Why Translation Studies Matters

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

EST Subseries The European Society for Translation Studies (EST) Subseries is a publication channel within the Library to optimize EST’s function as a forum for the translation and interpreting research community. It promotes new trends in research, gives more visibility to young scholars’ work, publicizes new research methods, makes available documents from EST, and reissues classical works in translation studies which do not exist in English or which are now out of print.

General Editor

Associate Editor

Honorary Editor

Yves Gambier

Miriam Shlesinger

Gideon Toury

Rosemary Arrojo

Zuzana Jettmarová

Rosa Rabadán

Michael Cronin

Werner Koller

Sherry Simon

Daniel Gile

Alet Kruger

Mary Snell-Hornby

José Lambert

Sonja Tirkkonen-Condit

John Milton

Maria Tymoczko

University of Turku

Bar-Ilan University Israel

Tel Aviv University

Advisory Board Binghamton University Dublin City University Université Paris 3 - Sorbonne Nouvelle

Ulrich Heid

University of Stuttgart

Amparo Hurtado Albir

Universitat Autónoma de Barcelona

W. John Hutchins

University of East Anglia

Charles University of Prague Bergen University UNISA, South Africa Catholic University of Leuven University of São Paulo

Franz Pöchhacker

University of Vienna

Anthony Pym

University of León Concordia University University of Vienna University of Joensuu

University of Massachusetts Amherst

Lawrence Venuti

Temple University

Universitat Rovira i Virgili

Volume 88 (Volume 6 in the EST Subseries) Why Translation Studies Matters Edited by Daniel Gile, Gyde Hansen and Nike K. Pokorn

Why Translation Studies Matters Edited by

Daniel Gile Université Paris 3 - Sorbonne Nouvelle

Gyde Hansen Copenhagen Business School

Nike K. Pokorn University of Ljubljana

John Benjamins Publishing Company Amsterdamâ•›/â•›Philadelphia

8

TM

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 Why Translation Studies matters / edited by Daniel Gile, Gyde Hansen and Nike K. Pokorn. p. cm. (Benjamins Translation Library, issn 0929-7316 ; v. 88) Includes bibliographical references and index. 1. Translating and interpreting--Study and teaching. I. Gile, Daniel. II. Hansen, Gyde. III. Pokorn, Nike K. (Nike Kocijancic) P306.5.W49â•…â•… 2010 418’.02071--dc22 2009039870 isbn 978 90 272 2434 7 (Hb ; alk. paper) isbn 978 90 272 8864 6 (Eb)

© 2010 – 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 36224 · 1020 me Amsterdam · The Netherlands John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa

Table of contents Preface

ix

Part 1.╇ Does TS matter? Why interpreting studies matters Franz Pöchhacker What matters to Translation Studies? On the role of public Translation Studies Kaisa Koskinen

3 15

Part 2.╇ Translation and society Translators as cultural mediators: Wish or reality? A question for Translation Studies David Limon

29

Censorship in the translations and pseudo-translations of the West Carmen Camus Camus

41

A world without God: Slovene Bambi Nike K. Pokorn

57

Manipulating the matricial norms: A comparison of the English, Swedish and French translations of La caverna de las ideas by José Carlos Somoza Yvonne Lindqvist

69

Knowledge in Translation Studies and translation practice: Intellectual capital in modern society Hanna Risku, Angela Dickinson and Richard Pircher

83

Part 3.╇ Language issues Is Translation Studies going Anglo-Saxon? Critical comments on the globalization of a discipline Mary Snell-Hornby Slowakisch: Brückensprache zur slawischen Welt? Möglichkeiten und Grenzen einer kleinen EU-Sprache Martina Vankúšová

97

105



Why Translation Studies Matters

Translation Studies and mass media research Rachel Weissbrod

115

Register shifts in translations of popular fiction from English into Slovene Marija Zlatnar Moe

125

Getting the ACCENT right in Translation Studies Ian A. Williams

137

Die Kirche im Dorf oder die Regierung im Wald lassen: Zum Übersetzungsproblem der Namen von Ämtern und Ähnlichem für Nachrichtenzwecke im Medium Radio Dieter Hermann Schmitz

153

Part 4.╇ Assessment and training Magnifying glasses modifying maps: A role for translation theory in introductory courses Heloísa Pezza Cintrão

167

Effects of short intensive practice on interpreter trainees’ performance Magdalena Bartłomiejczyk

183

Corpora in translator training: A program for an eLearning course Kerstin Kunz, Sara Castagnoli, Natalie Kübler

195

Part 5.╇ Psychology Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung im Rahmen der interdisziplinären Forschung Caroline Lehr

211

Interpreting Studies and psycholinguistics: A possible synergy effect Agnieszka Chmiel

223

fMRI for exploring simultaneous interpreting Barbara Ahrens, Eliza Kalderon, Christoph M. Krick, Wolfgang Reith

237

Part 6. Postscript Why Translation Studies matters: A pragmatist’s viewpoint Daniel Gile

251

Name index

263

Concept index

267

Preface This volume is a collection of papers around the theme “Why Translation Studies matters”. Some authors address the topic explicitly and point to contributions to training, to translation practice, to society at large, to other disciplines. Others prefer to focus on particular aspects of Translation, Translator training or Translation Studies, leaving it to the readers to identify the contribution of TS on the basis of the studies presented. The first paper, by Franz Pöchhacker, is one of the few which focus on the theme from a wide, comprehensive angle. It refers explicitly to Interpreting Studies, but applies equally well to Translation Studies as a whole. Pöchhacker argues that the existence of a separate body devoted to Interpreting Studies meets an epistemological need. He defines the “market” for IS, starting with interpreter training, moving on to the IS community and to practitioners of interpreting. He highlights the potential importance of IS to the community interpreting environment and suggests that interpreting scholars could help make IS more meaningful by stressing ‘development’, inter alia by participating in interdisciplinary committees which could improve the working environment of interpreters. This idea is also taken up by Kaisa Koskinen, who also expresses her opinions on the desirable contribution of TS. Her paper is an example of interdisciplinary import: she refers to ideas from sociologist Michael Burawoy, who divided sociology into a two-by-two matrix showing the four possible combinations of types of audience (“academic” vs. “non-academic”) and knowledge (“instrumental” vs. “reflexive”). She takes this matrix as a source of inspiration for her own reflection, explains that Translation scholars have not addressed sufficiently the non-academic audience and joins Pöchhacker in calling for direct, active engagement of TS in society. While such active intervention may help TS matter more, by essence, scholarly activity is primarily about exploring. Various papers in this collective volume explore the role and position of Translation in society. Here, TS attempts to discuss and reveal the underlying hegemonic structures that find their expression in a particular translation practice, and thus provides an insight into the position of the translator, translation or particular genres in the observed society. In his contribution, David Limon says that translators are increasingly referred to as “cultural mediators” – which implies an active intervention in the communication process beyond the conduit model – but finds that in the field, they do not necessarily play such a role. He illustrates this with the example of translations of texts about Karst, a type of limestone which plays a role in the natural heritage of Slovenia. He finds that in English translations of such texts there is no cultural explicitation. Limon speculates

