
into account cultural differences — and therefore a new approach to the
concept of translating LSP texts. Is there an essential difference between
'technical terms' and phraseology and other context-dependent, subject-
dependent etc. parts of speech?
The question of novice translators' knowledge acquisition should be

examined from the perspective of knowledge as a purpose-oriented and
contextual, i.e. task-specific activity. This provides the basis for explaining the
nature of translatorial expertise as a dynamic, active process.
Translating is always intercultural — not just in the case of isolated texts
or text elements.
References
Beaugrande, Robert de. 1994. "Cognition, communication, translation, instruction: The

geopolitics of discourse". In: R. de Beaugrande/A. Shunnaq and M. H. Heliel, (eds.)
Language, Discourse and Translation in the West and the Middle East. Amsterdam:
Benjamins, 1-22.
Costazza, Markus. 1993. "Die Abkehr vom Wahrheitsparadigma in der Wissenschafts-
philosophie". In: R. Fischer/M. Costazza and A. Pellert (eds.) Argumentation und
Entscheidung. Zur Idee und Organisation von Wissenschaft. München/Wien: Profil, 193-
242.
DUDEN Deutsches Universalwörterbuch, 1983. Drosdowski, Günther (ed.) Mannheim:
Dudenverlag.
Grossmann, Reinhardt. 1992. The Existence of the World. An Introduction to Ontology.
London: Routledge.
Hoffmann, Lothar. 1987. "Der Fachtext als strukturierte und funktionale Ganzheit". In: L.
Hoffmann (ed.) Fachsprachen: Instrument und Objekt. Leipzig: VEB Verlag Enzyklopä-
die, 49-63.
Hoffmann, Lothar. 1988. "Makrostruktur und Kohärenz als Fachtextsortenmerkmale". Wissen
schaftliche Zeitschrift der Karl-Marx-Universität Leipzig, Gesellschaftswissenschaftliche
Reihe, 37/6, 552-565.
Jumpelt, Rudolf. 1961. Die Übersetzung naturwissenschaftlicher und technischer Literatur.
Berlin-Schöneberg: Langenscheidt.
Kalverkämper, Hartwig. 1978. "Die Problematik von Fachsprache und Gemeinsprache". In:
R. Schützeichel (ed.) Sprachwissenschaft, Bd 3 (1978). Heidelberg: Carl Winter
Universitätsverlag, 406-444.
Kalverkämper, Hartwig. 1987. "Vom Terminus zum Text". In: M. Sprissler (ed.) Standpunkte
der Fachsprachenforschung. Tübingen: Gunter Narr, 39-78.
Keil, Frank C. 1989. Concepts, Kinds and Cognitive Development. Cambridge MA/London:
The MIT Press.
Lakoff, George. 1987. "Cognitive models and prototype theory". In: U. Neisser (ed.)
Concepts and conceptual development: Ecological and intellectual factors in categoriza
tion. Cambridge and New York: Cambridge University Press, 63-100.
Lakoff, George. 1986. Cognitive Semantics. Cognitive Science Program, Institute of Cognitive
Studies, University of California at Berkeley.
Lakoff, George. 1982. Categories and cognitive models. Trier: LAUT.
Langacker, Ronald W. 1988. "A View of Linguistic Semantics". In: B. Rudzka-Ostyn (ed.)
Topics in Cognitive Linguistics. Amsterdam, Philadelphia: John Benjamins, 49-90.
Munz, Peter. 1993. Philosophical Darwinism. On the origin of knowledge by means of natural
selection. London: Routledge.
Picht, Heribert and Draskau, Jennifer. 1985. Terminology: An Introduction. Guildford, Surrey:
The University of Surrey.
Picht, Heribert. 1988. "Fachsprachliche Phraseologie". In: R. Arntz (ed.) Textlinguistik und
Fachsprache. Akten des Internationalen übersetzungswissenschaftlichen AILA-Sym-
posiums, Hildesheim, 13.-16. April 1987. Hildesheim, Zürich, New York: Georg Olms,
187-196.
Plotkin, Henry. 1994. The Nature of Knowledge. Concerning Adaptations, Instinct and the
Evolution of Intelligence. London/New York: Allen Lane/Penguin.
Quine, Willard van Orman. 1960. Word and Object. Cambridge MA: The M.I.T. Press.
Quine, Willard van Orman 1969. Ontologicai Relativity and Other Essays. New York,
London: Columbia University Press.
Quinn, Naomi and Holland, Dorothy. 1987. "Culture and Cognition". In: N. Quinn and D.
Holland (eds.). Cultural Models in language and thought. New York and Melbourne:
Cambridge University Press, 3-42.
Reiß, Katharina. 1984. "Methodische Fragen der übersetzungsrelevanten Textanalyse. Die
Reichweite der Lasswell-Former. Lebende Sprachen 1, 7-10.
Rosch, Eleanor. 1978. Cognition and categorization. Hillsdale NJ: Lawrence Erlbaum
Associates.
Rosch, Eleanor. 1976. "Basic Objects in Natural Categories". Cognitive Psychology 8, 382-
439.
Rosch, Eleanor. 1973. "Natural Categories". Cognitive Psychology 4, 328-350.
Rosch, Eleanor and Mervis, Carolyn B. 1975. "Family Resemblances: Studies in the Internal
Structure of Categories". Cognitive Psychology 7, 573-605.
Schmitt, Peter. 1986. "Die 'Eindeutigkeit' von Fachtexten: Bemerkungen zu einer Fiktion".
In: M. Snell-Hornby (ed.) Übersetzungswissenschaft: Eine Neuorientierung — Zur
Integrierung von Theorie und Praxis. Tübingen:Franke, 252-282.
Schmitt, Peter. 1990. "Kulturspezifik von Techniktexten: Ein translatorisches und terminologi-
sches Problem". In: H. J. Vermeer (ed.) Kulturspezifik des translatorischen Handelns.
Vorträge anläßlich der GAL-Tagung 1989. Heidelberg: Universität Heidelberg, 49-88.
Spitzbardt, Harry. 1972. "Die Vielschichtigkeit des Problems wissenschaftlicher und
technischer Übersetzung". In: H. Spitzbardt (ed.) Spezialprobleme der wissenschaftlichen
und technischen Übersetzung. Halle (Saale): VEB Max Niemeyer, 13-32.
Tabakowska, Elzbieta 1993. Cognitive Linguistics and Poetics of Translation. Tübingen:
Gunter Narr.
Talmy, Leonard. 1986. The Relation of Grammar to Cognition. Berkeley CA: Cognitive
Science Program, Institute of Cognitive Studies, University of California at Berkeley.
The Concise Oxford Dictionary of Current English. 1990. H.W. Fowler/F.G. Fowler (eds.)
Oxford: Clarendon Press.
Trillhaase, Günther. 1972. "Polysemie" und "Kontext" in der Translation". In: H. Spitzbardt
(ed.) Spezialprobleme der wissenschaftlichen und technischen Übersetzung. Halle (Saale):
VEB Max Niemeyer, 87-122.
Wallner, Fritz. 1990. Acht Vorlesungen über den Konstruktiven Realismus. Wien: WUV.
Wallner, Fritz. 1993. "Der Konstruktive Realismus. Theorie eines neuen Paradigmas". In: F.
G. Wallner/J. Schimmer and M. Costazza (eds.) Grenzziehungen zum Konstruktiven
Realismus. Wien: WUV, 11-23.
Wierzbicka, Anna. 1985. Lexicography and conceptual analysis. Ann Arbor: Karoma.
Wüster, Eugen. 1979. Einführung in die Allgemeine Terminologielehre und Terminologische
Lexikographie. Schriftenreihe der TU Wien, Bd. 8, Teile 1 u. 2. Wien/New York:
Springer.
Zelinsky-Wibbelt, Cornelia. 1988. "From Cognitive Grammar to the Generation of Semantic
Interpretation in Machine Translation". In: E. Steiner/P. Schmidt and C. Zelinsky-
Wibbelt (eds.) From Syntax to Semantics. Insights from Machine Translation. London:
Pinter, 105-132.
A model for translation of legal texts
Dorte Madsen
The purpose of this paper is to present an approach to translation of legal texts

that focuses on the relationship between the legal text and the extralinguistic
reality. It is assumed that the essential factors that are relevant to translation
of legal texts can be accounted for on the basis of an analysis of three
universes (cf. Madsen 1994), that is
— a legal universe
— a textual universe
— and a translator's universe
An analysis of the legal universe presupposes a model description of the
extralinguistic reality, i.e. the reality in which legal actions are performed and
state-of-affairs are created. It is also the reality in which legal texts serve a
function. In the textual universe the actions performed in the legal universe are
described; the actions are fixed, so to speak, in a text. And finally in the
translator's universe, the texts describing the legal actions are translated.
By way of introduction I place these universes within the prototypological
framework of Snell-Hornby, i.e., the diagram (1988:32) in which a system of
relationships is established between basic text-types and the crucial aspects of
translation. As Snell-Hornby states (1988:31f.) the diagram represents a
stratificational model which proceeds from the most general level (A) at the
top, downwards to the most particular level (F) at the bottom. The translation
type I am dealing with within the conventional areas of translation at level A
is Special Language Translation, and the text type among the basic text types
focussed on in the prototypology at level B is legal language texts. Level B
corresponds to the textual universe of my model. Level C in the Snell-Hornby
diagram shows the non-linguistic disciplines, i.e. the areas of extralinguistic
reality which are reflected in the legal universe of my model. And finally,
level D in Snell-Hornby's diagram is the level dealing with important aspects
and criteria governing the translation process itself (cf. Snell-Hornby 1988:34).
This level (D) is represented in my model as the translator's universe.
The extralinguistic reality
As mentioned, the legal universe of my model reflects level C in the Snell-

Hornby diagram showing the areas of extralinguistic reality. Snell-Hornby
states (1988:34) that the extralinguistic reality is "inseparably bound up with
translation." This statement is indeed true of translation of legal texts, as has
been shown in Madsen (1994) which gives an account of the legal, textual and
translator's universes within an action theoretical framework.
Along the lines suggested by Baumann and Kalverkämper (1992:13f.),
who emphasize the necessity for interdisciplinarity when analysing complex
forms of LSP communication, Madsen (1994) analyses the ties between the
legal text and the legal reality in terms of the internal structure of the three
universes and their inter-dependence, as well as the consequences of these ties
for.translation.The analyses have revealed that the cornerstone of a model for
translation of legal texts must be the rooting of the legal text in a legal system.
In the present paper I will first illustrate the interdisciplinary approach
giving an account of the rooting of a legal text. The text used as an example
is an 'agreement', which can be seen as the textual manifestation of an act
pattern <AGREEMENT>, cf. below. Then I will address the issue of how
the rooting of a legal text relates to translation.
The action theoretical framework
The action theoretical framework developed for translation of legal texts is

based on the action theory of Jochen Rehbein (1977). The central principle of
Rehbein's theory is that actions are performed within patterns. With the
concept of act pattern, Rehbein establishes a relationship between the dynamic
actions performed by specific agents in specific situations on the one hand and
the social embedding of the action in question on the other.
The action theory of Rehbein refers to human actions in general, and thus
it has not been developed with a specific view to translation (cf. also Holz-
Mänttäri 1984:26). In the present study, the Rehbein action theory serves as
my starting point for an analysis of the legal actions taking place in the legal
universe. The reason why the Rehbein theory has been chosen as a point of
departure is that Rehbein bases his analysis of actions on the reality in which
Dorte Madsen 293
the actions are taking place; in other words Rehbein's basis for an analysis of
actions is a model of reality reflecting the structures of a given society. These
structures are, in turn, reflected in the static aspect of the act pattern.
The dynamic aspect of an act pattern, on the other hand, is reflected in the
implementation of the pattern, i.e. when it is actually put to use by specific
agents performing specific actions in specific situations. First I shall deal with
the static aspect, and I shall then consider the act pattern <AGREEMENT>
from a dynamic perspective.
The existence of an act pattern implies that a line of actions has achieved
some degree of stability. According to Rehbein (1977:126) a pattern has an
inherent purpose (Zweck). Consequently, it is the purpose that constitutes the
pattern. The purpose constituting an act pattern must be considered from the
angle of society, i.e., the more static perspective. Rehbein's theory is based
on the assumption that a community is organized in such a way that the
performance of actions has superior social purposes. Consequently, agents in
a given society apply an act pattern when they pursue superior social purposes.
The static aspect of the pattern, i.e., the structures of society existing at
a given moment in a given society are illustrated in my basic model shown as
figure 1 below. The basic 3-level structure (A, B and C) illustrates what
Rehbein calls a Handlungsraum (1977:12ff), the notion of space of action.
FIGURE 1: Space of action (Handlungsraum)

The whole idea of operating with a space of action is to provide a framework
for the factors that influence the actions performed within this particular space
of action. This means that the conditions that in one way or another determine
the actions performed in that space of action can be dealt with theoretically.
And the point is that each time an action is performed within the same space
of action, a great part of the conditions of the action will be more or less the
same over a certain period of time, cf. Rehbein (1977:12).
As will appear from figure 1 above, all cooperation in a given society is
arranged on three levels according to the nature of agents performing actions
and according to the level of abstraction, cf. Rehbein (1977:102f.). Hence
level A represents the direct interaction between individual members of
society, level B illustrates more stable types of cooperation, as is found in e.g.
institutions. And finally, level C, which is the most abstract level, represents
all cooperation in a given society, cf. gesamtgesellschaftlicher Handlungsraum
in Rehbein's theory (1977:117). Roughly, it is level C that ensures the
relationship and interaction between the two levels of cooperation A and B. It
is by virtue of level C that levels A and B can interact.
The basis of this static model is a dynamic, historical process during
which act patterns have been institutionalized, i.e. in the course of time the
behaviour of individuals has assumed the form of patterns; consequently, when
an individual pursues an act pattern he also pursues its inherent purpose. Thus,
the institutions at level B can be seen as the means by which superior social
purposes are obtained.
As a consequence of Rehbein's differentiation between individuals and
institutions, I operate with macro-level purposes on level B and micro-level
purposes on level A. The latter I designate goals. This division of purposes is
in accordance with Rehbein and his differentiation between Zweck, which
designates the superior social purpose of the pattern, and Ziel, which
designates the micro-level goal of an individual person, cf. Rehbein
(1977:137).
The legal universe
To analyse the legal universe I transfer the abstract basic model shown in
figure 1 above to a specific section of society designated the legal universe, cf.
figure 2 below, as it is assumed that the legal universe can be accounted for
as a space of action in Rehbein's sense. The only changes made in figure 2
compared with the basic model of a space of action, are that in the legal
Dorte Madsen 295
universe the only institution of interest at level B is the legal system, and the
agents at level A are not only individuals but also legal persons or entitites.
FIGURE 2: Legal universe