 Why Translation Studies Matters

about the reasons and suggests inter alia that raising the translators’ status would help empower them to intervene more actively as cultural mediators in their work. Carmen Camus Camus looks at another aspect of social constraints on translation, in this case censorship in Spain during Franco’s period. She examines the censorship files regarding Western narrative texts, which were very popular in Spain during that period, including both translations and pseudo-translations, and provides information on publications during various phases of Franco’s regime, including the 1946 Book promotion law and the subsequent economic incentives given by the Spanish government to the publication of Western narratives. She concludes with examples of censorship on texts which she analyzes. On the other side of the political spectrum, Nike K. Pokorn analyzes the influence of self-censorship in a Socialist country. Her research covers the translation of children’s literature in Slovenia from 1945 to 1963. The case study presented in her paper focuses on Felix Salten’s Bambi. The one aspect she chooses to highlight here is selftriggered attenuation of religious overtones in the translation of the last passages of the story, which, she argues, is found throughout translations of juvenile literature and reflects the attitude of communism towards the Judeo-Christian paradigm at the time. Another type of literature, namely crime fiction, is studied by Yvonne Lindqvist, with a literary polysystem-oriented analysis and a comparison of British, Swedish and French translations of José Carlos Somoza’s La caverna de las ideas. Lindqvist says that the English translation was the most manipulated of the three, and that this is due to differences in the position of translated texts within the three literary systems, the English system being more closed and requiring more information in the translation. She concludes that studies with both micro-level and macro-level components could help gain more insight into the translator’s habitus. Hanna Risku, Angela Dickinson and Richard Pircher’s paper takes a more economic view of the role of translation in society in terms of knowledge production and sharing as an economic asset. They discuss approaches to Knowledge Management and then go on to present translators as knowledge experts who generate intellectual capital and are good communicators. They look at KM instruments for codifiable and noncodifiable aspects of knowledge required for translation. They conclude that both translation and Translation Studies are relevant to companies as producers and sharers of intellectual capital which can add value to their organizations’ production, but that they are also relevant to society at large, inter alia in terms of intercultural communication. Translation is about knowledge, but obviously also about language. TS can help reveal the forces that shape different language policies, the position of particular languages in given societies as well as linguistic change. Several contributions in this volume focus on language issues, starting with Mary Snell-Hornby’s paper on globalization and English. While the existence of a lingua franca can be seen as a positive factor which facilitates communication, Snell-Hornby highlights limitations and drawbacks in the increasing use of less-than-perfect English for academic exchanges worldwide.



Preface 

Listeners sometimes find it difficult to follow non-native speakers taking the floor in English, and scholars with no English have no chance to participate in many debates. Snell-Hornby proposes that no single natural language should monopolize global communication in TS and is in favour of passive knowledge (understanding, as opposed to producing) of several languages, as well as the use of bridge languages for better communication. The concept of bridge languages is precisely the focus of Martina Vankúšová’s contribution, which addresses the specific case of translation and interpreting for the European Union where a large percentage of translations are done from Euro-English into other languages – but problems may arise with occasional use of non-English. Bridge-languages could help individuals communicate with speakers of other related languages. While there are solutions for Romance languages (French) and for Germanic languages (English), the situation is more difficult for Slavonic languages. Russian is considered unsuitable, if only because of its use of the Cyrillic script. Slovakian and Slovenian are other possibilities. Hebrew is the main language in Israel. It is understood and used actively by all, but strong minority languages which the mass media use include Arabic and Russian. On the other hand, English has also permeated strongly Israeli society. This intense interpenetration of several linguistic systems in a small population raises interesting questions for sociolinguists with respect to national identity and the integration of minorities. Rachel Weissbrod argues that Translation Studies could benefit from closer scrutiny of phenomena around multilingualism in the media, and suggests that, conversely, TS can provide insights into the construction of social identity in a multicultural society. Moving from language and society to language in translation, Marija Zlatnar Moe studies language shifts in the translation of popular fiction texts into Slovene. In her sample of translations of seven books from different genres, she finds a general trend towards a more neutral, unmarked formal register which does not reflect the style of the original.€She argues that this can make a well-written book uninteresting and disappointing to readers who do not have access to the source language. By highlighting such phenomena, TS could help make society at large aware of the risks associated with poor translation. Beyond awareness-raising, it could look at the reasons behind the losses and perhaps help find some remedies, especially in translator training. Ian Williams proposes a corpus-based approach and methodology to look at linguistic profiles of texts. When comparing such texts in source- and target languages, regular but subtle differences in collocation, colligation, semantic and semantic prosody preferences which can well go unnoticed in traditional human observation can be identified and quantified, and thus provide input for possible improvement of translation. The method, called Application of Corpus-based Contrastive Evaluation for Natural Translation (ACCENT), is illustrated by a comparison of the use of the first person verb use in English and Spanish biomedical research papers.



Why Translation Studies Matters

A paper with direct applications for translator training is offered by Dieter Hermann Schmitz, who looks at the problem of translating names of organizations. More specifically, he reflects on his daily practice as a translator trainer involved in the translation of Finnish radio broadcasts at his university in Tampere into German. Taking the name of a Finnish industrial association as a starting point, he shows that literal translation of such names may not be a good solution, discusses possible options and formulates general strategies for the translation of names. Training and translation assessment are the focus of the following section of the book. Heloísa Pezza Cintrão discusses the role of theory in translator training. She offers data from a case study where concepts from functional theory, from text linguistics and from cognitive research into inferencing processes and problem-solving were presented to students at an introductory class. A controlled experiment was designed to test the effect of such awareness-raising through theoretical concepts on actual translation performance. Findings suggest that presenting declarative knowledge about translation-related theoretical elements does contribute to the students’ progress, and rather rapidly at that. Introducing theoretical concepts is one potential way to accelerate learning. Another is a short internship-like experience. Magdalena Bartłomiejczyk reports on the results of such an operation, organized for her conference interpreting students. Students interpreted simultaneously at a real conference in silent booths (no output was available to delegates) for a total number of hours roughly equivalent to a term and a half in the booth at their university. A before-after experiment done on a set of two English speeches to be interpreted into Polish assessed the actual effect in terms of propositional accuracy gains (“fidelity”) and in terms of autonomous output quality. The findings serve as input for further improvement of the training programme. Kerstin Kunz, Sara Castagnoli and Natalie Kübler describe an e-learning course on the use of corpora for translation targeting both professional translators and trainees, which is designed to give vocational training in the application of new technologies such as Translation Memories, Markup Languages, Machine Translation and Corpora which span activities going from traditional translation to localization, terminology management, information management and project management. This course was developed by the EU-funded MELLANGE project which, according to the authors, fills a gap associated with the lack of awareness by translators of corpora and associated technologies. From within the ranks of Translation Studies, it has often been claimed that investigating translation-related phenomena has something to offer to other disciplines which address language and communication problems. Psychology and more specifically psycholinguistics has been interested in the cognitive development of bilinguals. Caroline Lehr reports on a study in which linguists submit translation students to a German and French lexical decision task, the purpose being to investigate linguistic data processing in bilinguals. She elaborates on further prospects for the extension of interdisciplinary cooperation between psychology and Translation Studies.



Preface 

Agnieszka Chmiel follows a similar line, starting with the import of concepts and theories from cognitive psychology and psycholinguistics, more specifically ideas about the mental lexicon and about memory. While Lehr refers to the contribution translators can make to psycholinguists, Chmiel focuses on interpreting. She introduces her longitudinal project aimed at developing effective aptitude tests for interpreter training programs on the basis of cognitive factors. She reports on two pilot studies, one on the processing of cognates and the other on semantic verbal fluency, and suggests that further studies might put to good use potential synergy between Interpreting Studies and psycholinguistics. Barbara Ahrens, Eliza Kalderon, Christoph Krick and Wolfgang Reith are engaged in neurophysiological studies of the human brain using functional magnetic resonance imaging. Again, they are interested in the particular bilingualism of conference interpreters which they compare to foreign language students and to students who have little contact with foreign languages, and examine activation patterns in the brain during language switching operations in the three groups. In the study reported here, they looked at the applicability of fMRI in studying the relative activity of various parts of the brain during simultaneous interpreting vs. free speech production. They view their findings as a suitable starting point for further exploration about the evolution of brain activity and structures over the interpreters’ career. The last paper in this volume takes a deliberately conservative view but offers an optimistic conclusion. Daniel Gile challenges the idea that TS takes no interest in the practitioners’ needs. He concedes that its “scientific” contribution to the practice of Translation is difficult to measure – one might note in this respect that most contributions about the scientific effects of TS included in this volume relate to prospects rather than to achievements, but argues that its contribution to Translator training is already felt in the field. The paper ends with a comment: some authors may argue that TS is an epistemological requirement for translators and interpreters and helps address various aspects of society, of language, of translator training, of cognitive processes, but perhaps the community of practitioners could be reminded that TS as an academic discipline also has a role to play in defending and sometimes raising the Translators’ social status. The editors

Does TS matter?

Why interpreting studies matters Franz Pöchhacker

University of Vienna, Austria

This paper takes up the theme of the Ljubljana EST Congress from the perspective of interpreting studies. The basic assumption that the existence of a distinct research community dedicated to the study of interpreting corresponds to an epistemological need is illustrated with regard to related work in psychology and sociology. On the assumption that the research output of interpreting scholars needs a ‘market’, interpreter education, professional practice, and institutional user contexts are examined as domains for which interpreting studies might matter. With special emphasis on community settings, the potential of interpreting research to shape social practices in relevant institutional contexts is discussed with reference to recent examples. Given the obstacles to interdisciplinary publication and impact, it is suggested that interpreting scholars need to invest also in ‘D’ (development) rather than ‘R’ if they wish their findings to matter to those who shape the social practices in which (community-based) interpreting is embedded. Keywords: discipline, sociology, epistemology, needs, social usefulness

1. Introduction In response to the challenge posed by the theme of the EST Congress in Ljubljana, my aim in this paper is to consider why interpreting studies as a (sub)discipline, and the research produced by scholars affiliated with it, could be said to matter, and to whom. My fundamental assumption, that the existence of a distinct research community dedicated to the study of interpreting corresponds to an epistemological need, will be argued from a historical perspective, using examples from such more established disciplines as psychology and sociology to illustrate how interpreting used to be seen and studied – or not. Moving on to the idea that research output also needs a ‘market’, I will discuss in more detail to whom interpreting studies might matter, referring to such major domains as interpreter education, professional practice, and institutional (user) contexts. In particular, I will examine the claim that research on community-based interpreting addresses a social need and could therefore be said to be (more) ‘socially



Franz Pöchhacker

useful’. Using some recent examples, I will show how this may well be the case but, at the same time, point to some serious limitations in the capacity of interpreting research to matter quite enough.

2. A ‘science’ for interpreting My first and fundamental point is obvious enough: Interpreting studies as a scientific discipline – that is, a field within academia, a community whose members are engaged in creating (more) knowledge about their object of study by working according to what is broadly (though not uncontroversially) defined as the ‘scientific method’ – matters because it meets an epistemological need: ours is the only scholarly community that claims and takes charge of interpreting as its object of study. This amounts to saying that we exist because we have a unique object of study, and that we matter because we exist. Quite so! It matters that there is a scientific community for the study of interpreting as a distinct phenomenon – on the assumption that human society maintains a consensus that the ‘web of science’ should be woven as densely as possible. In our day and age, this consensus is largely taken for granted, but one need only to recall, say, China’s Cultural Revolution to realize that scientific endeavor as a social value is itself a social construction. But if this consensus holds, and if ever more knowledge is to be created about as many phenomena as possible, then interpreting studies simply has its raison d’être as a branch, or twig, of science. That is our function, our use to society: we ‘see’ – and study – things that others don’t.

2.1

Seen by psychologists

The issue of seeing interpreting as an object of study – or not – can be illustrated with reference to early research on interpreting from the vantage points of more established disciplines. Interpreting studies as such did not exist in the 1960s, so research on interpreting was done within other paradigms, mainly psychology and psycholinguistics. And that research, by people like Frieda Goldman-Eisler and David Gerver, did matter: interpreter subjects provided Goldman-Eisler with a rich source of spontaneous speech output to be examined for pauses and thus for indicators of cognitive processing activity (e.g.€Goldman-Eisler 1967); and Gerver was able to use his analysis of the task of simultaneous interpreting (SI) to propose a new conceptualization of the human cognitive processor, replacing the assumption of a fixed processing channel for one cognitive task only by that of a “fixed-capacity central processor”, whose activity could be distributed over several tasks within the limits of the total processing capacity available (Gerver 1971:€15f). Interpreting, and research on it, clearly mattered – to psychologists. Even the pioneering work of my Viennese colleague Ingrid Kurz, then Pinter, is an example of that phenomenon. Having been trained as an interpreter and