It is also assumed that the concept of act patterns applies to the legal universe;
the regularity of the actions performed within the legal universe is so
pronounced that it cannot be considered a mere coincidence. Thus, the pattern
of < AGREEMENT > is an example of an action frequently performed in the
legal universe. The existence of the pattern < AGREEMENT > implies that
the community provides for the possibility that individual members of the
community can in fact do business with each other — that one person can
confer rights on another. The community has even laid down rules governing
the implementation of the pattern <AGREEMENT> in specific situations.
What is characteristic of the actions performed in the legal universe
compared with other spaces of action is that they are subject to the law in force
at a given time. This is why the notion of 'valid law' appears at level B in the
model. Another way of expressing this is that the law in force determines first
which actions, or omissions, are possible or prescribed within the legal
universe in question; secondly valid law determines the legal effects of these
actions or omissions.
The textual universe
As mentioned above it is in the textual universe that the actions performed in

the legal universe are described. Consequently an 'agreement' as text is
considered a description of possible or prescribed actions and/or omissions that
have been or are to be performed by the parties to the contract as a result of
their contractual relationship. So, when two parties sign the document
'agreement' they also apply the act pattern <AGREEMENT> provided for
by the community. And vice versa, by means of a text, the act pattern
<AGREEMENT> is fixed in time and space. Hence it follows that the legal
text is the means by which the major part of legal actions are performed. This
instrumental notion of text reflects the relationship between the situations in the
legal universe where the actions are actually taking place and the function of
the legal text.
To further analyse the actions described in the text 'agreement' it is
necessary to refer to the Rehbein concept of Gesamthandlung, which is the
overall action required to implement an act pattern, cf. Rehbein (1977:85f.).
Briefly, it is the Gesamthandlung that represents the dynamic aspect of an act
pattern, i.e., the pattern-in-use by specific agents in specific situations. But one
single agent can rarely implement a pattern by performing only one specific
action; this specific action is most frequently only one part of the
Gesamthandlung, and is therefore called a Teilhandlung, cf. Rehbein (1977).
Thus the agent performing partial actions will have to cooperate with other
agents to be able to perform a Gesamthandlung. For instance, the major partial
actions to be carried out in order to implement the pattern <AGREEMENT>
are <delivery> and <payment>. The partial actions may be seen as a string
of actions which all together realise the Gesamthandlung and consequently
represent the pattern < AGREEMENT > as they are carried out within the
same pattern. Thus, the Gesamthandlung AGREEMENT may be seen as the
dynamic counterpart of the static pattern < AGREEMENT > . This also means
that the notion of Gesamthandlung may be seen as the link between the basic
model reflecting the structures of a given society, i.e. the static aspect of a
pattern and the partial actions that contribute to the realisation of the pattern
in a specific and dynamic situation.
According to Rehbein (1977:82f.) a Gesamthandlung is constituted by
three phases: prehistory — history — posthistory. An analysis of the Gesamt
handlung AGREEMENT shows that the prehistory is the conditions for the
conclusion of a valid agreement; the history is the performance of the
agreement and the posthistory is the legal effects of the agreement. So with this
Dorte Madsen 297
notion of Gesamthandlung the actions that are described in legal texts can be
analysed in terms of their legal effects determined by valid law.
By way of example, figure 3 below introduces a model of the textual
universe of an 'agreement'. As will appear, the model builds directly upon
figure 2. It is assumed that the parties to the agreement referred to in figure
3 have signed a contract of sale. When two parties sign a contract, they apply
the act pattern <AGREEMENT>. This also means that the act pattern
< AGREEMENT > is fixed in time and space by means of a text, as the text
is the means by which the agreement is concluded.
FIGURE 3: Textual universe: "Agreement"

The space of action illustrated above is the Danish legal universe, and the
parties to the contract are Danish legal persons. In the specific situation the
text using the pattern <AGREEMENT> is written in the Danish language.
Consequently, the Danish contract is the specific use of a pattern rooted in the
Danish legal universe. That the pattern is rooted also means that the actions
described in the text, for instance <payment> and < acquisition > , cf. figure
3, are subject to Danish law.
As a consequence of the determination of 'valid law' from level B to level
A the model includes the names of two central Danish acts, i.e., the Danish
Sale of Goods Act and the Danish Contracts Act which apply to the contractual
relationship. This determination implies that the legal effects of the actions
performed on level A are established by means of 'valid law' at level B.

The example shows an act pattern <AGREEMENT> corresponding to
a Gesamthandlung which is performed by means of a text 'agreement'. As the
agreement is concluded in the Danish legal universe, Danish law applies to the
contractual relationship, which also means that it is Danish law that determines
the effects of the legal actions described in the 'agreement'. The text is
therefore rooted in the Danish legal universe.
Translation
The question is now what does this rooting mean to translation? Imagine the
following two apparently similar situations, in which the agreement is
translated from Danish into Spanish. In both situations the 'agreement' is
concluded in Denmark between a Spanish company and a Danish company,
respectively.
In the first situation (1) the conclusion of the agreement, the performance
of the agreement and the legal effects of the agreement, i.e. the whole
Gesamthandlung, is subject to Danish law, and in the second situation (2) the
conclusion, the performance and the legal effects of the agreement is subject
to Spanish Law.
(1) If the contract is subject to Danish Law, the Danish Sale of Goods Act
and Danish Contracts Act will apply to the contractual relationship. (2) If, on
the other hand, the contract is subject to Spanish Law, the Spanish Código
Civil and Código de Comercio will apply to the contractual relationship.
Consequently the rooting of the 'agreement' will depend on the choice of law.
Considered from the point of view of translation theory, in both (1) and
in (2) the source text is Danish and the target text is Spanish. But as a
consequence of the rooting of the target texts in different legal universes, the
Spanish target text will have to be orientated toward the Danish legal universe
in (1), because it is Danish law that determines the conditions for the
Gesamthandlung AGREEMENT, i.e. the conclusion of a valid agreement, the
performance and the non-performance as well as the legal effects of the actions
performed by the parties. And in (2) the Spanish target text will have to be
oriented towards the Spanish legal universe, because in that case it will be
Spanish law that determines the conditions for the Gesamthandlung
AGREEMENT, i.e. the text will be rooted in the Spanish legal system.
So the two situations outlined above are different in that the choice of law
determines the rooting of the Spanish target text and therefore its orientation
Dorte Madsen 299
toward either of the two legal systems involved. The problem is, however, that
in translation theory, situation (1) outlined above would be treated as source
language orientation, and (2) as target language orientation. The point is,
however, that if we take seriously the fact that the extralinguistic reality is
"inseparably bound up with translation" (cf. Snell-Hornby 1988:34), it will not
suffice to account for situation (2) as an instance of target language orientation,
as it is the entire legal universe of the target text that determines its orientation
towards the sources of law applicable. It is therefore suggested that the
traditional dichotomy between source language orientation and target language
orientation does not apply to legal translation.
References
Baumann and Kalverkämper. 1992. "Kontrastive Fachsprachenforschung — ein Begriff, ein

Symposium und eine Zukunft. Zur Einführung." In: Klaus-Dieter Baumann and Hartwig
Kalverkämper (eds.) Kontrastive Fachsprachenforschung.Tübingen'. Narr (Forum für
Fachsprachenforschung; Bd. 20), 9-25.
Holz-Mänttäri, Justa. 1984. Translatorisches Handeln. Theorie und Methode. Helsinki:
Suomalainen Tiedeakatemia
Madsen, Dorte. 1994. Den juridiske teksts forankring i en retsorden. En handlingsteoretisk
analyse af oversætterens beslutningsgrundlag. Ph.D. dissertation, Handelshøjskolen i
København, ARK nr. 77.
Rehbein, Jochen. 1977. Komplexes Handeln. Elemente zur Handlungstheorie der Sprache.
Stuttgart: Metzler.
Benjamins.
New ideas from historical concepts: Schleiermacher
and modern translation theory
Irene Rübberdt/Heidemarie Salevsky
Part I of this contribution (Heidemarie Salevsky) is designed to present some

of the main ideas of Schleiermacher's "Hermeneutics" and "Criticism" and to
establish its relevance for translation criticism.
In Part II (Irene Rübberdt), these hypotheses will be illustrated, using a
Biblical text (Gen 16 and 21:8-21) as an example.
Part I
Schleiermacher's hermeneutics and the problem of translation
Schleiermacher felt that the interpretation of a text might not take place on the
basis of an established canon and looked upon understanding as a circular
movement (cf. Schleiermacher in: Schreiter 1988 and in: Frank 1993). The
hermeneutic circle means a repeated return from the whole to the parts and
vice versa, grasping the sense of the parts from the whole, which is subject to
constant development. The ever-increasing spiral opens up and incorporates
ever new sense connexions, encompassing them like a vortex. Schleiermacher
sought to describe understanding as a provisional and unlimited process whose
starting point can only be grasped from the perspective of the individual's life
(cf. Schleiermacher in: Schreiter 1988:277).
According to Schleiermacher there are two forms of interpretation/under-
standing: "divinatory" (die divinatorische Methode) which aims at intuitive
perception of what is individual and "comparative" (die komparative Methode)
which proceeds from the general to the particular by way of comparison.
Neither one can be separated from the other because divination depends on a
supportive comparison for corroboration, and comparison alone does not
guarantee unity. The general and the particular must penetrate each other, and
this invariably takes place through divination. The idea of a work can only be
grasped by taking into account two aspects: the material and the setting. The
material alone does not necessitate a particular form of implementation.
Schleiermacher's hermeneutics has not been sufficiently turned to account
so far in the context of translation science and, in a more narrow sense, for
translation theory and criticism (but cf. first ideas on the subject in Vermeer
1994).
As far as Schleiermacher's lectures on hermeneutics are concerned (cf.
Schleiermacher in: Schreiter 1988), I consider the following main ideas to be
of special relevance for the problem of translating.
A. Any utterance can only be understood from the perspective of the entire
life context to which it belongs, as an aspect of the speaker's life that is
dependent on all other aspects of his life, and the latter can only be
determined by taking into account the sum total of the settings which
determine his development and future existence (Schleiermacher in:
Schreiter 1988:256).
B. Any speaker can only be understood through the prism of his nationality
and the age in which he lives (cf. Schleiermacher in: Schreiter 1988:256).
C. Hermeneutics is art in the sense that the activity is invested with the
character of art because the rules provide no recipe for application and
therefore do not permit any mechanization (Schleiermacher in: Schreiter
1988:257f.).
This means that the translator requires both psychological and linguistic talent.
According to Schleiermacher psychological talent is of two kinds. The
extensive talent can easily recreate or even anticipate other people's manner of
acting. The intensive talent involves an understanding of the (real) significance
of a human being and his peculiarities in relation to the concept of human
being. Both are necessary, but almost never combined in one and the same
person.
Linguistic talent is — according to Schleiermacher — a sense of analogy
and difference. A comparative grasping of languages in terms of their
differences (extensive linguistic talent) is to be distinguished from penetrating
the innermost recesses of language in terms of thinking (intensive linguistic
talent). Both are necessary, but almost never combined in one person
(Schleiermacher in: Schreiter 1988:258).
Irene Rübber dt/Heidemarie Salevsky 303
Schleiermacher and translation criticism
As Schleiermacher sees it, any act of understanding is the inversion of an act

of speaking so that hermeneutics and criticism belong together. Like
Schleiermacher's reflections on hermeneutics, his theses about criticism are
univeral in character. Those that are most important for the problem of
translation criticism are listed here.
(i) The assessment of human actions takes place on the basis of what ought
to be in relation to certain laws, ways of life, etc. Invariably, the first
point is to agree on maxims or axioms before further aspects can be
considered for the purpose of assessment. This includes the function of the
text. Schleiermacher distinguishes between four categories of the "purpose
of the work"/Bestimmung des Werkes (quoted from Frank 1993:272f.):
texts
(a) for aesthetic enjoyment (zum ästhetischen Genuß),
(b) for educational purposes (zum Schulgebrauch),
(c) for philological use (zum philologischen Gebrauch) and
(d) for the critical reader (für den kritischen Leser).
(ii) Criticism means establishing the result of decisions within the framework
of existing options. The difference between the option and the result may
be large or small, but it is always there. It is necessary to look at the way
the option corresponds with reality (the result) and then explore in what
kind of ways the assumed relation of identity could be lost, producing a
difference.
What matters in making an assessment is the genesis of the difference.
The differences concern omissions, additions and changes of various
kinds. Among the latter, it is important to distinguish between intentional
and unintentional changes.
"We may attribute any of these, more or less, to the following two cases:
1. If someone introduces something of his own making into the text,
whatever it may be, this invariably amounts to intentional falsification. 2.
If someone adds something because he feels that what he finds in the text
is inadequate and needs correcting, this amounts to a deliberate
change...Such a change may be meant as an improvement, and it may
really be one, but may also be based on error. In all these cases, the
changes are intentional, albeit in different ways" (Schleiermacher quoted
after Frank 1993:276).
Schleiermacher felt that criticism should first proceed from the assumption
that the changes are unintentional and only then contemplate the likelihood
of intentional changes (cf. Schleiermacher quoted after Frank 1993:280f.).
(iii) As criticism can only be practised on a one-off basis, all but the most
general rules may be formulated. What matters, invariably, is to what
extent the whole state of affairs can be grasped. However, we need clues
to proceed from, and on the other hand, a point that has emerged from the
connexions with what is to be explained. So the assessment of a
translation is always a concrete affair and must never take place in
isolation from its genesis. In other (modern) words: For a translation to
be properly assessed it is necessary to establish the connexion between the
translation process and its product.
The implication of the subjective in the process of perceiving and anticipating
(i.e. understanding the ST and producing the TT) turns the structural aspects
of translation as an activity — in a manner of speaking — into structural
aspects of the assessment via its reflection in the target text. The assignment
(laying down the function of the TT) and the prevailing circumstances, together
with the human factor (translator), determine the one-off character of every
translation and hence the weighting needs within the predetermined limits of
decision-making.
Now what are these decision and test regions like which have so far been
ignored or insufficiently taken into account in the models of translation or
translating proposed so far?
In his writings, Friedrich Daniel Ernst Schleiermacher dealt with various
antinomies in the process of understanding, which I believe may be regarded
as crucial for the translator's decisions. They are to be viewed as a network
(that can hardly be represented here). In other words, the translator has to
make decisions not only within the context of one antinomy, but redefines its
quality ever anew by relevance attributions within the fabric of all antinomies.
These antinomies can be seen as decision or problem regions by the
translator and may therefore, differ in number, kind and relevance attribution
with regard to the assignment, the text type, the addressees as representatives
of a cultural, linguistic and communicative community (including certain norms
and conventions) in the source and target text area and other factors affecting
the translation process. They can be part of the planning and of the basis of
assessment.
Irene Rübberdt/Heidemarie Salevsky 305
Here are some of the antinomies which Schleiermacher considered relevant:

— Purpose: congruity (Bestimmung:Stimmigkeit)
What is the relationship between (a) text function and congruity in the ST and
TT (congruity ranks higher than effect) and (b) between the requirements of
the translator's assignment and his freedom of decision?
— Impact situation production situation (Wirkungsmoment: Entstehungs
moment)
In order to fully comprehend the intended effect the text must be traced back
to the situation in which it originated and to its significance in the author's life.
The text and its structure thus permit conclusions as to activity and its
structure.
— Whole:part (Ganzes:Teile)
What are the (historical, literary, religious, social, cultural, linguistic) contexts
of the text? Which of these are relevant and to what extent? How are partial
texts interwoven with other partial texts and with the overall text, or how
autonomous are they?
— Intentional : unintentional (Absichtliches: Unabsichtliches)
What is intended by the author in the text, what is not intended and thus open
to interpretation ("offen får die freie Handlung" in the words of
Schleiermacher)?
— Thinking:presentation (Denken:Darstellen)
What is the idea behind the theme, the person, the situation, and how is it
explicitly presented in the text? What implicit pieces of information,
associations and trains of thought are suggested by way of connotation?
— Significant:insignificant (Bedeutendes: Unbedeutendes)
What is significant or insignificant for whom, when, where and why? What is
moved into the focus, e.g. through titles and headings? What is omitted or
added?
— General particular (Allgemeines: Besonderes)
What are the specific features of the text as compared to general or typical
features of language, text type and text content?
— Psychological : technical (Psychologisches: Technisches)
What is the relationship between the emotional and the rational element in the
text or how does it present itself in text production and reception? What
weighting is required with regard to function in a specific situation and how
can it be achieved by the individual translator?
— Permanent: mutable (Stetes: Bewegtes)
Are the intentions pursued of an innovative or conservative kind? What is the
relationship between tradition and topicality?
— Analogy:difference (Analogie: Differenz)

What is perceived as familiar or unfamiliar, and by whom? Is the unfamiliar
to be retained in the text or is it better to resort to analogy? To what extent is
the translator (intra- and interlingually) under the influence of other translations
relating to the same ST?
Part II
Now the importance of the antinomous relations mentioned above will be

pointed out for the modelling of the translation process and for translation
criticism using Gen 16 and 21:8-21 of the Old Testament as an example.
According to the Genesis narratives of Gen 16 and 21:8-21 the Egyptian
slave woman Sarai (Sarah) becomes the mother of Abram's (Abraham's) first-
born son Ishmael who, as it turns out, is not the son promised by God. Hagar
is oppressed and finally expelled by Abraham and Sarah because of the rivalry
between the two women/sons.
Using selected concrete decisions made by translators, we shall now
attempt to illustrate the dynamics of the network, which is made up of the
decision regions and which, time and again, either supports the translator's
decisions or calls them into question.
As far as the Biblical story chosen here and its potential for interpretation
are concerned, the antinomies whole:part, significant: insignificant,
psychological: technical, thinking: presentation, analogy:difference,
purpose:congruityand permanent: mutable appear to be of special relevance for
this one-off concrete web of antinomies, each rendering serving a different
function.1 The dominant aspect is the whole:part antinomy, which applies to
the text of Gen 16 and 21:8-21 at several levels, namely 1. in the way it is
interwoven into the interreligious dialogue (cf. Kahl/Salevsky 1996), 2. the
way it is interwoven into the Bible with regard to the story of Abraham, the
Book of Genesis and the link between the Old and the New Testament, and
3. in its quality as a textual entity of its own, which enters into a relationship
with its elements. Therefore, the whole:part antinomy, which must be defined
differently each time, and its correspondent or dominant relationship with other
antinomies are at the centre of the following reflections.
1
Not only translations in a narrow sense have been included, but also translated quotations
in scholarly and popularized editions (e.g. Procksch 1924; Lutz/Timm/Hirsch 1970),
renderings for artistic purposes (e.g. Kaegi 1914) and adaptations for children (e.g.
Belloso 1990; Weth 1992).
The relevance of the whole:part antinomy in the context of the story of