Why interpreting studies matters

at the same time worked on a degree in psychology, she devoted her doctoral thesis in psychology (Pinter 1969) to demonstrating what she knew from immediate experience, i.e. that it was in fact possible to learn to listen and speak at the same time. She took her cue from no less an authority in psychology than Donald Broadbent, one of the pioneers of cognitive psychology, whose study of flight control communication in pilots had led him to the conclusion that “the saying of even a simple series of words interferes with the understanding of a fresh message” (1952:€271). This example hints at what may be gained by viewing one’s object of study from different intellectual (theoretical and experiential) premises. But it is still an example of the normal workings of the scientific process, within a given scientific community, in this case, psychology. Some members of that scientific community, after it had fully embraced the ‘cognitive turn’, maintained an interest in interpreting; others did not. The most prominent example of the latter is pioneer time-lag researcher Pierre Oléron, who, in a 1992 autobiographical essay, referred to his 1965 study with Nanpon as “une sorte d’aberration” (…) une sorte de contre-exemple par rapport aux recherches ‘normales’” (1992:€153). A prominent example of psychologists holding the opposite view is Giovanni Flores d’Arcais, who, in his contribution to the 1977 Venice Symposium, described (simultaneous) interpreting as ideally suited for psychological experiments: […] there are probably very few “real-life” situations which are more similar to a laboratory of psychological experimentation than the situation of the interpreter in a conference booth, both for the control of external variables and for the “artificiality” of the task in comparison to normal linguistic performance. (Flores d’Arcais 1978:€393)

Note that the focus of psychological interest is simultaneous conference interpreting as practiced in a booth. With the exception of some comparative work by Gerver, consecutive interpreting, which after all involves a formidable memory task, at least in its ‘classic’ form as practiced since the 1920s, was not seen by experimental psychologists as a phenomenon worth studying. Two issues emerge quite clearly from this brief review of the way influential psychologists related to interpreting: one is sustaining interest in interpreting as an object of study, and the other is defining that object in the first place. In either case, the underlying attitude could be summarized as “We study what we are interested in!” or “We study what matters to us!” The example of early research on interpreting by psychologists cuts both ways: on the one hand, interpreting was at least seen as an object of study (and interpreting studies owes significant initial input to those psychologists of the 1960s and 1970s); on the other hand, it was viewed very narrowly, leaving much of the phenomenon unexplored. Oléron, for instance, had published on deaf education since the late 1940s but showed no awareness of, or interest in, signed language interpreting. And as even consecutive interpreters’ feats of memory did not attract the attention of cognitive scientists, it is





Franz Pöchhacker

not surprising that the more mundane forms of interpreting remained ignored and unseen. And not only by psychologists, as the following example will show.

2.2

Seen by sociologists – or not

As early as the 1970s, one of the most accomplished and inspiring representatives of the field of sociology, Aaron Cicourel, was doing pioneering work on language in medicine. In a review article published in 1981, he discussed two medical interviews from his own empirical work. One of them involves a 15-year-old Mexican boy suspected of suffering from epilepsy. The patient is accompanied by his mother and uncle, with the latter serving as ad hoc interpreter for the former. The uncle’s command of English is quite limited, however, so the interview proceeds with great difficulty. This is demonstrated in Cicourel’s transcript, where the uncle is unable to render the Spanish term “sarampión” and is in fact assisted by the researcher, presumably Cicourel himself, who is fluent in Spanish. The researcher is thus directly aware of the issue of translation and of the ad hoc interpreter’s insufficient competence. Here is a brief excerpt, adapted from Cicourel (1981:€424f), to illustrate what we now know are typical features of ad hoc interpreters’ performance, such as reduced renditions and an active third-party approach to the task. (18) D: Was it the three-day or ten-day measles? (19) U: ¿No te acuerdas cuánto te duró? ¿Tres días o diez días? Do you remember how long it lasted? Three days or ten days? … (27) U: Duró mucho tiempo? Did it last a long time? (28) M: Y dijo el doctor que si no me quitaba la fiebra me moría yo y (U: uh huh) And the doctor said that if I couldn’t get rid of the fever I would die and Que me vi mal (a?) con el sarampión. That I (saw myself?) was (really) sick with the measles. (29) U: Sí pero. she feel. sick through that time. Yes but. The doctor’s question (18) is rendered by the Uncle in a somewhat more personal style (19) and taken up again several turns later, after the Mother had failed to give a precise answer. The Mother again describes her illness rather than specifying its duration (28), and the Uncle reconsiders his urge to press her for an answer (“Yes, but...”) and renders her utterance with considerable abridgment. And yet, what looks like an early example of discourse-based research on interpreting in healthcare is not about interpreting at all. Cicourel is particularly interested in the written record of the interview and, from this perspective, concludes that nothing is said in the medical report about the difficulties of conducting the interview and



Why interpreting studies matters

obtaining accurate information. His analysis of the mediated communication is very succinct: “Some of the communication gets a little garbled in the translation.” (Cicourel 1981:€423) Towards the end of his analysis, after pointing out that “[t]here are many episodes that could be described as ‘hilarious’”, Cicourel also mentions media reports about “communication barriers” which result in children not receiving adequate medical attention. But “interpreting” as a phenomenon that was taking place, or that might have taken place differently, with a different outcome for the communicative encounter, remains unseen – and unstudied. It is true that those were early days for healthcare interpreting, and for research on it. (There are only a few substantial papers on this topic from the 1970s.) Much has changed in this field, as we know, but not the fact that interpreting may remain invisible in sociological or sociolinguistic studies on communication in healthcare. In the inaugural issue of the journal Communication & Medicine, for instance, a paper by the selfsame Aaron Cicourel (2004) analyzes “cognitive overload and communication” in healthcare. In this article, which is part of a larger study of three US healthcare settings, the focus is on the workplace routines of nonmedical personnel in charge of scheduling appointments (and much more). And why not. But the issue of language barriers does not feature. The issue of “cognitive overload” that commands so much attention in interpreting research is not applied to interpreters charged with facilitating cross-cultural communication but to administrative hospital staff. My intention here is of course not to fault Aaron Cicourel for not being an interpreting scholar (though our discpline could certainly use him only too well). On the contrary, the point is that a given research community is entitled – and expected – to cultivate its particular parcel of scientific territory, be it as extensive as human interaction and coexistence, or as specific as mediated real-time communication across languages and cultures. Though only in relatively recent times, interpreting scholars have come to stake out their claim on the scientific landscape for the study of interpreting. By doing so, they are addressing a basic human (epistemological) need, taking charge of a phenomenon at the intersection of language, cognition, interaction and culture that is socially relevant and therefore clearly worth studying. In my discussion so far I have attempted to show the difference between having ‘scientific research on’ vs having ‘a science for’ interpreting. To put my conclusion that the latter matters in a nutshell: If interpreting studies did not exist, it would be necessary to invent it! Even so, the epistemological rationale illustrated here may not be considered sufficiently convincing in our increasingly utilitarian, economicized world in which it is hard to avoid questions about the ‘practical use’ or ‘benefit’ of what one is doing. Having asserted that interpreting studies matters, we would be challenged to state, economically speaking, what sort of a market there is for the output of interpreting scholars. In short: interpreting research – cui bono?