Abraham
The story of Hagar is part of the story of Abraham, and even if Abraham
hardly ever speaks himself here, it revolves around a test of faith for the man
who has since been regarded as the very epitome of unshakeable belief (Rom
4, Hebr 11:8-19). In an interpretation focussing on the history of the
patriarchs, the Hagar story provides a "spannendes und verzögerndes Motiv
[...] zwischen Verheißung und Erfüllung" (exciting and delaying motive ...
between promise and fulfilment; Procksch 1924:113). This would help to
explain the line taken in the "significant: insignificant" decision region by
numerous translators and adapters who, through the use of subheadings in the
context of the story of Abraham, relate the women's tale of Sarah and Hagar
so expressly to the men (who are less important in the context of the Hagar
story and hence in a different part:whole relation): "Abraham (!) will Gott
nachhelfen: Hagar und Ismael" (Abraham wants to lend God a hand: Hagar
and Ishmael) in Die Gute Nachricht 1991; "Abrams neues Versagen und Gottes
treues Zurechtbringen" (Abram's renewed failure and God's unfailing
correcting hand) in Bruns (1962)}
The relevance of the whole:part antinomy in the overall Biblical context
When we look at the Hagar story in the context of the Old and New
Testament, it can now be interpreted against the background of the Christian
tradition according to Gal 4:29-30, which leads to the main emphasis being
laid on the figure of Sarah.3 A clear sign of the Sarah bias is the
disambiguation of the Hebrew word zachak in Gen 21:9 as mocking or
scoffing.4 Let us recall the situation: The "wild ass" Ishmael (Gen 16:12), the
very embodiment of the humiliation suffered, remains a thorn in Sarah's side
even after the birth of Isaac. At the feast given on the day Isaac is weaned,
Sarah sees the son of Hagar mocking, which triggers her protective maternal
2
See also Weth (1992) and Kaegi (1914).
3
Cf. for example the subheadings of Gen 16 in Menge (1933); Luther (1928); Károli
[Hung.] (1991).
4
Text versions after Luther I checked, cf. Buber/Rosenzweig (1925/1976); Neue Welt
(1971); Simon (1976); Schlachter (1923); Bruns (1962); Weth (1992); New King James
(1985); Károli [Hung.] (1842); Russian Bible (1904-07); Polish Bible (1980); Czech Bible
(1991).
instincts and/or arouses fears about her prestige and inheritance. However this
may be, Ishmael must be expelled.
The neutral German word lachen (laugh) is ambivalent like the Hebrew
zachak.5 The translator has three options: 1. to choose a technological-rational
approach in the psychological:technological decision region and translate the
ambiguous zachak into the equally ambiguous and neutral lachen, 2. opt for
a psychological-emotional approach and substitute for lachen a semi-
synonymous word such asplay,Jest or romp about, thus interpreting Ishmael's
behaviour as childlike and playful, but harmless, or 3. to disambiguate
Ishmael's laughing unequivally as an act of mocking, scoffing (directed against
Isaac).6 The two latter options show how the whole:part antinomy corresponds
with other decision regions. The disambiguation of zachak, taking into
consideration implicit text information (in the decision region
thinking: presenting) is, on the one hand, attributable to subsequent re-writing
in an overall Biblical context (sanctioning of the Sarah bias in Gal 4:29f.),
implicit information being made explicit by decisions in the
thinking: presentation region. Admittedly, theanalogy:difference antinomy also
plays a part in the decision-making process. It is well known that the Protestant
Gâspâr Kâroli, while working on the Hungarian Bible, was strongly influenced
by Luther and may have taken his cue from him in Gen 21:9.7 (As regards the
decisions in theanalogy:difference region in connection with Gen 16:3, cf.
Kahl/Salevsky 1996: 151-156). But the disambiguation of zachak along the lines
of scoffing is also attributable to a striving for congruency within the text of
Gen 16/21:8-21, which may also be taken as a textual entity. Negative
5
The same Hebrew word can be found elsewhere in the context of our story, namely in
Gen 21:6: Ein Lachen hat mir Gott bereitet (God had made me to laugh) says Sarah on
learning that she is pregnant. Ein jeder, der es hört, wird mir zulachen. (So that all that
hear will laugh with me.) (Arenhoevel/Dreissler/Vögtle 1965). The note in
Arenhoevel/Dreissler/ Vögtle describes the laughing in 21:6 as ein Lachen aus Freude
(laughing for joy). But not all translators have seen it that way. In Luther (1992) we read
instead: Gott hat mir ein Lachen zugerichtet; denn wer es hören wird, der wird über mich
6
lachen. (God has bestowed a laugh on me, for whoever will hear it will laugh at me).
In fact, not all translators relate Ishmael's behaviour explicitly to Isaac. This is clearly
done only in Bruns (1962); the Zurich Bible (1960); Rad (1949); Lutz/Timm/Hirsch
(1970); Arenhoevel/ Dreissler/Vögtle (1965); Procksch (1924) and in the adaptation of
Weth (1992).
7
Und Sara sahe den son Hagar der Egyptischen den sie Abraham geborn hatte das er ein
spötter war. (Luther 1534/1983); Mikor látta vólna pedig Sarah az Àgyptumbéli
Hágárnac fiát [...] hogy czufolnâ (az Ishákot) (Károli 1590/1981).
evidence is provided, inter alia, by the Old Testament as selected and rendered
by Jörg Zink, which does not contain the Hagar story at all.
The relevance of the part: whole antinomy for the story of Hagar as a textual
entity
Within the framework of the Hagar story as a textual entity of its own, Gen
21:9 corresponds with other Biblical passages in close proximity. Thus
Ishmael's character is predicted already in Gen 16:12: "And he will be a wild
ass: his hand will be against every man, and every man's hand against him... "
Gen. 16:12 and Gen 21:9 link up with Gen 25:18 to form a logical chain of
thought: Ishmael, the wild ass, mocks little Isaac, and his rebellious nature
remains typical of his descendants (Gen 25:18)8. Against this background, it
is possible to explain why, for example, the translation of Buber/Rosenzweig
(1925f/1976) disambiguates the Hebrew zachak as spottlachen (mock) contrary
to all expectations. The decision in the purpose:congruity region, which in
literal translations (cf. classification in Kassühlke 1976:168) would suggest
closer adherence to the original because of this function (purpose), is here
outweighed by the decision in the whole:part region taken in the interest of
congruity.
But an interpretation focussing on the figure of Hagar also presupposes
that the story is invested with an added weight of its own and with a
significance going well beyond that of a catalyst within the story of Abraham.
The emphasis laid on Hagar by means of subheadings9 can be evidence of this
(cf. also the reflections on the relationship between the whole:part and the
significant:insignificant antonomy in 3.). However, when zachak in Gen 21:9
is rendered by the neutral word laugh or, even more important, disambiguated
by choosing harmless words such as play or romp about, Sarah's decision is
unmistakably ascribed to the high-handedness of someone in authority. In the
current context of growing anti-foreign sentiment and nationalistic tendencies
in Europe, Hagar thus comes to symbolize all those who are strangers and
deprived of rights. Future translators of the Bible will hardly be able to ignore
8
Since Gen 25:18 is almost identical with the last part of the sentence in Gen 16:12 (And
he shall dwell in the presence of all his brethren), Gen 15:18 can be hardly interpreted
geographically in this context as, e.g., Kautzsch (1894) or Menge (1933) do. The latter,
in a note, at least draws attention to the alternative translation option. Incidentally, neither
Kautzsch nor Menge render zachak as spottlachen (mock), but use the neutral word
lachen (laugh) (Kautzsch) or the more harmless romp about (Menge).
Reuß 1908, Procksch (1924); Lutz/Timm/Hirsch (1970); Slovak Bible (1991).
the topical relevance of the Hagar story when making decisions in the
permanent: mutable region. Kahl (1994:31) has referred to the significance of
the Hagar story (especially Gen 16:11-14) as a "theological provocation".
Similarly, in the interpretation of the Spanish Children's Bible by Belloso,
the Hagar story assumes an importance of its own, which corresponds with the
way it is woven into the interreligious dialogue. Gen 16/21:8-21 is woven here
into an internationalist image of God. This reading places Ishmael, as the
ancestor of the Arabian tribes and of Islam, in the focus of attention and sees
the Hagar story against a Muslim background. The fact that in this context
Hagar's expulsion in Gen 21:10-14 not only marks the cruel end of a dramatic
conflict, but also signals the beginning of a new, promising chapter of history
likewise initiated by a divine promise is also suggested by the French landscape
painter Claude Lorrain's artistic "translation" of Gen 21:8-21 in his painting
The Expulsion of Hagar (1668), which rather than conveying a wild sense of
drama as in Pieter Lastman's painting Hagar's Farewell (1612) tells of faith
in God and the future as expressed in the calm and harmony of the landscape.
Finally, the picture also recalls that in Gen 21 Hagar is not only expelled, but
granted her freedom. The road to freedom, however, always involves a
journey into the unknown.
As we have tried to demonstrate, the ambivalence of the story surrounding
the two women, Hagar and Sarah, not only calls for a sophisticated rational
understanding of the text on the part of the translator, but also — over and
above linguistic ability in the sense of Schleiermacher — for a psychological
ability to project oneself emotionally into all dimensions of the text with due
regard for the different whole:part antinomies and their implications. The tasks
of translation criticism, which need to be redefined, will therefore have to
reflect the body-and-mind issue to a far greater extent.
References
Humboldt, Wilhelm von. 1836-1839/1962. "Über die Kawi-Sprache auf der Insel Java". In:
Werke in 5 Bänden. Bd. 1, Teil 1. Darmstadt: Wissenschaftliche Buchgesellschaft.
Kahl, Brigitte. 1994. "Hagar: Gott denken aus der Perspektive der anderen". Ökumenischer
Informationsdienst 4, 30-31.
Kahl, Brigitte/Salevsky, Heidemarie. 1996. "Auf der Suche nach Hagar". In: H. Salevsky
(ed.) Dolmetscher- und Übersetzerausbildung gestern, heute und morgen. Akten des
internationalen wissenschaftlichen Kolloquiums anläßlich des 100jährigen Jubiläums der
Dolmetscher- und Übersetzerausbildung Russisch ander Berliner Universität (1894-1994),
veranstaltet an der Humboldt-Universität zu Berlin am 12. und 13. Mai 1995.
Frankfurt/Main: Lang, 141-162.
Kassühlke, Rudolf. 1976. "Deutsche Bibelübersetzungen des 20. Jahrhunderts". In: S. Meurer
(ed.) Der Bestseller ohne Leser. Evangelisches Bibelwerk. Stuttgart: Deutsche
Bibelstiftung, 168-171.
Schleiermacher, Friedrich Daniel Ernst. 1993. Hermeneutik und Kritik. Hrsg. und eingeleitet
von Manfred Frank. Frankfurt am Main: Suhrkamp (Taschenbuch Wissenschaft 211).
Schreiter, Jörg. 1988. Hermeneutik, Wahrheit und Verstehen. Berlin: Akademie-Verlag.
Vermeer, Hans J. 1994. "Hermeneutik und Übersetzung(swissenschaft)". In: TEXTconTEXT
9, 3/4, 161-182.
Bibles (quoted)
Czech
Bible svatá aneb všecka svatá Písma Starého i Nového Zákona. Podle posledního vydámí
Kralického z roku 1613. Česká Biblická Společnost, 1991.
English
New King James 1985: The Holy Bible. The New King James Version. Nashville; Camden;
New York: Thomas Nelson Publishers.
German
Arenhoevel/Dreissler/Vögtle 1965: Die Bibel. Die Heilige Schrift des Alten und Neuen Bundes.
(Herder-Bibel). Deutsche Ausgabe mit Erläuterungen der Jerusalemer Bibel. Hrsg.
von Diego Arenhoevel, Alfons Deissler, Anton Vögtle. Freiburg i.B./Basel/Vienna:
Herder.
Belloso 1990: Die neue Patmos Bibel. Erzählt von J. M. Belloso. Mit Bildern von Carme Solé
Vendrell. Deutsch von Hans Hoffmann. Düsseldorf: Patmos.
Bruns 1962: Das Alte Testament. Neu übertragen mit neuen Überschriften und Erklärungen
von Hans Bruns. Giessen/Basel: Brunnen.
Buber/Rosenzweig 1925f./1976: Die fünf Bücher der Weisung. Verdeutscht von Martin Buber
gemeinsam mit Franz Rosenzweig. Gerlingen: Deutsche Bibelgesellschaft. Verlag
Lambert Schneider.
Gute Nachricht 1991: Die Bibel in heutigem Deutsch. Die Gute Nachricht des Alten und Neuen
Testaments ohne die Spätschriften des Alten Testaments (Deuterokanonische
Schriften/Apokryphen). Stuttgart: Deutsche Bibelgesellschaft.
Hamp/Stenzel/Kürzinger 1992: Die Heilige Schrift des Alten und Neuen Testaments. Nach dem
Grundtext übersetzt und herausgegeben von Prof. Dr. Vinzenz Hamp, Prof. Dr.
Meinrad Stenzel, Prof. Dr. Josef Kürzinger. Augsburg: Pattloch.
Kaegi 1914: Die Bibel. Eine moderne Bearbeitung und Nachdichtung von Paul Kaegi. Der
Bibel erster Band: Israel und Juda. Munich: Delphin.
Kautzsch 1894: Die Heilige Schrift des Alten Testaments. Übersetzt und herausgegeben von E.
Kautzsch. Freiburg i.B./Leipzig: Akademische Verlagsbuchhandlung von J. C. B.
Mohr (Paul Siebeck).
Luther 1534/1983: Biblia/das ist/die gantze Heilige Schrijft Deudsch. Frankfurt am Main:
Röderberg. Faksimile-Ausgabe der ersten vollständigen Lutherbibel von 1534 in zwei
Bänden.
Luther 1912/1982: Die Bibel oder die ganze Heilige Schrift des Alten und Neuen Testaments.
Nach der deutschen Übersetzung Martin Luthers. Textfassung 1912. Stuttgart:
Deutsche Bibelgesellschaft.
Luther 1928: Die Bibel oder die ganze Heilige Schrift des Alten und Neuen Testaments nach
der deutschen Übersetzung D. Martin Luthers. Neu durchgesehen nach dem vom
Deutschen Evangelischen Kirchenausschuß genehmigten Text. Mit erklärenden
Anmerkungen. Stuttgart: Privileg. Württembergische Bibelanstalt.
Luther 1989: Die Bibel mit Erklärungen nach der Übersetzung Martin Luthers.
Berlin/Altenburg: Evangelische Haupt-Bibelgesellschaft.
Luther 1992: Stuttgarter Erklärungsbibel Die heilige Schrifl nach der Übersetzung Martin
Luthers. Mit Einführungen und Erklärungen. Stuttgart: Deutsche Bibelgesellschaft,
1992.
Lutz/Timm/Hirsch 1970; Das Buch der Bücher. Altes Testament: Einführungen, Texte,
Kommentare. Hrsg. von Hanns-Martin Lutz, Hermann Timm, Eike Christian Hirsch.
Munich: Piper.
Menge 1933: Die Heilige Schrifl Alten und Neuen Testaments. Übersetzt von D. Dr. Hermann
Menge. Handbibel. Stuttgart: Privileg. Württembergische Bibelanstalt.
Neue Welt 1971: Neue-Welt-Übersetzung der Heiligen Schrift. Übersetzung nach der
revidierten englischen Wiedergabe von 1970 unter getreuer Berücksichtigung der
hebräischen, aramäischen und griechischen Ursprache. New York: Watchtower Bible
and Tract Society/Vienna: Wachturm Bibel- und Traktat-Gesellschaft.
Procksch 1924: Die Genesis. Übersetzt und erklärt von D. Otto Procksch. (Kommentar zum
Alten Testament I). Leipzig/Erlangen: Deichert.
Rad 1949: Das erste Buch Mose. Übersetzt und erklärt von Gerhard von Rad. Berlin:
Evangelische Verlagsanstalt.
Reuß 1908: Die Bücher der Bibel. Hrsg. von F. Rahlwes. Band 1: Überlieferung und Gesetz.
Das Fünfbuch Mose und das Buch Josua. Nach der Übersetzung von [D. Eduard]
Reuß. Braunschweig: George Westermann.
Schlachter 1923: Die Heilige Schrift: Miniaturbibel. Nach dem Urtext und mit
Berücksichtigung der besten Übersetzungen hrsg. von Franz Eugen Schlachter. 17.
Aufl. bearbeitet von K. Linder und E. Kappeier. Stuttgart: Privileg. Württembergische
Bibelanstalt.
Simon 1976: Die Bibel oder Die Heilige Schrift des Alten und Neuen Bundes. Nach den
Grundtexten übersetzt und mit Überschriften und Erklärungen herausgegeben von
Ernst Simon Missionar i.R. Eigenverlag.
Weth 1992: Irmgard Weth: Neukirchener Kinder-Bibel. Mit Bildern von Kees de Kort.
Anhang: Einführung in die Bibel und ihre Geschichten. Neukirchen-Vluyn:
Kalenderverlag des Erziehungsvereins.
Zink 1966: Das Alte Testament. Ausgewählt, übertragen und in geschichtlicher Folge
angeordnet von Jörg Zink. Stuttgart/Berlin: Kreuz-Verlag.
Zürcher Bibel 1991: Die Heilige Schrift des Alten und des Neuen Testaments. Zurich: Verlag
der Zürcher Bibel.
Hungarian
Kämory 1870: Biblia. O és új testamentum. Az eredetiböl, heber, aram és görögböl forditotta
Kámory Samuel. Pest.
Károli 1590/1981 [Vizsolyi-biblia]: Szent Biblia az az Istennec ó es wy testamentvmanac
prophétác es apostoloc ältal meg iratott szent könyuei. Magyar nyelwre fordíttatott
egészlen és wijonnan, Az Istennec Magyar országban való Anya szent Egyházánac
epülésére. Visol (Vizsoly) [Faximilé 1981].
Kåroli 1842: Szent biblia azaz: Istennek ó és üj testamentomában foglaltatott egész Szent Irás.
Magyar ny el vre fordíttatott Károli Gáspár ál tal. Köszeg.
Károli 1991: Szent Biblia, azaz istennek ó és üj testamentomában foglaltatott egész Szent irás.
Magyar nyelvre forditotta Károli Gáspár. Budapest: Magyar Bibliatanács.
Polish
Pismo śwliete Starego i Nowego Testamentu w przekladzie z jezykow oryginalnych. Redaktor
odpowiedzialny Ks. Kazimierz Dynarski. Poznań-Warszawa: Wydawnictwo
Pallottinum 1980.
Russian
Tolkovaja Biblija Ui Kommentarij na vse Knigi Sv. Pisanija Vetchago i Novago Zaveta. St
Petersburg. Vol. 1: 1904-1907; vol. 2: 1908-1910, 1913; vol. 3: 1911-1913.
Slovak
Biblia. Písmo sväté Starej a Novej Zmluvy. Slovenská Biblická Spoločnost.' UBS 1991.
A matter of life and death: Gender stereotypes in
some modern Dutch Bible translations1
Anneke de Vries
The Bible has had a profound influence on Western civilization. Significantly,