Franz Pöchhacker

3. To whom it matters 3.1

Interpreter education

It is fair to say that interpreting studies largely owes its present position in academia to the very practical concern of training qualified professionals, which has been institutionalized at university level since the 1940s. While much development in interpreter training was achieved without the benefit of scientific research, the idea that a more thorough understanding of the interpreter’s task and skills would facilitate teaching and learning has never been in doubt, making interpreter trainers the prime consumers of insights gained from research. But the link between research findings and actual teaching practices in the interpreting classroom is rather tenuous. Daniel Gile (1990:€33), referring specifically to training-oriented scientific research, concluded that [...] it does not seem to have had any significant effect on training methods and results except in courses given by the researchers themselves, and sometimes in the schools where they teach, but on the whole, interpretation instructors prefer to keep their personal, most often traditional methods, and take no heed of research.

My survey among fellow teachers of simultaneous interpreting at the University of Vienna (Pöchhacker 1999) sadly bore out Gile’s point nearly a decade later. In recent years, there has been considerable progress in training-oriented research, with studies (usually at the doctoral level) on strategy use by trainee simultaneous interpreters; on note-taking practices in consecutive interpreting; on the role of preliminary exercises in SI training; or on the assessment criteria applied to final-exam performances. But Gile’s doubts may still linger. Aside from “training of trainers” initiatives like Barbara Moser-Mercer’s course in Geneva, little is known about the actual uptake of research findings by the teaching staff of interpreter training institutions.

3.2

Interpreting research community

Rather than teachers, the prime users of research output in interpreting studies seem to be other interpreting scholars (who, incidentally, may often be teachers trying to secure a university position by earning a doctoral degree). That research should matter mainly to other researchers is in fact quite normal and as it should be, science being a collective endeavor in which individual contributions build on, or at least ‘depart from’ the state of the art, adding to it and becoming part of it. This self-referential view of where, to whom and how interpreting studies matters is perfectly in line with the field’s raison d’être as stated earlier – that of cultivating a particular parcel of scientific territory. Still, as interpreting is very much a real-life social practice, the view of researchers doing research on interpreting for the benefit of other interpreting researchers is likely



Why interpreting studies matters

to conjure up the image of the ivory tower, of people who fail to connect with the ‘real world’ surrounding them, that is, the world of professional practice.

3.3

Practice and profession

So the other major domain in which one might expect interpreting studies to matter is the professional practice of interpreting. And does it? My answer would be two-fold, depending on the timeline along which professional practice and research have evolved. Much of the (twentieth-century) history of professional interpreting suggests that in this field ‘research follows practice’, or even that practice can do without research. Examples include the 1920s trials with SI without any researchers’ guidance; the switch from consecutive to simultaneous in the UN, without conclusive research findings informing that decision; or, most pointedly, Gerver’s (1969) experimental finding that conference interpreters can comfortably handle an input rate of up to 120 words per minute, confirming the recommendation made by an AIIC meeting in the 1960s. In more recent times, one of the main challenges facing the interpreting profession has been remote interpreting. Here again, initial implementation efforts could hardly be said to have been based on research findings. At the same time, though, the topic of remote interpreting in international conference settings provides evidence to warrant an affirmative response to the question at hand: Increasingly, interpreting research does matter when it comes to responding to new professional challenges. The review of research on remote interpreting experiments by Mouzourakis (2006) is a case in point. On a much smaller scale, there is also the challenge posed to the traditional consecutive mode by the use of digital recording devices. SCIC interpreter Michele Ferrari, who pioneered this ‘simultaneous consecutive’ mode of interpreting in 1999, has called for, and carried out, research on the feasibility and broader adoption of this technique. His call has been heeded by some interpreting scholars (e.g.€Hamidi and Pöchhacker 2007), whose findings can be expected to inform further developments. For a well-known example of what might be seen as the profession following research, I would recall the study of user expectations, which was launched (unwittingly) by an individual scholar (H. Bühler), implemented in several small-scale studies by Ingrid Kurz (e.g.€1993) and ultimately taken up in a large-scale AIIC-sponsored research project (Moser 1996). An initiative that came from within academia thus developed into a concern among practitioners, as represented by AIIC, giving us a case of ‘research guiding the profession’. At this point we might also reflect on what the relationship between academic research and a profession should be like, for instance by looking at the situation in other professions. We usually like comparing ourselves with the medical profession, or the legal profession, but I would suggest that it is also appropriate to take, say, journalism and communication studies as a paragon. Do professionals there need the output of researchers? Do research findings always have an impact on, or a certain relevance





Franz Pöchhacker

for, the profession? The answer invariably brings us back to the university education of such professionals, where curricula and teaching content, if not also instructional practices, should be research-based and, through the newly graduated generation of practitioners, inform and shape professional practices. This affords the profession a means for the constant development and critical appraisal of its unique body of knowledge and expertise, and it is this mechanism of collective self-reflection that constitutes one of the hallmarks of a profession. These reflections on the relevance of research to the profession so far have focused on international conference interpreting in spoken languages. I would suggest, without arguing this point at length for lack of space, that research informing practice can also be seen in the field of signed language interpreting (e.g.€Harrington and Turner 2001), much of which is done in community-based settings. As regards the field of community interpreting in general, assessing the role of research for professional practice is more difficult: community interpreting as a profession is still emerging, if it exists at all, in various countries or for particular institutional domains. There is indeed an impressive flow of studies on what interpreters in community settings do, but to consider the impact of such research on ‘the profession’ and its affiliated practitioners is possible only where such a profession exists. A case in point is the focus group study by Angelelli (2006) on the CHIA standards of practice for healthcare interpreters. Though Angelelli in her article laments the fact that professional developments, such as the drafting and dissemination of a code of practice, are often not sufficiently informed and guided by research, the fact that CHIA commissioned her to do the study does suggest a healthy interaction between research and the profession. And yet, this example from the Californian context cannot be generalized to other states or countries, where professionalization is often lacking and where the role of research can therefore be discussed not so much in relation to the interpreters as with regard to the institutional contexts in which individuals serving as interpreters are used.

3.4

Institutional (user) contexts

To what extent and how, then, does (community) interpreting research matter to and address the needs of users in particular institutional contexts? A typical kind of research on community-based settings are studies to assess interpreting needs and communicative practices in a given domain. My survey among healthcare professionals in Vienna (Pöchhacker 2000) is a simple example of such research addressing a need, or rather, helping to actually address the need by supplying the necessary data and research findings. More sophisticated is the stocktaking study by Erik Hertog and Jan van Gucht on public service interpreting in Flanders (e.g.€Lannoy and van Gucht 2006). Aside from such ‘applied’ empirical studies yielding data for policy-makers, much research on dialogue interpreting in community settings, often by authors with a background in linguistics or discourse analysis, has focused on the sociolinguistic and



Why interpreting studies matters

interactional intricacies of interpreters’ behavior (e.g.€Mason and Stewart 2001). But to what extent has this type of research been of relevance in shaping a given social (and in it, communicative) practice? If we can assume that a social practice – such as (professional) interpreting – is informed by a body of basic and applied research, via training or through findings bearing directly on an aspect of practice, the same would hold also for other social practices, such as medical interviewing, police questioning, psychotherapy, or refugee status determination. In other words, there would be scientific communities producing basic as well as applied research to inform and guide the respective social practices. So, how, if we find it difficult enough for interpreting research to shape practices in our very own domain, can we hope to have an impact on practices in a different social institution, such as, broadly, medicine, law enforcement or the judiciary? Within the realm of science, the logical path is to cross over with our research findings to the other discipline(s), make them part of ‘their’ literature and, by influencing ‘their’ state of the art, have an impact on relevant social practices. If we manage this, research alone will be sufficient, and we may simply sit back and be cited. Such interdisciplinary cross-fertilization, however, is easier said than done. Indeed, it can be surprisingly difficult for interpreting researchers to gain access to the relevant fora (conferences, journals) of other disciplines. While a natural side-effect of increasingly differentiated and specialized scientific paradigms, the various barriers to extradisciplinary publication greatly limit the potential impact of our research on interpreting as practiced in certain institutional contexts. As the problem of ‘translating’ our research findings into the published canon of other fields of knowledge will most likely persist, our ambition to matter to those controlling the social practices in which interpreting is embedded will have to (also) take a different path. This, I suggest, has relatively little to do with publishing papers based on research. It is not ‘R’, but ‘D’ – the kind of development work that is done when interpreting scholars participate in committees or interdisciplinary working groups or projects designed to change a given social practice. It is in such time-consuming contacts that research-based expertise stands a good chance of being accepted and allowed to have an impact, but not so much through the dissemination of publications as through a translational process that relies on human qualities like openness, flexibility, commitment and communicative skills. In other words, in such development efforts it is not interpreting studies that matters, but the people who promote and leverage its findings in social contexts, ideally by moving toward collaborative research. While the efforts invested in ‘development’ are often not apparent from the published literature, I will conclude this discussion with a few examples based on experiences shared by colleagues. An impressive case of community-based and institutional networking leading to high-impact research is Jemina Napier’s study for the New South Wales Law Reform Commission proving the effectiveness of sign language interpreting for Deaf jurors (Napier and Spencer 2006). No less groundbreaking and influential (and labor-intensive), Miriam Shlesinger’s initiatives to establish and





Franz Pöchhacker

professionalize healthcare interpreting services in Israel exemplify the preponderance of much-needed development efforts over research per se (Shlesinger 2007). Last, but not least, and closer to home, mention should be made of Sonja Pöllabauer’s achievements for interpreting in Austrian asylum proceedings: Following her doctoral work on first-instance asylum hearings in Graz (Pöllabauer 2005), she joined an interdisciplinary working group that drafted “Minimum Procedural Standards for Interpreter Use in Asylum Hearings” as input for the Interior Minister and lawmakers preparing to amend the Austrian Asylum Act. The fact that the conservative government of the day (a coalition including the Haider party) did not take these proposals on board was disappointing, but it cannot detract from the ‘development work’ that was accomplished. These are only a few examples of ‘R & D’ in interpreting studies, focused on community interpreting, where, by definition, institutional contexts largely constrain communicative (and interpreting) practices. Since interpreting is invariably a situated social activity, though, the message here should apply to other domains of interpreting as well.

4. Conclusion Taking up the challenge of the Ljubljana Congress theme for the (sub)discipline of interpreting studies, I have argued that interpreting studies matters fundamentally for being the only scientific community to claim and take charge of interpreting as its unique object of study. This means that interpreting studies mattered even before the growing interest in community-based interpreting came to increase the field’s perceived social relevance, as discussed for such domains as interpreter education and professional practice. With regard to practices in institutional user contexts in which community interpreting is embedded, I have suggested that research alone may not matter enough to shape the (largely underdeveloped) situation on the ground. Unless interpreting scholars manage to publish their findings in the journals of disciplines that underpin the relevant social practice, be it medical interviewing or refugee status determination, they may need to complement their ‘R’ by collaborative ‘D’ in order to secure a certain impact. Thus, to the extent that interpreting research is intent on describing and explaining its object of study also to stakeholders in other institutional contexts, interdisciplinarity in research is always a goal, but ‘development’ efforts in practice seem no less important to ensure that interpreting studies matters.

References Angelelli, C.V. 2006. “Validating professional standards and codes: Challenges and opportunities.” Interpreting 8 (2): 175–193.