however, it has been translations, not the original, which have been read and
studied. Even the Vulgate, which used to be considered a kind of source text,
actually is a (Latin) translation. This is why studying Bible translations is
relevant, not just for theologians and translation scholars, but for all those who
are interested in matters of culture and society.
In my research project, I have been investigating gender-specific elements
in recent Dutch translations of the Old Testament. Here I will present a few
ideas about the occurrence of male and female stereotypes in some recent
Dutch translations of one particular chapter: Genesis 27. More specifically, I
will discuss four concepts that can help to detect these stereotypes.
Stereotypes are cognitive structures that contain one's beliefs about groups
and their members. Gender stereotypes contain beliefs and expectations about
male and female character and behaviour. First, females and males are
believed to think and behave according to the stereotype; consequently, they
are expected to do so. Stereotypes influence our thinking and talking on the
subconscious level, and therefore can be found everywhere, and also in Bible
translations.
The translations under discussion are the following. Groot Nieuws (G) is
the counterpart of the English Good News for Modern Man. It was published
in the early eighties and was aimed at people with no religious background. It
is a fairly free translation, based on the translation principles as stated by Nida
& Taber (1974). The Start Bijbel (S) is a translation of parts of the Bible,
made for the young. It follows the source text rather closely, but uses simple
1
I wish to thank Arian Verheij for valuable comments on an earlier draft and for
improving my English.
language to facilitate understanding by children. The fifth edition was

published in 1994. The Willibrord Translation (W) is a Roman Catholic
Translation from the seventies.2 This translation is rather heterogeneous: some
parts are rendered fairly literally, others rather free. Het Boek (B), finally, is
not a translation but a paraphrase in easy Dutch. It originates from the
Evangelical Organisation 'Living Bibles'. It was published in the late eighties.
A remarkable characteristic of this version is that it was based mainly on a
number of other translations. See Jaakke & Tuinstra 1990 and Hollander 1994
for more information on these translations.
In all prefaces it is explicitly claimed that the translators have tried to
make translations that can be easily understood and are put in modern
language. This is relevant because it demonstrates that the translators had a
target-oriented approach, which makes the appearance of gender stereotypes
even more significant.
In this paper I will concentrate on gender-specific stereotypes to which the
source text would not seem to give rise. This is not to say that it does not
contain gender stereotypes. On the contrary, the source text came into being
in largely androcentric societies and as a result it does contain gender
stereotypes which often turn out to be negative for females and positive for
males. But these are the subject of other research (Brenner 1994-).
Genesis 27, a summary
The story begins with a father's words to his eldest and favourite son. Isaac,
old and blind, tells Esau that he might die at any time. He asks Esau to go
hunting and prepare a meal for him, that he may bless him before his death.
Rebekah, the wife and mother, overhears the father's request and his
promise. But she has received special information from God that her younger
son Jacob, and not Esau, should receive the eldest's rights and blessing, and
prepares to act accordingly. She plans to quickly cook a meal and to make
Jacob pretend he is Esau, so that Isaac will bless him instead of his brother.
Jacob hesitates, not out of ethical compunctions, but for fear that he might be
found out and be cursed rather than blessed. In the end, however, he does as
his mother has told him. The trick works and Jacob receives his father's
blessing.
2
A revised edition of the Willibrord Translation was published late 1995, which was too
late to be taken into account in this paper.
Anneke de Vries 315
When Esau finds out he is furious and plans to kill Jacob after Isaac's
death. Rebekah advises Jacob to flee the country, till Esau's fury is over. In
order to obtain Isaac's permission for him to leave, she tells Isaac that she is
disgusted with the women around them and does not want Jacob to marry any
of them.
Gender-specific elements
This is a story in which a woman, Rebekah, plays an important role, yet it is

not a woman's story. It depicts 'normal' family life and therefore it is ideal
for gender stereotypes to creep in. In this section I will discuss four concepts
that can help to detect gender-specific elements in the translations: topos,
speech, an argumentational connector, and focalization.
Topos
Anscombre and Ducrot have developed a theory of argumentation which says

that language does not in the first place give information, but primarily refers
to topoi. A topos is an underlying, generally accepted idea, shared by larger
groups in society. E.g.: 'A good bargain is a pickpurse'. Everyone 'knows'
in a way that very cheap objects are no good. According to Anscombre and
Ducrot, there are always counter topoi. A counter topos here could be that
paying three times as much for an object is overdone and not necessary. Which
topos is referred to depends on the words chosen (Anscombre & Ducrot
1986:88-89; Anscombre 1989:39). This concept is useful because with the help
of it the importance of the terminology used can be demonstrated and gender
stereotypes in the translations can be discovered.
This idea of language primarily referring to topoi is used here to interpret
the occurrence of concepts of life and death in the translations. It might be the
case that something like target culture topoi associating women with death and
men with life are underlying some remarkable and consistent differences
between the source text (ST) and the translations: where ST associates men
with death, translations tend to refer to life and conversely, where ST
associates women with life, we often find death in the translations.
In verse 2 Isaac, who is very old, says to Esau: lo' yada 'ti yom moti —
/ do not know the day of my death. Two of the translations remove the
reference to 'death' and mention 'life' instead. Het Boek gives a paraphrase,
also without mentioning explicitly Isaac's death:
(1) W: "I do not know how long I have to live" (Dutch: "leven")
S: "I will not live long any more" (Dutch: "leven")
B: "Each day can be my last" (Dutch: "laatste dag")
At the end of the story Esau discovers that Jacob has cheated him by 'stealing'
in a way his blessing. He then wants to kill Jacob, after his father's death. He
says to himself (verse 41): yiqrebu yemey 'ebel 'abi — the days of mourning
for my father are near. Here the concept of death is implied in the word
'mourning'. In the Start Bijbel this reference, not the word 'ebel — mourning
is translated. However, a reference to 'life' is added:
(2) S: "My father will not live long any more. When he has died ..."
(Dutch: "leven", "sterven" resp.)
Rebekah learns about Esau's plan, and she warns Jacob (verse 42), saying:
'esaw ... mitnaxem lexa lehorgexa — Esau is consoling himself (with the
thought of) killing you. In Het Boek, the concept of 'death' implied in
lehorgexa — killing you is eliminated and replaced by the explicit mention of
'life':
(3) B: "She told him that Esau was after his life" (Dutch: "leven")
So the association of men with death in the source text is reversed in the
translation into the association with life.
In verse 46 the concept of 'life' is used in relation to the only woman in
the story. Rebekah tells Jacob to flee in order to be safe from Esau's plan to
kill him. However, Isaac seems to be in charge and has to give his permission
for Jacob to leave. Rebekah, instead of telling the truth, pretends she wants
Jacob to go abroad to find himself a wife. In verse 46 she says to Isaac: qatsti
bexayay mipney benot xet — 1 am disgusted with my life because of the
daughters of Chet. Two translations significantly eliminate 'life':
(4) G: "I have an aversion to the daughters of Chet" (Dutch: "afkeer")
B: "I can not stand those girls" (Dutch: "uitstaan")
In the second part of the same verse the pattern becomes even clearer. Here
Rebekah says to Isaac: "if Jacob will marry one of them" lamah li xayim —
what [good] will life do me? Here we do not just find elimination of the
reference to 'life' in our translations, but instead explicit mention is made of
her death:
(5) G: "It will be my death" (Dutch: "dood")
B: "I'd rather die" (Dutch: "sterven")
Related to this gender-specific pattern is the translation of verse 4. Isaac is
Anneke de Vries 317
about to die, and before that he wants to bless his eldest son. He tells him to
catch game and make him the food he loves, ba 'abur tebarekxa naphshi —
that 1 may bless you. Surprisingly, the Willibrord Translation inserts 'strength'
(and does so again in the verses 10, 19, 25 and 31):
(6) W: "through it I will receive the strength to give you my blessing"
(Dutch: "kracht")
So, where the concept of 'death' is used in the source text in relation to men,
we regularly find translations containing references to the concept of 'life'.
Where 'life' is used in relation to the woman, we find translations without
'life' or a replacement by a reference to 'death'. And where a man talks about
his dying wish, the translation talks about 'strength'. The target culture topos
that all these translations may reflect is, in my view: wittingly or unwittingly,
women are associated with weakness and death, and men are associated with
life and strength.
Speech
The second concept distinguished here is 'speech'. Originally, it was assumed

that men's speech and women's speech differ (Lakoff 1975:57). According to
this view, men's speech and women's speech are two extremes on a conti-
nuum: men's speech would contain more instances of characteristic x and
women's speech would contain more instances of characteristic y. Lakoff's
ideas were based on intuition. Empirical research did not confirm or only
partly confirmed her ideas (Graddol & S wann 1989:83).
This type of research demonstrated that the idea that women's speech and
men's speech differ, is based upon presuppositions (Kramer 1977:157). These
presuppositions are normative ideas about characteristics of the speech of
certain groups, irrespective of whether these characteristics really occur or not.
All presuppositions together form a language attitude.
Brouwer, a Dutch sociolinguist, has demonstrated that women's speech is
considered to be "gentle, emotional, trivial and polite". Men's speech, on the
other hand, is supposed to be "demanding, boastful, dominating, loud and
authoritarian" (Brouwer 1987:213-214). Some of these presupposed
characteristics may have influenced the translators of Genesis 27. I limit myself
to the rendering of quotation verbs.
In verse 8 Rebekah has started to talk to Jacob about her plan to let him
receive Esau's blessing: shema 'beqoli la'asher' ani metsawah 'otax — Listen
to my voice, to what I command you. The verb of speech metsawah —
command has the connotation of authority. All four translations render it in a

stereotyped way, apparently inspired by the fact that it is a woman who is
quoted here:
(7) G: "Do what I ask you" (Dutch: "vragen")
W: "Listen to what I say to you" (Dutch: "zeggen")
B: "Rebekah advised Jacob" (Dutch: "raadgeven")
S: "Do what I say to you" (Dutch: "zeggen")
Here the strong woman, who is in charge of the situation and commands Jacob
to do certain things, is changed into a woman who takes a weaker stand.
In verse 13 we find a similar phenomenon. Here it is told that watomer
lo 'imo — and his mother said to him [Jacob], This completely neutral
sentence about a woman talking is translated in a female-stereotyped way:
(8) B: "But Rebekah eased his mind" (Dutch: "gerust stellen")
To 'ease minds' is considered to be specific female behaviour (Brouwer
1987:213-214). A man talking neutrally, on the other hand, is presented as
rebellious in one translation. This occurs when Rebekah has commanded
(source text) Jacob to go to the flocks and get her two goats, so that she may
cook a meal for Isaac. The source text goes on: wayomer ya'aqob 'el ribqah
'imo — and Jacob said to Rebekah, his mother. Het Boek has a male-
stereotyped translation:
(9) B: "But Jacob protested" (Dutch: "protesteren")
So the translation of quotation verbs demonstrates various occurrences of
gender stereotypes related to presuppositions about women's and men's speech.
Argumentational connector
Another idea developed by Anscombre and Ducrot is that many connectors

between sentences carry or suggest conclusions. In other words, these
connectors are not neutral means of linkage. They influence the interpretation
of the reader (Anscombre & Ducrot 1986; Anscombre 1989). In the sentence:
"we had a female professor, but we could not have been better off" the
connector 'but' carries the following conclusion or interpretation: a female
professor will evidently never be first choice. The use of 'but' creates a
contrast between the first and the second part of the sentence.
In Hebrew, by far the most frequent conjunction is wa-. This wa- can
imply contrast, but does not necessarily do so: the most obvious translation,
therefore, is 'and' or elimination.
Anneke de Vries 319
In verse 11 Rebekah has proposed that Jacob would go to his father

pretending to be Esau, in order to receive the blessing. The passage continues:
wayomer ya 'aqob' el ribqah — and Jacob said to Rebekah. In the source text
Jacob's reaction to Rebekah's proposal is connected to it by wa-. The
translations, however, have 'but':
(10) W: "But Jacob said to his mother" (Dutch: "maar")
G: "But Jacob said to his mother" (Dutch: "maar")
B: "But Jacob protested" (Dutch: "maar")
In the translations, the use of 'but' suggests that Jacob opposes his mother's
plan (and not just the way it will be carried out). The result is gender-specific:
Rebekah is depicted as the wicked woman, and Jacob as the good guy. Later
on in the translation it becomes clear that it was not the plan itself he did not
agree with, but the important thing is that, due to the use of 'but', the reader
has 'heard' the other suggestion already, which may sink in. So if one
translates by 'but' this is not 'wrong', but it brings about a specific meaning:
here it creates the gender-specific suggestion just mentioned.
Focalization
The fourth and last concept to be mentioned here is focalization. In a story,