Why interpreting studies matters Broadbent, D.E. 1952. “Speaking and listening simultaneously.” Journal of Experimental Psychology 43: 267–273. Cicourel, A.V. 1981. “Language and medicine.” In Language in the USA, C.A. Ferguson and S.B. Heath (eds), 407–429. Cambridge: Cambridge University Press. Cicourel, A.V. 2004. “Cognitive overload and communication in two healthcare settings.” Communication & Medicine 1 (1): 35–43. Flores d’Arcais, G.B. 1978. “The contribution of cognitive psychology to the study of interpretation.” In Language Interpretation and Communication, D. Gerver and H.W. Sinaiko (eds), 385–402. New York/London: Plenum Press. Gerver, D. 1969/2002. “The effects of source language presentation rate on the performance of simultaneous conference interpreters.” In The Interpreting Studies Reader, F. Pöchhacker and M. Shlesinger (eds), 53–66. London/New York: Routledge. Gerver, D. 1971. Aspects of Simultaneous Interpretation and Human Information Processing. D. Phil. thesis, Oxford University. Gile, D. 1990. “Scientific research vs. personal theories in the investigation of interpretation.” In Aspects of Applied and Experimental Research on Conference Interpretation, L. Gran and C. Taylor (eds), 28–41. Udine: Campanotto. Goldman-Eisler, F. 1967. “Sequential temporal patterns and cognitive processes in speech.” Language and Speech 10 (3): 122–132. Hamidi, M. and Pöchhacker, F. 2007. “Simultaneous consecutive interpreting: A new technique put to the test.” Meta 52 (2): 276–289. Harrington, F.J. and Turner, G.H. (2001) Interpreting Interpreting: Studies and Reflections on Sign Language Interpreting. Coleford: Douglas McLean. Kurz, I. 1993/2002. “Conference interpretation: Expectations of different user groups.” In The Interpreting Studies Reader, F. Pöchhacker and M. Shlesinger (eds), 313–324. London/New York: Routledge. Lannoy, K. and van Gucht, J. 2006. “Babel rebuilt: A survey of social welfare institutions and interpreting and translation services in Flanders.” In Sociocultural Aspects of Translating and Interpreting, A. Pym, M. Shlesinger and Z. Jettmarová (eds), 191–200. Amsterdam/ Philadelphia: John Benjamins. Mason, I. and Stewart, M. 2001. “Interactional pragmatics, face and the dialogue interpreter.” In Triadic Exchanges, I. Mason (ed.), 51–70. Manchester: St. Jerome. Moser, P. 1996. “Expectations of users of conference interpretation.” Interpreting 1 (2): 145–178. Mouzourakis, P. 2006. “Remote interpreting: A technical perspective on recent experiments.” Interpreting 8 (1): 45–66. Napier, J. and Spencer, D. 2006. “A sign of the times: Deaf jurors and the potential for pioneering law reform.” Reform 90: 35–37, 74. Oléron, P. 1992. “Autobiographie.” In Psychologues de Langue Française. Autobiographies, F. Parot and M. Richelle (eds), 135–159. Paris: Presses Universitaires de France. Pinter, I. 1969. Der Einfluß der Übung und Konzentration auf simultanes Sprechen und Hören. Doctoral dissertation, University of Vienna. Pöchhacker, F. 1999. “Teaching practices in simultaneous interpreting.” The Interpreters’ Newsletter 9: 157–176. Pöchhacker, F. 2000. “Language barriers in Vienna hospitals.” Ethnicity & Health 5 (2): 113 119. Pöllabauer, S. 2005. “I don’t understand your English, Miss.” Dolmetschen bei Asylanhörungen. Tübingen: Gunter Narr. Shlesinger, M. 2007. “Making the most of settling for less.” Forum 5 (2): 147–170.



What matters to Translation Studies? On the role of public Translation Studies Kaisa Koskinen

University of Tampere, Finland

In this article, the field of Translation Studies is mapped with the help of a matrix developed by Michael Burawoy for sociology. His four fields of professional, critical, policy and public sociology are used to chart similar trends in TS. Burawoy argues for a more visible engagement in public sociology. This paper argues for a similar focus on public Translation Studies. Keywords: sociology, social relevance, academic activism, Public Translation Studies

1. Introduction The growth of Translation Studies (TS) as an independent discipline during recent decades has been a success story: new journals, publication series, training programmes, academic events and research projects proliferate. However, the recent crisis in the humanities amidst the commodification and neoliberalisation of higher education has, at least in my own country, Finland, created an unexpected backlash. Faced with a need to increase productivity and cut down personnel costs, several Finnish universities are planning or already executing structural reforms where the longstanding independence of translator training (which has just had its 40th anniversary) is at risk, and severe cuts have been proposed. It seems that the apparent success of TS had gone largely unnoticed by several neighbouring disciplines, and the idea of the separate training mission and research agenda we had cherished was not regarded as a valuable asset either by our colleagues or by the university administration. The success story seems to be more of an internal narrative than an actual achievement which has been externally recognised. At the same time, the rapidly changing field of the language industry is presenting us academics with constant challenges, as we try to keep up with the developments. Professional practices are evolving in new directions, and stretching and even breaking the boundaries of translation. Firms involved in localisation, subtitling, transediting and many other activities often make a point of not using the term ‘translation’ at all.



Kaisa Koskinen

The new constellations in the field have often resulted in translators ending up in less than satisfactory working conditions (Abdallah and Koskinen 2007). Research in TS has to some extent kept up with the developments, widening its focus to follow the latest moves (Delabastita 2003), but research has seldom taken the initiative. We may have been able to observe the developments, but perhaps we have been less successful in using research-based interventions for directing the activities within the field (see, however, Pöchhacker’s article on Interpreting Studies in this volume). This feeling of a less than perfect communication of our aims, values and research results – either beyond the disciplinary boundaries within academia or within the professional field of translation and language mediation called by other names – raises some important questions: Are there some inherent weaknesses in TS, in spite of its apparent successes, and if so, can these be re-enforced? Is the discipline too narrowly focused? Does it have blind spots that need to be covered? What matters to Translation Studies as a discipline? Does it, or should it, care about the practitioners? Does it support and defend translator training? Do we reach out from or own academic niche? Are we willing and able to take social responsibility? These are not simple questions. But Translation Studies is not alone in searching for its role in the new constellation of public and private interests. Many disciplines are presently asking similar questions. In this article, the state and status of Translation Studies will be discussed with reference to recent debates about the task of another discipline, sociology.

2. Division of academic labour In his much-debated presidential address to the American Sociological Association in 2004, Michael Burawoy (2005) divides the field of sociology into four subfields, each characterized by its particular combination of audience (academic vs. extra-academic) and type of knowledge (instrumental vs. reflexive). The matrix of the four subfields of professional, critical, policy and public sociology is shown in Table€1. According to Burawoy, these four subfields are “antagonistically interdependent” (2005:€4). That is, their tasks and values are different, but the field is not properly functional without them all. In a perfect world the flourishing of each type would be the precondition of the flourishing of the others, but Burawoy warns us that each subfield can also develop into a pathological form, or become excluded or subordinated by the others. Table 1.╇ Division of academic labour. Source: Burawoy 2005