there is always a narrator and a focalizer. The narrator, being an internal (one
of the participants) or an external (omniscient) authority, reports the story. The
focalizer focalizes the story, that is: provides the point of view from which the
events are seen. Very often narrator and focalizer coincide. This is the case in
our ST.
It is important for the reader to realize who actually focalizes the events,
because it makes a difference who does so: a young child and a psychothera-
pist, for example, both watching the same relational conflict between two
partners, will give very different reports of what was going on. This is because
the child has another point of view, other background information, other
possibilities for understanding what is going on, etc., than the therapist. If the
child is the focalizer, we will be given quite a different description of the
conflict from the one that the therapist will give.
Another reason for concentrating on the focalizer is that the reader usually
tends to watch the events from the point of view taken by the focalizer (Rim-
mon-Kenan 1983:71-85; Bal 1990:113-129).
The events reported in Genesis 27 are presented by an external omniscient
authority, the narrator, who is also the focalizer. The reader of the text views
the characters as does the narrator/focalizer. In the source text, this is also the
case in verse 46: watomer ribqah el yitsxaq — and Rebekah said to Isaac,
However, in the Willibrord Translation we come across a very specific
addition:
(11) W: "Once, Rebekah said to Isaac" (Dutch: "eens")
Through the addition of 'once', which marks discontinuity and the introduction
of a new episode, the translator/narrator momentarily lends his focalization to
Isaac. Isaac is the only character for whom this 'once' is suitable. He is the
only one who does not know that Esau wants to kill Jacob, so for him there is
no continuity with earlier events. For him, Rebekah's words come out of the
blue and mark a new episode. The translator has not stuck to the focalization
of the narrator here but has clearly identified with Isaac. Identification with the
woman in the story has not been found.
I have presented four concepts that can be used to detect gender-specific

translations: topos, speech, an argumentational connector and focalization. By
using them I have demonstrated that some Dutch translations of Genesis 27
contain gender-specific stereotypes to which the source text does not seem to
give rise, such as associating the male participants with life and strength and
the female participant with weakness and death, quoting the male and female
participants in stereotyped ways, pushing the female participant in a negative
role and the male in a positive, and lending focalization to a male participant.
References
Anscombre, Jean-Claude. 1989. "Théorie de l'argumentation, topoi, et structuration

discursive". Revue Québécoise de Linguistique 18 (1), 13-56.
Anscombre, Jean-Claude & Oswald Ducrot. 1986. "Argumentative et Informativité". In: M.
Meyer (ed.), 79-94.
Bal, Mieke. 19905. De theorie van vertellen en verhalen. Inleiding in de narratologie.
Muiderberg: Coutinho.
Brenner, Atalya(ed.). 1994-... The Feminist Companion to the Bible (I-..). Sheffield: Sheffield
Academic Press.
Brouwer, Dédé. 1987. "Language Attitudes and Sex Stereotypes". In: D. Brouwer & D. de
Haan (eds.), 212-224.
Brouwer, Dédé & Dorian de Haan (eds.) 1987. Women's Language, Socialization and Self-
image. Dordrecht: Foris.
Graddol, David & Joan Swann. 1989. Gender Voices. Oxford: Basil Blackwell.
Hollander, H.W. (ed.) 1994. Spectrum van bijbelvertalingen. Een gids. Zoetermeer:
Boekencentrum.
Jaakke, A.W.G. & Evert W. Tuinstra. 1990. Om een verstaanbare bijbel. Nederlandse
bijbelvertalingen na de Statenbijbel Haarlem: Nederlands Bijbelgenootschap/Brussel:
Belgisch Bijbelgenootschap.
Anneke de Vries 321
Kramer, Cheris. 1977. 'Perceptions of Female and Male Speech'. Language and Speech 20,
151-161.
Lakoff, Robin. 1975. Language and Woman's Place. New York: Harper & Row.
Meyer, Michel (ed.) 1986. De la Métaphysique à la Rhétorique. Bruxelles: l'Université de
Bruxelles.
Nida, Eugene A. & Charles R. Taber. 1974. The Theory and Practice of Translation. Brill:
Leiden.
Rimmon-Kenan, Schlomith. 1983. Narrative Fiction: Contemporary Poetics. London/New
York: Methuen.
Part III
Panel Discussions
Translation as intercultural communication —
Contact as conflict
Christina Schäffner/Beverly Adab
This workshop tested the validity of a hypothesis, namely the possible

existence of a hybrid text. This hypothesis was developed by the convenors
as follows:
Hypothesis (A): Hybrid texts are a feature of contemporary intercultural
communication. They result from cultures and languages being in contact. Such
contacts are initiated for differing communicative reasons: informative,
commercial, political or propagandist, proselytising or educational, entertaining
or thought-provoking. A hybrid text is provisionally defined as follows:
A hybrid text is a text that results from a translation process. It shows
features that somehow seem 'out of place'/'strange'/'unusual' for the
receiving culture, i.e. the target culture. These features, however, are not the
result of a lack of translational competence or examples of 'translationese', but
they are evidence of conscious and deliberate decisions by the translator.
Although the text is not yet fully established in the target culture (because it
does not conform to established norms and conventions), a hybrid text is
accepted in its target culture because it fulfils its intended purpose in the
communicative situation (at least for a certain time).
Assuming such texts do exist, hypothesis (A) poses several questions in
terms of contact as conflict, where conflict should be taken to refer to a
situation of change in the target culture requiring some reaction by this target
culture:
(1) Why do hybrids occur?
(la) Socio-political changes in a given culture create the need for new or
modified text types. A distinction can be made between text types for
which models already exist in the target culture, and text types which
are introduced into the target culture only through translation.
(lb) The increasing internationalisation of communication processes breaks

down text type areas.
We are not concerned here with text typologies (for example, Reiß's (1971)
typology which included an "audio-medialer Text" as a kind of blend, replaced
in Reiß/Vermeer (1984) as "multi-medialer Texttyp"). This system of
categorization was rejected by Mary Snell-Hornby (1988:31) because "the vast
majority of texts are in fact hybrid forms, multi-dimensional structures with a
blend of sometimes seemingly conflicting features [...] What is wrong is the
use of box-like categories as a kind of prescriptive grid, creating the illusion
of clear-cut objectivity. " Snell-Hornby argues for a "prototypology, a dynamic,
gestalt-like system of relationships, whereby the various headings represent an
idealised, prototypical focus and the grid-system gives way to blurred edges
and overlapping".
What we are concerned with is based on the fact that translation and
interpreting tie up with cultural, social and political realities. This is linked to
a convergence between cultures, a partly inevitable tendency. Also institutional
patterns of behaviour may occur in similar forms in different cultures as the
result of international strategies (in multinational companies, international
alliances for defence, trade, education, etc.). An example is the way
interlingual transfer operates in the European Communities (cf. Dollerup
1996). Translators at the Commission of the European Communities work with
different types of source texts. Hybrids may occur with common
communications (e.g., 'circulars to staff) and also with legal documents.
Another example would be advertising, which also helps to create a new
global culture (cf. Cook 1992). Advertisements often rely on knowledge,
recognition and acceptance of social conventions and/or taboos. For example,
the French version of the advertisement for Qualcast lawn mowers assumes
that target readers share the British consumer's desire for a well-kept lawn
with neat stripes or criss-cross patterns. Does this reflect accurate marketing
research or is the translated text a hybrid of social values as well as of
linguistic patterns?
(2) Who is responsible for their creation?
(2a) The stimulus may come from society as a whole, a specific sub-
group, an individual (including the translator). Translated texts
maintain a primary position when a society or a specific discursive
practice is young, in the process of being established, or is
experiencing a crisis or turning point (e.g. societies after the collapse
of Communism).
Christina Schäffner/Beverly Adab 327
(2b) Hybrid texts are also created in a multi- or supra-cultural

environment.
Cultural boundaries change due to social and political integration and
internationalisation processes. Multilingual texts that are created in a process
of multinational and multilingual negotiation and which are equally valid (e.g.,
EU documents) are evidence of such (overt or hidden) internationalisation
processes. Such texts are usually created simultaneously in the working
languages of international bodies, and subsequently translated. In this case,
even the source text could be called a hybrid text. In explaining such
translation practices, the notion of the source text would need to be
reconsidered, and also an explanatory model need not start from a source
culture (cf. Pym 1993:81).
(3) What are their identifying features?
(3a) A new text type is created in the target culture by adopting some or
all of the features of a text type in the source culture. An example
would be annual reports of companies in Malaysia which have been
modelled on the English reports. Venuti's (1994) foreignising
strategies would apply here.
(3b) Hybrids reflect specific textual features (vocabulary, syntax, style,
etc.) which may clash with target language conventions. Cultures not
only express ideas differently, they shape concepts and texts
differently (Jakobsen 1993:158). Hybrid texts have features that are
somehow contradictory to the norms of the target language and
culture.
Again, EU texts are an example of this. In the process of establishing political
unity, linguistic expressions are levelled to a common, (low) denominator.
Eurotexts reflect a Eurojargon, i.e. a reduced vocabulary, meanings that tend
to be universal, reduced inventory of grammatical forms (cf. Schütte 1993;
Pym 1993). When EU legal texts are translated and subsequently become
embedded in national legislation, they are formative elements in language
change in the national languages of the member states.
(4) What factors influence their reception in the target society?
(4a) Acceptance is due to the limited communicative functions of the texts.
EU texts, for example, function within the Community within which
they are created (e.g., for the staff, or for meetings of the respective
bodies). This means that there are clearly defined user needs. The
multinational EU institutions as such are the target culture, hybrid
texts are formative elements in creating a (truly) supranational
culture. What counts as knowledge or norms is determined by the

respective internal discourse practices.
(4b) There may be a hidden political or ideological agenda governing
acceptance and institutionalisation. When, for example, EU texts are
translated for use in the member states, different criteria for
acceptance may apply. The receiving culture may adopt a defective
attitude (enriching its own culture with foreign elements), a
transcultural attitude (a kind of carelessness in adopting Target
Language (TL) elements in the Source Language (SL)) or a defensive
attitude (rejecting the "otherness" and doing everything to stop it) (cf.
Robyns 1994).
Acceptance is in general due to the fact that all cultures are open and adaptive
systems. Once a text is accepted, it is no longer a hybrid text. Thus, hybrid
texts are a transitional and historical phenomenon. Many cultures have received
their first written documents through translation (for example, Bible
translations gave many languages their first written form; or commercial
contacts with the outside world helped developing countries to create functional
text types which were borrowed from the chief vehicular languages of traders
or colonial powers).
Hybrid texts allow the introduction into a target culture of hitherto
unknown and/or socially unacceptable/unaccepted concepts through a medium
which, by its non-conformity to social/stylistic conventions and norms,
proclaims the otherness of its origin and thereby legitimises its right to be
heard. There is freedom of expression which is unhindered by said
conventions.
Discussion was invited on the following questions: Is the notion of a
hybrid text a useful one to explain texts that are produced in the translational
reality? Where do hybrid texts fit into the spectrum of categorisation of
translated texts? Or can the specific functional and textual features be
accounted for within existing models of translation (e.g. functional approaches,
such as Skopos theory)? If yes, then hypothesis (B) would apply: There is no
such thing as a hybrid text.
Four invited panellists each reacted to the concepts raised in the discussion
paper, speaking from their relative research interests and perspectives:
Anna Trosborg looked at the role of cultural norms and genres in
translation as mediation between two cultures. The concept of culture as a
totality of knowledge, proficiency and perception is fundamental to any
approach to translation. If translation is defined as source text induced text
production, translation into a foreign language will always be an instance of
Christina Schäffner/Beverly Adah 329
intercultural communication. The translator will have to bridge the gap, small
or large, between two cultures. Culture is to be understood not in the narrower
sense of man's advanced intellectual development as reflected in the arts, but
in the broader anthropological sense of all socially conditioned aspects of
human life.
Culture has thus to do with common factual knowledge, usually including
political institutions, education, history and current affairs. The problem for
the translator is how to comply with cultural norms, i.e. to decide which
norms take priority, whether the cultural norms of the SL community as
reflected in genre conventions, the cultural norms of the TL community, or
perhaps a combination of the two, a compromise between two or more
cultures? The choice of cultural strategy may result in source-culture bound
translation (the translation stays within the SL culture), target-culture bound
translation (the translation stays within the TL culture) or in a hybrid, where
the translation is a product of a compromise between two or more cultures.
International texts are usually the product of the dominant culture. Factors
likely to influence the choice of cultural strategy are:
— the text type: some texts, such as political speeches and legal documents,
will be culture-bound. Some text genres are more likely to develop
international norms than others, e.g technical and scientific texts are
representative of technology in international fields.
— the purpose of translation: texts may be translated for different functions:
perhaps metatextual, with the function of reporting exactly what is
conveyed in a particular text (e.g. political speech); alternatively, the text
may function as a text in its own right, as in the case of advertisements.
— the status of the ST or ST author may require loyalty to the ST: some
texts are representative of a dominant culture to be conveyed in the source
culture or written by a prestigious author whose idiolect has to be retained
in the translation.
— the evolution of existing genres: a genre may change over time, for
example in order to adapt to dominant norms. Scientific research is an
example of a genre which has progressed over time so as to conform to
Western norms and conventions.
— the creation of new genres: the formulation of standardised treaties may
involve the creation of a hybrid text.
Hybrid texts are produced through intercultural negotiation, as well as through
translation. They come into existence as a compromise between various
cultures. Thus they are not hybrids in the sense that they are a mix of various
text types in the rhetorical sense of narration, description, exposition,
argumentation, expressive or directive (appellative) communicative features.
Instead, hybrids are arrived at as an outcome of negotiations between different
languages and cultures and may involve features which are contradictory to TL
and target culture norms.
In contrast to culture-specific political texts, there are other political texts
which are interactively negotiated in a supra-national setting for the purpose
of achieving and reflecting consensus. Such texts involve standardised
formulations of treaties and legal documents (e.g. for Nato and the EU).
Contracts and treaties, as a genre, display special conventions. Thus
translations into different languages represent different versions intended to
fulfil an identical function and to pursue identical political aims in the
respective target cultures with regard to their addressees, who are expected to
possess almost identical background knowledge. For example, EU documents
have developed a specific Eurocrat language for the purpose of Community
negotiations, which may include the creation of new concepts as well as new
terms. The result is a jargon, known to EU staff, translators and interpreters.
In this context the ST could be described as a pseudo-text, which does not in
itself fulfil a communicative function, since there is no single primary
communicative situation but several parallel ones.
A hybrid text is a recognisable response to the exigencies of the situation.
A sociocognitive theory of genre developed by Berkenkotter and Huckin
(1995:4-24) comprises a theoretical framework consisting of five principles,
which are highly relevant to the concept of the hybrid text: dynamism,
situatedness, form and context, duality of structure, community ownership.
Like genres, hybrid texts may change over time in response to the
sociocognitive needs of the users. They are a form of "situated cognition",
deriving from participation in the communicative activities of professional life
in a particular setting. Both form and content are adjusted to what is
appropriate to purpose and context. Through the process of writing and
translation, hybrid texts are both constitutive of social structure and generative
in their professional, institutional and organisational contexts of production.
Such texts are a product of community, representing several languages and
cultures.
Sonja Tirkkonen-Condit discussed the definition of the hybrid and the
process of hybridisation, basing her observations on a Finnish research project.
Intercultural communication gives rise to the development of new text types
and genres and particular stages of this development can be described as
hybridisation. These are the stages at which the new text types and genres have
not yet fully established themselves as forms of communication in a
Christina Schäffher/Beverly Adab 331
sociocultural setting: they manifest linguistic and rhetorical features which are
felt to be foreign. Hybridisation can be seen as a process comparable to
pidginisation: while pidginisation in the course of time may result in the
emergence of new languages, i.e. creoles, hybridisation may result in the
emergence of new domestic text types and genres.
Different rhetorical norms can clash, as can be seen in examples drawn
from EU project proposal texts compiled in English by Finnish applicants.
Project proposals are submitted to the EU Commission by individual applicants
or by international consortia. The Commission appoints a team of evaluators
to select those projects to which the Commission should grant money. Thus a
proposal should be written in a style which convinces the evaluators of the
viability of the project. This is where two different rhetorical norms may clash.
The rhetorical norm governing the proposals written in English is close to the
one prevailing in Anglo-American scientific rhetoric, especially as regards
grant applications. In these, the style is assertive and straight to the point. It
does not hide the merits of the applicants. The text is reader-friendly in that
it uses metatext and other structural signals to guide the reader. The Finnish
rhetorical tradition is different. It is more implicit and impersonal. It starts
from a background and tends to leave it to the reader to infer the aims of the
project as well as the merits of the researchers. Praising oneself is felt to be
impolite, and metatext is frowned upon as a sign of underestimating the
reader's intelligence. The 'point' of the text tends to be left towards the end
of the text. Thus a Finnish applicant or a Finnish translator who is not aware
of the rhetorical difference may end up producing an English text which is
grammatically correct but rhetorically deviant.
The blurring of target cultures is another feature of hybridisation. If the
notion of target culture is defined as the receiving culture in which the new
text types and genres ultimately develop, there are in fact two target cultures
present. Both the Finnish scientific community and the EU scientific
community are in a sense target cultures. One can imagine that Euro-rhetoric
absorbs rhetorical, lexical and even grammatical features from the various
linguistic communities which participate in its functions. Thus the EU can also
be regarded as a target culture. The other target culture is the Finnish
academic community, which must gradually learn to write its scientific prose
according to the Anglo-American rhetorical tradition, or, perhaps, according
to a shared Euro-English rhetorical tradition. This learning may spread to the
Finnish-language academic rhetoric as well. It is also possible that the Finnish
scientific community will continue to maintain a dual rhetorical norm: one for
international communication and another for national communication.
Euro-rhetoric can also be seen as an example of the hybrid and in some

instances existing EU texts can be described as hybrids, for example, in the
instructions for applicants published by the EU Commission. The following
extract is from the guidelines for applicants, viz. from the instructions on how
to write a summary:
Give a brief description of the industrial and technical objectives and the
proposed approach to meet them. Please note that the summary is a very
important part of your proposal and should be completed carefully. It should give
a quick reference to the areas covered by the proposal, a concise description of
the proposal content and could be used for publication purposes. It should
therefore be non-confidential and should be completed by the proposal co-
ordinator, preferably in English.
The last two sentences of this passage manifest a syntactic and stylistic
anomaly which suggests that the text itself may be a translation.
Analysis of the EU project proposals indicates that English-language texts
produced by Finns represent a hybrid which vacillates between three varieties:
the Finnish rhetorical norm, the prototypical target norm (= Anglo-American
scientific rhetoric), and the hybrid target (= EU) norm. Three parallel source
norms coexist, the national prototype, the Anglo-American prototype, and the
Euro-rhetorical hybrid. The natural sciences are more international in scope
and tend to use the second type, whereas the more national based sciences
(such as Finnish history and education research) still tend to stick to the first.
The concept of the hybrid text seems useful as a tool for describing the
transition stages in which many discourse norms interact and coexist as a result
of growing internationalisation.
Candace Séguinot addressed the concept from both intercultural and
intracultural perspectives, taking examples from the world of advertising,
adopting a functional and pragmatic, rather than textlinguistic, standpoint.
Translation is part of a continuum, and as such forms part of a wider
discipline. For example, in the case of advertising, translation falls within the
discipline of marketing, and the translator has a duty to be aware of all social,
psychological and legal aspects relating to the production of a marketing text.
He or she also has a duty to understand the mechanisms of marketing and to
acquire sufficient subject knowledge to enhance the effectiveness of his/her
contribution to the overall marketing performance as a communicative act. In
this perspective, there is not really a dichotomy between a minor and a major
participant, between a powerful coloniser and a colonised society. A useful
concept from marketing is the identification of "same, other and wanna-be".
In other words, text users need to perceive what is being presented as
representing a desirable self-image, one which it is desirable to project,

regardless of where it came from. In the case of advertising, some mini-
cultures within the same language culture will react positively to certain
concepts, such as the housewife using certain kinds of washing powder or
bathroom cleaner, and other sub-sectors of that same culture will not be at all
convinced (Séguinot 1994).
In any discussion of translation there is at least a 3-way identification
process going on, from source culture to target culture, via the translator and
his or her perception of what segment of the target culture will correspond to
the target segment of the source culture. Culturally appropriate advertising
material for one socio-linguistic community may require minor or even
extensive adaptation for a new target community through the translation
process and the mediation of the translator. For example, the Danish version
of the television advertisement for Carlsberg beer, featuring Brigitte Nielsen,
was banned in the UK for its overtly explicit sexual overtones. There is need
for some debate about the difference between translation and adaptation and the
circumstances or production situations in which each should be the most
appropriate technique. Conformity to culture-specific norms may involve the
modification of information content and of manner of presentation, in order to
harmonise the manner of representation of the production according to target
culture conventions of interpretation, for visual and written forms.
Achieving equivalent effect or communicative function, that of incentive
to purchase through attraction and persuasion, may necessitate a re-positioning
of the product in the new market, a process of adaptation, not simply or solely
one of translation. For example, Horlicks is sold as a relaxing night-time drink
in the UK but as an energy-giving morning drink in Thailand. It is, however,
arguable whether such modifications constitute a process of hybridisation or,
as stated, one of adaptation in order to fulfil a similar communicative function
within the new target community.
Translation can also occur on an intralingual basis, when texts are adapted
for different sectors of the same language community. For example, a product
may be sold as a useful aid to digestion to the elderly but be perceived by the
(mainly female) teenage market as an aid to dieting. The rapid spread of
satellite communications has resulted in the development of global market
segments which cross national community boundaries. Séguinot argued that the
study of hybridisation, in terms of the process of translation, should also take
into account the circumstances of production, the characteristics of the target
culture and the intended communicative function.
334 Translation asInterculturalCommunication
leva Zauberga gave examples from Latvia's current climate of political

and cultural transition to illustrate how hybrids can be welcomed and serve a
useful purpose in the target community. Hybrid texts are inevitable features of
contemporary intercultural communication and they should not necessarily be
treated as the product of linguistic and cultural interference but rather as a
natural consequence of crossing cultural barriers.
In a sense, all translations are hybrids since they can be viewed as a
transplant of the source text into an alien, target culture environment. As
pointed out by Levy, in the process of translation the form-content unity of the
ST is disrupted. Inevitably some pressure is exerted upon the TL as the
transfer of foreign elements is impossible without a certain "violence". Levy
(1974:83) claims that the translator's style always bears the imprint of the ST.
The influence may be direct and obvious (e.g. transcribed neologisms), but
often it is indirect. While the ST emerges as a homogeneous entity, the TT
needs to be adjusted to the long-standing TL system and cultural context,
which necessitates the implementation of linguistic compromise in the
translated text. Consequently a translated text can easily be recognised by
words, word combinations and structures that are semantically and
grammatically correct but seem artificial. Levy compares the translator to the
actor who, if not highly professional and talented, reproduces clichés which he
calls "translator's jargon" (Levy, 1974:153). The degree of artificiality depends
on the translator's competence as well as approach to the ST, e.g. the
translator's intention to remain faithful to the original or the prioritising of the
readability and acceptability of the TT in the new cultural context.
Levy spoke of hybrid texts prior to the general move towards a target-
culture oriented approach and the redefinition of the translation process as
rewriting, but similar ideas have been conveyed in the light of modern
theories. Duff calls the language of translations the "third language" which
lies, as it were, in between the source and target languages: all words are
known but put together in an unfamiliar way (Duff 1981:122). Korzeniowska
and Kuhiwczak (1994:112) apply the term "hybrid" to translations that
manifest signs of inconsistency of some kind; texts with contradictory stylistic
features, undefined readership and blurred intonation. Toury (1995:28) also
admits that translated texts tend to have features that render them distinct from
non-translated texts. A TT can never be fully adequate or fully acceptable; it
tends to deviate from sanctioned patterns. However, he claims that these
deviations are not necessarily "mere production mishaps". More than one
writer has observed that the identification of a text as a translation "protects"
the reader from misinterpreting the writer's intention.
In Latvian, faithfulness to the ST as an approach that enhances cultural

and linguistic differences still remains the dominant strategy. The closer the
translation to the original, the more vivid the ST imprint. Accordingly, the
majority of Latvian translations contain a great number of hybrid features
which, however, does not mean that they infringe the TL cultural norms. The
general level of acceptance of foreignising strategies has been historically and
socially conditioned, for the following reasons:
— Within the Latvian literary polysystem translated literature has tended to
dominate. Consequently the ST tends to be taken as a model and
translations aim to imitate it.
— The Latvian cultural scene has often been perceived as defective, the
Latvian language designated inferior (by invaders of different origins);
national self-assertion has therefore been one of the major functions of
Latvian translations. Translators have, from very early in the 19th century
until the present, felt obliged to prove that the concepts as those expressed
in "major" languages can also be expressed in Latvian.
— In the current climate of radical political and economic transformation, an
unconditional acceptance of Western mass culture is taking place. This has
led to extensive borrowing of both linguistic units and cultural patterns.
Consequently, source lexical units are being transcribed with growing
frequency and behavioural patterns are directly transferred: this can be
seen in the increased use of four-letter words which had previously been
ousted from the literary language. Traditional forms of expression and
social conventions have been altered, leading to further internalisation and
to a liberalisation of the environment.
From the above, the conclusion can be drawn that the degree of hybridisation
and attitudes to hybrid texts depend on the concrete cultural situation and on
the status of translations in that society.
Key points arising from the general discussion concerned more specific
aspects as well as additional features :
1) The concept of the hybrid text is not new. The question is rather what can
be learned from such a concept and how this can contribute to Translation
Studies. Any research into this question should be systematic and clearly
defined. The concept has to be studied in the context of Translation Studies,
in terms of what the concept may contribute to translation theory but also what
it may offer of relevance for the work of the translator, at the analysis stage,
at the processing stage and at the stage of synthesis to recreate effect.
2) Translators have to watch for and be conscious of any attempt to

superimpose foreign or universal values (linguistic and cultural) upon a local
culture, to modify existing discourses or to introduce new genres of discourse.
However, decisions relating to choices and the imposition of the values of the
"other" may be the result of power struggles beyond the control of the
translator.
Increasing internationalisation of marketing and other media forms of
communication may also be responsible for a tendency on the part of
translators or translation clients to stop transferring concepts and simply
borrow or transfer them wholesale into the target text. However, this tendency
is often counterbalanced by an attempt to localise imported concepts or forms
in order to facilitate their integration into and acceptance by the target culture.
Another important point to consider is the way in which the hybrid may
be a deliberate creation of content or form, or both, designed to provoke
change in the target culture's linguistic or cultural system.
3) The hybrid must also be seen in its historical context, since there may be
times in the history of a culture when people are more open to hybrid texts.
Hybrids allow the translator to expose the weaknesses in his/her own language
and at the same time to expose weaknesses in the original language, so that the
text becomes a site of conflict, not just of concepts but also of concepts within
and across cultures. In other cultures the "foreigness" of a translation may be
appreciated, especially those whose members have reason to reject existing
values because these have previously been imposed on them by yet another
foreign culture. Some may claim that historically, a hybrid will always be a
hybrid even if people who use it are unaware of its etymology. The users of
the hybrid may have a deep-rooted, hostile reaction to such texts — something
arising out of political and national agendas, not just linguistic issues. Others
believe that once the hybrid enters the target culture it is accepted because it
fulfils a particular communicative situation or need and at this stage it is no
longer a hybrid but a new text type. Reactions to a text depend to a large
extent on whether it is seen by the receiving culture as being externally
imposed or internally desirable, as much as on whether or not this culture can
actually identify with the hybrid as representing a potential reality .
4) It is also essential to study the process of hybridisation, not just the
result. The main objective should be to develop a more precise description and
explanation of the nature of the phenomenon, its form, origin and adaptability
to further influences. Norms can thus be established and the evolution of the
process studied so that the life-span of the hybrid can be defined. Prior to
Christina Schäffner/Beverly Adab 337
entering the target culture it does not exist except in the mind of the translator.
5) The context of production is also important. A source text may itself be a
hybrid construct resulting from collaboration between members of different
cultures (e.g. EU documents).
In conclusion: it would be valuable to study hybrid texts as defined here,
because such texts are identifiable examples of translation strategies. They
could thus offer a useful perspective from which to study translation strategies
and also reception research. A good starting point might be to evaluate
reactions to hybrid texts and the strategies involved in their production, as
opposed to other ideologies. It is intended to develop the concept of the hybrid
text and the process of hybridisation in a forthcoming publication.
References:
Berkenkotter, Carol/Huckin, Thomas N. 1995. Genre Knowledge in Disciplinary

Communication, Cognition, Culture, Power. Hillsdale, New Jersey: Lawrence Erlbaum
Associates.
Cook, Guy. 1992. The Discourse of Advertising, Routledge: London.
Dollerup, Cay. 1996. "Language Work at the European Union". In: M. Gaddis Rose (ed.)
Translation Horizons Beyond the Boundaries of Translation Spectrum. Binghamton: State
University of New York, 297-314.
Duff, Alan. 1981. The Third Language. Oxford: Pergamon.
Jakobsen, Arnt Lykke. 1993. "Translation as Textual (Re)Production". Perspectives: Studies
in Translatology 2, 155-165.
Korzeniowska Aniela/Kuhiwczak Piotr. 1994. Successful Polish-English Translation. Warsaw:
Wydawnictwo Naukowe Pwn.
Levy Jiri. 1974. Iskustvo perevoda. Moscow: Progress.
Pym, Anthony. 1993. Epistemological Problems in Translation and its Teaching. A seminar
for thinking students. Calaceit: Caminade.
Reiss, Katharina. 1971. Möglichkeiten und Grenzen der Übersetzungskritik. Kategorien und
Kriterien für eine sachgerechte Beurteilung von Übersetzungen. München: Hueber.
Reiss, Katharina & Vermeer Hans. 1984. Grundlegung einer allgemeinen Translationstheorie.
Robyns, Clem. 1994. "Translation and Discursive Identity". In: C. Robyns (ed) Translation
and the (Re)Production of Culture, Leuven: The CERA Chair for Translation,
Communication and Cultures, 57-81.
Schütte, Wolfgang 1993. "'Eurotexte' — Zur Entstehung von Rechtstexten unter den
Mehrsprachigkeitsbedingungen der Brüsseler EG-Institutionen". In: J. Born/G. Stickel
(eds.) Deutsch als Verkehrssprache in Europa (Jahrbuch /Institut für Deutsche Sprache;
1992). Berlin: de Gruyter, 88-113.
Séguinot Candace. 1994. "Translation and Advertising: Going Global". In: C. Schäfmer/H.
Kelly-Holmes, (eds) Cultural Functions of Translation, (Contemporary Issues in
Language and Society, vol.1, no 3). Clevedon: Multilingual Matters, 249-265.
Benjamins.
Toury Gideon. 1995. Descriptive Translation Studies and Beyond. Amsterdam: Benjamins.
EST Focus: Report on research training issues
Daniel Gile/José Lambert/Mary Snell-Hornby
EST was set up to help foster translation research, which means inter alia
helping to increase the volume of translation research (but not necessarily
increasing the volume of publications), and helping to improve research
quality, in particular by increasing the proportion of good research in the total
mass of translation studies.
In most established academic disciplines, research has the following
attributes:
— It is done by "research professionals", i.e. academics and full-time
researchers.
— Research productivity is a function of professional requirements, career
requirements and personal ambition, of the researchers' interest in their
subject and their problem-solving drive, as well as the intellectual pleasure
and social recognition they derive out of research.
— Research quality is a function of training, knowhow, research norms and
quality control. The latter comes from teachers and advisers during
training, and from "clients", supervisors, editorial committees and
selection during the subsequent research career. Quality is also fostered
by competition over positions, over publication space in journals and
elsewhere, and sometimes over money.
In the field of translation studies, there are very few academic translation
departments (although there are many 'vocational' training schools and many
translation courses in foreign language and literature departments). Only a
small proportion of the authors in the literature of Translation Studies are
academics. Most of them are Translation teachers and practitioners. For them,
research is not a professional requirement, nor a career requirement. Neither
are they highly available for such an academic enterprise, as their time is
thinly spread over their professional, training and professional life (see for
instance, as regards interpreting, Cenkova 1995; Gile 1995 and Strolz 1995).
Moreover, the vast majority of such authors have not had the benefit of formal
research training. Neither is there a research tradition per se in Translation
Studies as a whole, though in some of its sectors, including the literary, the
historical and the philosophical areas, previous training in the mother discipline
has carried over to produce some very solid work (as can be seen in particular,
albeit not exclusively, in the work done by Belgian scholars in and around
Leuven). Another point is that there is little quality control in most translation
publications, including journals and conference proceedings, and the
publication market being far from saturated, Translation Studies do not have
the benefit of the healthier side of competition. As a result, on the whole:
— The field has comparatively little knowledge and knowhow.
— Research careers frequently stop after an initial project, often an M.A. or
a PhD, which may be subsequently aired from time to time into a
conference or journal with a fresh layer of make-up.
— Many authors are blissfully ignorant of the rule that researchers should
make a serious effort to review existing studies in and around their topic
before embarking upon their own exploratory endeavours, and thus
experience for themselves the joy of creation while submitting their
readers to endless repetitions.
— A high proportion of studies conducted in the field are methodologically
weak (Toury 1991; Gile 1995).
Possible action on institutional environments could include setting up academic
Translation departments in universities. Another possibility would be to bring
translation and interpretation schools closer to the norms of academia, in
particular by creating academic requirements such as graduation theses in such
schools, which has proved rather efficient in Italy (see Gran & Viezzi 1995)
and in Scandinavian countries, setting academic standards for recruitment of
translation and interpretation trainers, and offering research training to
students. Such actions are by no means easy to implement or necessarily
legitimate, for the matter, if only because of the essentially vocational and
practical orientation of translation and interpretation schools. Another channel
is that of specialized summer programmes or other courses held outside the
institutional framework of individual schools of universities, as explained by
José Lambert below.
Possible action on individuals would seek to strengthen motivation, in
particular by stimulating the curiosity of translation and interpretation students
and practitioners, which requires solid, readable and interesting research. Once
motivation is awakened, an appropriate human environment is required to
maintain it. This is where communication, and in particular international
Daniel Gile/José Lambert/Mary Snell-Hornby 341
contacts, come in. This is also where the role of research supervisors, be they
direct institutional advisors or indirect co-advisors, becomes important.
In this context, EST can serve as a forum for communication, and as a
(provisional, it is hoped) second-best substitute for academic translation and
interpretation departments. Through its meetings, newsletter and membership,
it offers a wide range of resources. By calling on its resource persons, EST
could also help provide guidelines on research training, thesis supervision,
research methodology and research organization. More direct contributions
could include the organization and/or sponsoring of actual research courses and
facilitating meetings between young researchers and more experienced
colleagues who could help them as co-supervisors.
This EST focus was meant to be the first of a series of working sessions
devoted specifically to research methodology, policy and training issues. In this
first meeting, presentations focused on training. A short session with three
speakers can only scratch the surface of the very intricate subject matter.
Rather than providing a full description or analysis of the phenomena under
consideration, the following reports only aim at helping to start a discussion
on the relevant issues. Indeed, the speakers, having kindly restricted the
duration of their presentation to a few minutes, gave the audience ample time
to discuss such issues. The lively discussion which followed the panelists'
presentation will not be summarized here. However, when debates on the
subject become more systematic and focused, they should certainly deserve
significant publication space, possibly even as separate publications.
In the following sections of this report, two leading Translation scholars
discuss their experience and share their knowledge. José Lambert has been the
central personality and driving force behind the CE(T)RA summer training
programme in Leuven, Belgium, and is now pushing forward the idea of
distance training, thus taking full advantage of the most recent communication
technology. His report provides some information on the training programme,
which is organized internationally and inter-institutionally. Mary Snell-Hornby,
EST President and former Head of the translation and interpretation school of
the University of Vienna, explains the institutional structure of the University
system in Austria, emphasizing aspects of supervision at the various levels and
discreetly hinting at some fundamental problems.
The CE(T)RA programme
In 1989, a group of Translation scholars decided to set up a training center

for (young) scholars who intended to go into translation research. The roots of
this initiative can be found in an interuniversity Contact Group spanning the
"Low Countries", which was created in the 1980's under the auspices of the
Belgian Research Foundation. The group itself crystallized after the 1976
Literature and Translation symposium, held at the Catholic University of
Leuven (KUL), during which much emphasis had been put on the need to
systematize translation research and to integrate it into the academic world.
Scholars with an international record such as Holmes, van den Broeck,
Lefevere (the first President of the Group), and Lambert (the first Secretary
of the Group), felt the need to train their younger colleagues and successors
such as Hermans (the Group's second President), D'hulst, van Leuven-Zwart,
Naaijkens, Schoneveld, Delabastita (second Secretary), Korpel and many
others, and then to lead them to international channels and into scholarships.
The Contact Group was also used as a forum for larger international translation
initiatives. Members of the Group participated in international initiatives and
promoted collective research and publication beyond departmental and
institutional borders. One of the tangible results of this group's work was a
series of books on the history of translation in the Low Countries: the Sticking
Bibliographica Neerlandica published a bibliography of translation research in
Dutch culture, and a series of books on the history of Dutch translation have
also been published by the same group on other occasions. However, as
indicated by Hermans in an important article (Hermans 1996), this was not
sufficient to obtain recognition from official Belgian and Dutch institutions.
The CERA Chair for Translation, Communication and Cultures can be
considered one outcome of the development of the group's activity. It was
created within the "Penn-Leuven Institutre for Literary and Cultural Studies",
an intercontinental institute for doctoral research training (1987-1989), at the
same time Target (John Benjamins Publishing Company) was born. CERA, a
leading Belgian bank, offered its generous support as a way into larger
consortia. The idea was to attract and select young scholars and/or PhD
students worldwide, to discuss their projects while offering them
methodological advice and thus promoting them and preparing them for
international competition. Beside the CERA Professor, a staff of regular
supervisors and visiting scholars were asked to guide the young talents during
4 heavy weeks of intensive training in a typical university town, which in the
summer-time, is left to the visitors and tourists, and only books and seminar
rooms remain as signs of academic life. The session includes lectures by the
CERA Professor, who is selected every year for one summer session among
leading scholars in the field of Translation Studies, lectures and seminars with
the supervisors and visiting scholars, who also provide personal tuition to
trainees during their stay, and presentation by trainees of their own ongoing
projects. The programme includes lectures and seminars about research
methodology as applied to Translation research and about theoretical and other
issues in the field of Translation Studies. Besides its highly international and
interdisciplinary nature, the programme is unique in its very high ratio of
lecturer/supervisors to trainees, which often approximates one-to-one, giving
trainees the opportunity to consult personally some of the leading scholars in
the field.
The newly founded "research summer school" soon acquired worldwide
reputation, in particular with its impressive list of CERA Professors: Toury in
1989, Vermeer in 1990, Bassnett in 1991, Neubert in 1992, Gile in 1993,
Snell-Hornby in 1994 and Lefevere in 1995, and with its equally impressive
list of participants (more than 150 persons from 5 continents) and visitors,
including alumni of the programme, many of whom like to come back for
more. As a matter of fact, two former trainees of the programme have edited
the special series of CERA Papers (Robyns 1993; Jansen 1996). Some of them
already have a significant scholarly record, and though the CERA papers and
volumes are not widely distributed yet, they are taken seriously by experienced
scholars. Former trainees of the CERA programme are active in international
meetings. For example, they represented about 10 percent of the participants
in the 1992 Vienna conference (Snell-Hornby, Pöchhacker & Kaindl 1994),
and more than 12 percent of the participants in the Prague EST Congress in
1995 (this volume). Undeniably, the Center has been fulfilling an active
international role, as noted in a number of publications (Schjoldager 1995;
Pöchhacker 1995; Hermans 1996; Gile 1996). This is remarkable, as CERA
ia an independent center operating with an international staff, rather than part
of a University or Institute, and has been trying very hard, sometimes to no
avail, sometimes running against opposition from existing institutions, to obtain
institutional support.
The Programme changed its name from CERA to CETRA in the fall of
1995 as the CERA Bank withdrew and the Institut Libre Marie Haps in
Brussels and Dhaxley Translation Inc. jumped in, but this has not had any
impact on the Center's objectives, though CE(T)RA has gradually evolved,
extending its activities and widening its field from the initially dominant
literary orientation to the international world of mass media, popular culture

and business communication.
One of the obvious difficulties lies in maintaining contact and interaction
with former trainees, which actually is one of the ultimate goals of the
programme. This is where the role of supervisors and other colleagues is most
important. Meetings, correspondence and conferences are more than a
pleasurable academic activity; they are the best possible foliowup to a learning
process that has become "professional". However, maintaining such contacts
is resource consuming. How can one continue to interact with an ever-growing
number of young colleagues from more than 30 countries without permanent
human and logistic support throughout the academic year? The fact that
Belgian Universities (as opposed to specialized tertiary level schools) do not
train translators and interpreters and struggle with doctoral curricula, having
started late, makes it even more ironical that (private) sponsoring is needed for
the task most specific to universities in an area which is by nature
interdisciplinary. In this respect, it is more than symbolic that an institute for
translator training (the Institut Libre Marie Haps) and a translation agency have
decided to give financial and intellectual support to a center for research
training with a view to promoting the "training of the trainers".
On another matter, CERA considered the use of computers and electronic
mail a must from the beginning. This however was not enough to allow smooth
integration of the four-week summer sessions into the academic year of the
supervisors, given the absence of an adequate administrative infrastructure and
support. New solutions have recently been found in the telematics-supported
world of Distance Learning. CETRA has suddenly discovered that it was born
and has been growing as a "networking entity". The use of Distance Learning
opens radically new perspectives in which flexibility is a central asset, if not
a must. The development of TRANSCETRA during the academic year (as
opposed to the CERA summer session programme) as a combination of
videoconference and computer conference sessions between a small number of
centers indicates that consortia and interuniversity curricula for the training of
scholars will soon be the rule rather than the exception. The fact that networks
such as CETRA are not developing in an environment of canonized curricula
and disciplines may be a strong handicap. Perhaps they are doomed to
disappear from the face of the earth as soon as the (more) institutionalized
disciplines find out that they also have to struggle for survival. However, the
CETRA example shows at least that networking and training scholars in this
format is more than just a nice idea. All the feedback received so far seems to
indicate that participants in the process have never challenged the basic
principle after experiencing it. It is now clear that international centers for
research training and research services are the way to the future. Mobility of
people will not stop. The emphasis should now be on the mobility of concepts
and curricula. The question is how to get the most out of this situation and
how to involve EST as a partner and/or initiator of action.
Translation Studies research: A supervisor's perspective
Translation Studies research in Austria must be seen from two differing, even
conflicting points of view. On the one hand, the translation schools still see
themselves essentially as training schools for future practising professionals,1
and they have virtually no research background; on the other hand however,
they are fully-fledged institutes of their university Arts Faculties, which are
products of the specifically Austrian academic system. This too is an
ambivalent construct which must be seen in its own terms and is by no means
identical with the system in Anglo-Saxon countries. On the one hand it is
deeply rooted in the Humboldtian tradition of academic freedom and the
indivisibility of teaching and research; on the other hand however, the
university structure was radically changed by laws passed in the 1970s, putting
much of the power into the hands of committees and functionaries. This means
that while university professors theoretically have the right to teach what they
like, in practice they are bound by the curriculum and by regulations beyond
their control.
The Chair of Translation Studies2 at the University of Vienna was
established in 1989, and it was my declared policy as first Full Professor of
Translation Studies to promote a young generation of translation scholars for
the future development of our discipline. For Translation Studies this seems to
be absolutely essential: as Daniel Gile has already pointed out, there are still
very few academic Translation departments, and most of us are self-taught,
1
This is clear from the name (Institut für Übersetzer- und Dolmetscherausbildung) given
to all three schools (in Vienna, Graz and Innsbruck). Attempts to change the name, with
reference to both the lack of transparency for non-German speakers and to the emergence
of Translation Studies as an independent discipline, have so far been unsuccessful. Plans
for a curriculum reform, which have been under way for some years, have resulted in
proposals to change the name of the two degree courses Übersetzerausbildung and
Dolmetscherausbildung to Übersetzungswissenschaft and Dolmetschwissenschaft, but at
the time of writing these have not yet been passed by the government officials
responsible.
2
In the Austrian system, a university chair is an academic post of Full Professor, and not
an administrative one (Head of Department).
coming from other fields. Within the Austrian system there are two ways of
fostering research: by creating academic posts for young researchers (known
as Universitäts-Assistenteri) and by encouraging suitable candidates to take
post-graduate degrees. There are three levels of academic qualification in
Austria: the undergraduate level (resulting in the title of Magister3), the
doctoral level, and the post-doctoral level leading to the Habilitation4.
The undergraduate 'diploma thesis'
Since 1972 the diploma thesis (Diplomarbeit) has been a requirement for the
degree of Magister in Austrian translation schools. Before the establishment of
the Chairs5, students had to look for qualified supervisors from other
departments; in Vienna they still can and indeed often do so. Hence in theory
there is a potential for constructive interdisciplinary cooperation. According to
the regulations, the topic of the diploma thesis must fall under one of the
following headings:
— Theoretical problems of translation and interpreting
— Linguistic phenomena of the candidate's B or C language
— Translation critique and comparison
— Terminology and lexicography
— Contrastive cultural studies.
The regulations require only one supervisor (who must have the post-doctoral
Habilitation), but allow for co-supervision. This makes it possible, particularly
in the case of highly specialized terminological studies, for the student to be
supervised both by a translation scholar (or terminologist) and by an expert in
the subject area concerned. Using this opportunity for interdisciplinary
collaboration, I have supervised a number of diploma theses in cooperation
with colleagues from other institutes and faculties, with extremely satisfactory
results (see Snell-Hornby 1991:17 and 1995), and such possibilities should be
explored further.
Undergraduate diploma theses of this kind are sometimes valuable case
studies, unfortunately doomed to collect dust on the shelves of the institute
3
This should not be confused with a post-graduate Master's degree in the Anglo-Saxon
system.
4
This is the basic requirement in German-speaking countries for a professorship, but also
provides the official qualification for supervising undergraduate and doctoral theses.
5
The appointment to the Chair in Graz was made in 1988, in Innsbruck in 1990.
library. There is great potential for research here, 6 at present underdeveloped

in Austria due to the lack of research training at undergraduate level — and
hence the diploma thesis is for most students a foreign body in their course of
studies. This could be at least partially remedied by the curriculum reform
which has been in preparation for a number of years, but due to the constraints
of budget and bureaucracy, has still not been approved by the government
officials responsible.
The doctoral thesis
At doctoral level there are as yet no specific regulations for the translation
schools in Austria, hence there is a good deal of freedom from curricular
constraints, 7 and it is here that we have had the most satisfactory results so
far. On the negative side is once again the lack of training in scholarly work
at undergraduate level, so that candidates embark on their thesis with some
factual knowledge of the area that interests them, but with very little
knowledge of translation theory and the background literature, and hardly any
practice in academic writing or research methodology. So all this usually has
to be learnt from scratch, and the situation will not change basically until the
planned curriculum reform is eventually put into practice. To date 1995 five
doctoral degrees (Dr.phil.) have been completed under my supervision in
Vienna, and it is to the credit of those concerned that they managed to remedy
the above deficits to meet the high standards of quality required of them. 8
In Austria the doctoral candidate needs both a supervisor and a co-
supervisor; in accordance with the principle of academic freedom he/she
approaches the professors of his/her choice and presents a clearly defined
project, which is accepted or rejected according to the policy, research
interests or work load of the professors concerned. The interest in obtaining
a doctor's degree in Translation Studies in Vienna has been overwhelming, and
I must have been approached by well over 50 would-be candidates, many of
whom were however unaware of the demands scholarly research would make
on them, and hence their interest barely survived the first interview. All the
6
This is recognized by the University authorities, who provide scholarships for students
who need to do their research abroad.
7
The degree is awarded on the basis of the general university regulations, which simply
require in all four certificates from seminars recommended by the supervisor and co-
supervisor.
8
Some of these theses have meanwhile been published (e.g. Pöchhacker 1994; Kaindl
1995; Kurth 1995).
same there are close on 20 research projects currently in progress on a broad

range of topics from film dubbing and stage translation to quality management,
court interpreting, translatorial competence, text design, cross-cultural
communication and the problems involved in the special languages of law and
psychiatry. My basic approach is pluralistic, holistic and interdisciplinary, and
most of the projects are co-supervised by colleagues from the special field
involved, hence usually from other faculties. Apart from personal consultation,
the main forum for supervision is the doctoral seminar, where the candidates
present papers on their research in progress. At present this seminar has been
divided into two groups, to allow for a 'beginners' seminar on presentation
and methodology.
After the doctor's degree has been completed, anyone interested in a
university career can go on to the Habilitation, sending his/her application to
the Dean of the Faculty (on the basis of a lengthy research project plus other
scholarly publications), who appoints a special committee to deal with each
individual case. The candidate does not require a personal supervisor as such;
but the committee approaches two professors specializing in the area of his/her
research for a detailed assessment. There have been two Habilitationen in
Translation Studies at Vienna since my appointment (see Nord 1993 and Kurz
1996).
Future prospects
It cannot however be the final aim of anyone personally interested in doing

research only to pave the way for individual academic careers. In the Austrian
system the university professor has both the right and the duty, not only to
supervise, but also to carry out research him/herself — the posts of the
Universitäts-Assistenten were originally intended to support this (and in other
German-speaking countries still are). Although in the daily round of today's
university, with the paralyzing constraints of excessive administration and
bureaucracy, this duty often turns out to be wishful thinking, I still have not
abandoned the hope of creating a research centre in Translation Studies with
a team of dedicated and committed young scholars.
For such a purpose however, there are certain prerequisites. One is of
course adequate funding, which, if available at all, all too often evaporates
under the dictates of bureaucracy and power struggles. What is at present of
more urgent interest for the immediate future of Translation Studies research
is to create an intellectual infrastructure by adequate training at undergraduate
level. As has been pointed out above, this is what is still sadly lacking at the
Austrian translation schools: a university degree course where students

naturally acquire the habit of critical thinking and the ability to assess the vast
amount of factual information they are given, where they acquire skill in
producing convincing arguments and well-structured texts, and where the
foundations of scholarly work are laid.
Conclusion
As announced in the introduction, this report can only be a trigger for further
reflection on research training issues. However, a number of points were made
salient by the speakers:
— The theme of research training and its institutional and organizational
aspects are very important for the future of the (inter)discipline of
Translation Studies.
— The interdisciplinary nature of Translation Studies seems to call for an
appropriate organizational structure in which co-supervision plays a central
role.
— The development of Translation research has yet to struggle into official
existence, sometimes fighting opposition from vocational translator and
interpreter training institutions besides the opposition from academic in-
stitutions whose territory is threatened by the newcomers.
— The administrative burden on research training has been a significant
handicap in the activities of Translation Studies leaders.
— Imagination and innovation could be particularly useful assets for over-
coming barriers, as the CE(T)RA and TRANSCETRA examples show.
It is hoped that this first EST focus session will pave the way for more
systematic reflection on these issues, in particular through detailed and
systematic reports on the situation in the countries and institutions that EST
Members belong to and on solutions that have been attempted.
References
Cenkova, Ivana. 1995. "La recherphe en interprétation dans les pays d'Europe de l'Est: une
perspective personnelle". Target 7:1, 75-89.
Gile, Daniel. 1995. Regards sur la recherche en interprétation de conference. Lille: Presses
Universitaires de Lille.
Gile, Daniel. 1996. "La formation à la recherche traductologique et le concept CERA Chair".
Meta 41:3, 486-490.
Gran, Laura/Maurizio Viezzi. 1995. "Development of Research Work at SSLM, Trieste".
Target 7:1, 107-118.
Hermans, Theo. 1996. "Vertaalwetenschap in de Lage Landen". Neerlandici extra Muros

XXXII/3, 1-13.
Kaindl, Klaus. 1995. Die Oper als Textgestalt. Perspektiven einer interdisziplinären
Übersetzungswissenschaft. Tübingen: Stauffenburg.
Kurth, Ernst-Norbert. 1995. Metaphernübersetzung. Dargestellt an grotesken Metaphern im
Frühwerk Charles Dickens in der Wiedergabe deutscher Übersetzungen. Frankfurt: Lang.
Kurz, Ingrid. 1996. Simultandolmetschen als Gegenstand der interdisziplinären Forschung.
Wien: WUV.
Nord, Christiane. 1993. Einführung in das funktionale Übersetzen. Am Beispiel von Titeln und
Überschriften. Tübingen: Francke.
Pöchhacker, Franz. 1995. "'Those who do...': A Profile of Research(ers) in Interpreting".
Target 7:1, 47-64.
Pöchhacker, Franz. 1994. Simultandolmetschen als komplexes Handeln. Tübingen: Narr.
Schjoldager, Anne. 1995. "Interpreting Research and the 'Manipulation School' of Translation
Studies". Target 7:1, 29-45.
Snell-Horaby, Mary. 1991. "The professional translator of tomorrow: Language specialist or
all-round expert?" In: C. Dollerup/ A. Loddegaard (eds.) Teaching Translation and
Interpreting. Training, Talent and Experience. Amsterdam: Benjamins, 9-22.
Snell-Hornby, Mary. 1995. "Traduzione e interpretariato: un percorso interdisciplinare da Yin
e Yang a Don Giovanni". In: R. Arntz (ed.) La Traduzione — Nuovi approcci tra teoria
e pratica.
Strolz, Birgit. 1995. "Une approche asymptotique de la recherche sur l'interprétation". Target
7:1,65-74.
Toury, Gideon. 1991. "Experimentation in Translation Studies: Achievements, Prospects and
some Pitfalls". In: S. Tirkkonen-Condit (ed.) Empirical Research in Translation and
Intercultural Studies. Tübingen: Narr, 45-6
List of contributors
Sirkku Aaltonen
University of Vaasa, P.O. Box 297, 65101 Vaasa, Finland
Beverly, Adab
Aston University,Department of Modern Languages, Aston Triangle, B4 7 ET
Birmingham, United Kingdom
Rosemary Arrojo
Universidad Estadual de Campinas S. Paulo, Institute de Estudos da Linguagem,
Rua Marquês de Abrantes 382, 03060 020 Sao Paulo, S.P. Brazil
Leo Tak-hung Chan
Lignan College, Tuen Mun, HongKong
Andrew Chesterman
University of Helsinki, Department of English, P.O. Box 4 (Hallituskatu 11),
00014 Helsinki, Finland
Anneke de Vries
Uiterwaardenstraat 137, 1079 CK Amsterdam, The Netherlands
Cay Dollerup
University of Copenhagen, Center for Translation Studies and Lexicography,
Njalsgade 80, 2300 Copenhagen, Denmark
Daniel Gile
10, rue Pasteur, 92190 Meudon, France
Jean-Marc Gouanvic
Concordia University, Department of French Studies, 7141 Sherbrooke West,
Montreal H4B IR6, Canada
Theo Hermans
University College London, Centre for Low Countries Studies, Gower Street,
London WC1E 6BT, United Kingdom
Zuzana Jettmarová
Charles University, Institute of Translation Studies, Hybernskà 3, 11000 Praha
1, Czech Republic
352
Udo Jörg
132b Bravington Road, Westkillburn, London W9 3AL, United Kingdom
Mira Kadric
Universität Wien, Institut für Übersetzer- und Dolmetscherausbildung, Gymnasi-
umstr. 50, 1190 Wien, Austria
Klaus Kaindl
Christine Klein-Braley
Universität Duisburg, Institut für Anglistik und Amerikanistik, 47048 Duisburg,
Deutschland
Rainer Kohl may er
Johannes Gutenberg-Universität Mainz, FASK-Germersheim, An der Hochschule
2, 76711 Germersheim, Deutschland
Irena Kovacic
Univerza V Ljublijani, Department of English, Askerceva 2, 1000 Ljubljana,
Slovenia
Ingrid Kurz
Paul Kussmaul
José Lambert
Katholieke Universitet Leuven, Dept. Literatuurwetenschap, Blijde-Inkomstraat
21, 3000 Leuven, Belgium
Dorte Madsen
Copenhagen Business School, Department of Spanish, Dalgas Have 15, 2000
Frederiksberg, Denmark
Marta Mateo
Marqués de Pidal 7-2, 33004 Oviedo, Spain
353
Saliha Paker
Bogazici University Istanbul, Department of Translation and Interpreting,
Yadyok, P.K. 2, 80815 Istanbul, Turkey
Maria Piotrowska
Pedagogical University, NKJÀ, WSP, ul. Podcharazch 2, 30-084 Kraków,
Poland
Franz Pöchhacker
Renate Resch
Hanna Risku
Andersengasse 1/69/12, 1120 Wien, Austria
Irene Rübberdt
Wustrowerstr. 9, 13051 Berlin, Deutschland
Heidemarie Salevsky
Niebergallstr. 3, 12557 Berlin, Deutschland
Christina Schäffner
Aston University, Department of Modern Languages, Aston Triangle, B4 7 ET
Birmingham, United Kingdom
Veronica Smith
Universität Klagenfurt, Institut für Anglistik und Amerikanistik, Universitätsstr.
65-67, 9020 Klagenfurt, Austria
Mary Snell-Hornby
Ubaldo Stecconi
Department of English, Ateneo de Manila University, The Philippines
Maria Luisa Torres Reyes
Department of English, Ateneo de Manila University, The Philippines
Zehra Toska
Bogazici University Istanbul, Department of Translation and Interpreting,
Yadyok, P.K. 2, 80815 Istanbul, Turkey
Erkka Vuorinen
University of Tampere, Department of Translation Studies, PL 607, 33101
Tampere, Finland
354
Michaela Wolf
Universität Graz, Institut für Übersetzer- und Dolmetscherausbildung, Mer-
angasse 70, 8010 Graz, Austria
leva Zauberberga
University of Latvia, Department of Contrastive Linguistics, Visvalza 4A, 1011
Riga, Latvia
Benjamins Translation Library
A complete list of titles in this series can be found on www.benjamins.com
84 Monacelli, Claudia: Self-Preservation in Simultaneous Interpreting. Surviving the role. Expected April
2009
83 Torikai, Kumiko: Voices of the Invisible Presence. Diplomatic interpreters in post-World War II Japan.
2009. x, 197 pp.
82 Beeby, Allison, Patricia Rodríguez Inés and Pilar Sánchez-Gijón (eds.): Corpus Use and
Translating. Corpus use for learning to translate and learning corpus use to translate. x, 151 pp. + index.
Expected February 2009
81 Milton, John and Paul Bandia (eds.): Agents of Translation. 2009. vi, 337 pp.
80 Hansen, Gyde, Andrew Chesterman and Heidrun Gerzymisch-Arbogast (eds.): Efforts and
Models in Interpreting and Translation Research. A tribute to Daniel Gile. 2009. ix, 302 pp.
79 Yuste Rodrigo, Elia (ed.): Topics in Language Resources for Translation and Localisation. 2008.
xii, 220 pp.
78 Chiaro, Delia, Christine Heiss and Chiara Bucaria (eds.): Between Text and Image. Updating
research in screen translation. 2008. x, 292 pp.
77 Díaz Cintas, Jorge (ed.): The Didactics of Audiovisual Translation. 2008. xii, 263 pp. (incl. CD-Rom).
76 Valero-Garcés, Carmen and Anne Martin (eds.): Crossing Borders in Community Interpreting.
Definitions and dilemmas. 2008. xii, 291 pp.
75 Pym, Anthony, Miriam Shlesinger and Daniel Simeoni (eds.): Beyond Descriptive Translation
Studies. Investigations in homage to Gideon Toury. 2008. xii, 417 pp.
74 Wolf, Michaela and Alexandra Fukari (eds.): Constructing a Sociology of Translation. 2007. vi, 226 pp.
73 Gouadec, Daniel: Translation as a Profession. 2007. xvi, 396 pp.
72 Gambier, Yves, Miriam Shlesinger and Radegundis Stolze (eds.): Doubts and Directions in
Translation Studies. Selected contributions from the EST Congress, Lisbon 2004. 2007. xii, 362 pp. [EST
Subseries 4]
71 St-Pierre, Paul and Prafulla C. Kar (eds.): In Translation – Reflections, Refractions, Transformations.
2007. xvi, 313 pp.
70 Wadensjö, Cecilia, Birgitta Englund Dimitrova and Anna-Lena Nilsson (eds.): The Critical
Link 4. Professionalisation of interpreting in the community. Selected papers from the 4th International
Conference on Interpreting in Legal, Health and Social Service Settings, Stockholm, Sweden, 20-23 May
2004. 2007. x, 314 pp.
69 Delabastita, Dirk, Lieven D’hulst and Reine Meylaerts (eds.): Functional Approaches to
Culture and Translation. Selected papers by José Lambert. 2006. xxviii, 226 pp.
68 Duarte, João Ferreira, Alexandra Assis Rosa and Teresa Seruya (eds.): Translation Studies at the
Interface of Disciplines. 2006. vi, 207 pp.
67 Pym, Anthony, Miriam Shlesinger and Zuzana Jettmarová (eds.): Sociocultural Aspects of
Translating and Interpreting. 2006. viii, 255 pp.
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 and 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.

Translation As Intercultural Communication Selected Papers From The Est Congress Prague 1995 9789027216212 9781556197024 9789027285614 Compress

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Translation As Intercultural Communication Selected Papers From The Est Congress Prague 1995 9789027216212 9781556197024 9789027285614 Compress

Uploaded by

Copyright:

Available Formats

TRANSLATION AS INTERCULTURAL COMMUNICATION

BENJAMINS TRANSLATION LIBRARY

Mary Snell-Hornby, Zuzana Jettmarová and Klaus Kaindl (eds)

Translation as Intercultural Communication

JOHN BENJAMINS PUBLISHING COMPANY

Library of Congress Cataloging-in-Publication Data

Part I: Translation — Sociology, Culture and Ethics

Leo Tak-hung Chan

Saliha Paker/Zehra Toska

Part II: Translation and Beyond — Aspects of Communication

Part III: Panel Discussions

Christina Schäffner/Beverly Adab

List of Contributors 351

The first International Congress of the European Society for Translation

It is conspicuous that most of the contributions are written in English,

Vienna/Prague, March 1997

Translation — Sociology, Culture and Ethics

Let us imagine the following situation as real. We find ourselves in a packed

in his collection Philosophical Papers (1961). Austin's essay goes on to

Conventions and norms

Sense and self-reproduction

in 'the words on the page', decipherable by means of linguistic and other

speak on behalf of others, interpreters among them, re-express the original

communicative atmosphere. The result affects the communicative position of

Self-reference and metalanguage

A similar duality, or ambivalence, pertains to discourses about translation. As

Toury (1986) wondered why Jakobson listed two kinds of 'lingual'

other transformational operations not normally covered by the term

practices of other disciplines, allows us at least to pinpoint the issue, and to be

The "death" of the author and the birth of the reader

The notion of abusive fidelity

A similar paradox seems to underlie certain notions defended by a few

invisible and becomes "strong" and "forceful," requiring, as Venuti concludes,

translation. If we accept the impossibility of perfect repetition, then any

The limits of the translator's inevitable visiblity

Thus, an appropriate.poststructuralist model for the translator's inescapably

well be found in the full implications of Derrida's conception of translation as

own voice" (1995:287) The transparency idealized by tradition is not exactly

Díaz-Diocaretz. 1985. Translating Poetic Discourse: Questions on Feminist Strategies in

Sociologie des productions symboliques et "effet traduction"

Dans un espace culturel, tout se passe comme si s'opérait une division du

[L'] investissement (aux sens de l'économie et de la psychanalyse) c'est

littérature traduite, même la plus légitimée, n'est pas traitée comme la

certaine marge de manœuvre en fonction de sa légitimité sociale.

La traduction de la science-fiction américaine et l'émergence d'un champ

Avant 1951 en France les romans scientifiques indigènes et a fortiori traduits

cheval sur l'héritage des romans populaires d'avant-guerre et le modèle de la

La littérature canonique américaine dans l'espace culturel français après

On pourrait penser que la dominance américaine en science-fiction serait dans

géopolitique", mais dans la "distance historique". Cette "distance géo-

Senders and translation

'Imposition' vs. 'requisition' by means of 'cultural bridgeheads'

When translation is forced upon source texts, their realisations in target

culture, and therefore pays much attention to the intention or intentionalities

Large-scale translation: the 19th century

I have suggested that large-scale professional translation, in the sense that

Shelley. As gateway language, German provided source texts for minor

The present scene

translations of national classics.7 Even so, this procedure is fraught with

Nida, Eugene A. 1964. Toward a Science of Translation. Leiden: Brill.

Leo Tak-hung Chan

Much of the current evaluation of Chinese translation theory has tended