Instrumental knowledge Reflexive knowledge

Academic Audience

Extra-academic audience

Professional Critical

Policy Public



What matters to Translation Studies 

Professional sociology is the sine qua non for all the other types. This subfield is the “pure” or “hard”, “scientific” field of empirical and theoretical explorations, guided by scientific norms and assessed by peer review. It is the source of tested methods, accumulated knowledge, orienting questions and conceptual frameworks, i.e. of expertise and legitimacy for both policy and public sociology (Burawoy 2005:€10). In a pathological form, it can become wrapped in itself and get caught in methodological fads and trivial questions. Instrumental knowledge cannot thrive for long without the challenges posed to it by reflexive types of knowledge (ibid.: 19). Its values and research agendas need to be constantly re-examined and renewed. This is the task of critical sociology, the heart and the collective conscience of the discipline. Critical sociology has a strong ethical undercurrent, and it is geared towards a critical reflection on the foundations of the professional subfield (ibid.: 10–11). Both of the two subfields described above primarily address academic audiences. The particular strength of Burawoy’s matrix, however, lies in the other vertical axis. The explicit inclusion within the disciplinary boundaries of the two subfields addressing extra-academic audiences is a key to a new understanding of the varied roles and tasks of academics. As a social science, sociology has traditionally had close ties to decision-making bodies. Policy sociology provides solutions for the problems defined by clients. It is pragmatic and concrete, and in the service of the community (Burawoy 2005:€9). It is to be distinguished from public sociology, where the sociologist and the various publics enter a dialogic relation, and their agendas are discussed and adjusted (ibid.). Burawoy’s address, titled “For public sociology”, is in fact a defence of this particular subfield addressing (or rather, conversing with) extra-academic audiences in a reflexive manner. Traditionally, public sociologists have, for example, raised sociological issues in newspaper articles or popularised research findings for lay people. Training the new generation of sociologists also falls into this category in Burawoy’s matrix; students are the “first and captive public” of sociology (ibid.: 7). Organic public sociology, then, brings the sociologist into active cooperation with a thick, active local group (e.g.€action research, interventions). This kind of public sociology often remains invisible, as those involved are too busy doing public sociology to have time to write academic articles about it, and it is also often considered to fall outside the discipline because of its practice-orientation (ibid.: 7–8). Burawoy makes a case for a more visible inclusion of extra-academic audiences. The time has come, he argues, to embark on a systematic back-translation of the professional knowledge accumulated in sociology, “taking knowledge back to those from whom it came, making public issues out of private troubles, and thus regenerating sociology’s moral fiber” (ibid.: 5). Burawoy’s proposal for a more engaged and more public sociology has been widely debated.1 Its viewpoint is American, and it is also a necessarily schematic presentation of a complex field with many overlaps, dialogic 1. For an overview of Burawoy’s papers on public sociology and numerous critical commentaries of them, see https://webfiles.berkeley.edu/~burawoy/workingpapers.htm.



Kaisa Koskinen

exchanges and fuzzy borders. It is not always entirely clear where the borders are to be drawn, especially since Burawoy argues that each quadrant can be further divided along the same axis (i.e. that for example professional sociology can be further divided into pure professional sociology, critical professional sociology, public professional sociology and policy professional sociology). But the particular value of the matrix for the present purposes is that it provides a pragmatic scheme that can be used for analysing how any discipline is positioned in relation to its various potential functions. It also offers at least tentative answers to the difficulties described above: should it turn out that Translation Studies as a discipline has developed towards what Burawoy calls a “pathological” state, caused by an imbalance between the fields, conscious efforts to redirect the course can be made.

3. Division of labour in Translation Studies Translation Studies is often described as an interdiscipline (Snell-Hornby 1988), emphasizing its origins in a number of disciplines and its numerous borrowings from various sources. The discipline has a somewhat messy image, and its academic itinerary is convoluted. In recent years, another kind of narrative has also gained currency. It can be interpreted as an attempt to create coherence in the sometimes chaotic field, both by finding some shared ground between the opposing poles of “empirical” and “postmodern” research models (Chesterman and Arrojo 2000; see also Delabastita 2003) and by acknowledging the sometimes ill-reconcilable differences between the two main paradigms perceived in TS: the Liberal Arts Paradigm (LAP) and the Empirical Science Paradigm (ESP) (e.g., Gile 2006).2 These dualistic models are interesting from the point of view of Burawoy’s matrix since they seem to be charting a division of labour resembling the one between professional and critical sociologies. The proponents of the “empirical” school emphasise rigorous research, testable hypotheses and empirical testing as well as scientific norms and objectivity in a manner quite similar to professional sociology (see, e.g., Chesterman 2005:€198–200; cf.€Gile 2008). It would seem quite logical to include the ESP within “professional Translation Studies”. The proposed counterpart of the empirical paradigm is a more problematic case. There probably exists a unanimous agreement that not all work in TS falls within the scientific model of hypothesis-testing and causal predictions, but what this “other” of ESP actually consists of is less clear. Many trends in TS, from postcolonial studies to queer theory and beyond, follow research agendas and rationales different from the one valued in ESP, and not all research questions lend themselves to repeatable empirical testing. But is there a unified “postmodern” or Liberal Arts Paradigm in TS? And if there is, how far removed is it from its alleged polar opposite of ESP? 2. Interestingly, this latter division has mainly been promoted by the EST, in its research forums and on the EST website.



What matters to Translation Studies 

Although prominently portrayed, the ESP agenda alone is in fact far too narrow to cover sufficiently the entire field of TS research in order to provide instrumental knowledge for all the other subfields. The number of translational phenomena one can study by empirical testing is limited. The list of names Daniel Gile (n.d.) characterises as ESP authors is indicative: it largely consists of researchers interested in translators’ or interpreters’ mental and cognitive processes. Research in the linguistic aspects and the history of translation is often conducted within Descriptive Translation Studies (DTS), arguably the most popular research paradigm in TS during recent decades, both for those interested in literary translation and for those interested in corpus studies. The ethos of DTS is, as the name implies, against prescriptiveness, and similar to ESP it values empirical data, objectivity and rigorous analysis (distinguishing between the two is, in fact, rather a matter of interpretation). In Gile’s categorization (ibid.), prominent DTS figures such as Gideon Toury or Theo Hermans are, however, placed in the opposing camp of the liberal arts paradigm.3 One of the problems is that the proposed LAP paradigm has mainly been defined by commentators who are not personally affiliated with it, and the descriptions are perhaps for that reason somewhat sketchy. It also follows that it is largely defined by what it is not, i.e., the ESP: LAP research, according to Gile (n.d.), is more theoretical than empirical; it is not primarily based on data analysis; it does not aim at verification of theories or require facts to substantiate claims but can proceed intuitively; it does not value objectivity and is not opposed to prescriptiveness. The resulting picture partly resembles a caricature of bad research. Surely, rigorousness, fact-based argumentation and working with data can and should be expected from serious research within any paradigm. This is also, I think, the greatest achievement of ESP: together with its twin paradigm of DTS, the input provided by its adherents has greatly advanced the coherence, rigour and cumulativeness of all TS research, regardless of the paradigm. The other side of the coin is that the “ban” on prescriptiveness resulting both from the scientific ethos of ESP and from the historical emphasis on describing, not prescribing what the translators do in DTS may have participated in preventing Translation Studies from fully exploiting its reflexive and critical potential in pragmatic terms, in spite of the extensive input from various cultural studies approaches. The scientific ethos, and the perception of TS as a wannabe hard science, together with a particular understanding of what empirical research can be, has also resulted in a persistent disciplinary inferiority complex. If we took Burawoy’s four fields as our starting point, the picture of TS would be drawn somewhat differently, and the duality 3. There is often a marked difference in TS researchers’ own empirical research and the research paradigms they propagate for (I wish to thank the anonymous reviewer for drawing my attention to this point). For example, the seminal work of Gideon Toury often functions as the backbone of ESP research projects, but his own research might be more appropriately categorized as LAP. Similarly, Andrew Chesterman has vigorously argued for ESP research, with a strong emphasis on causality and hypothesis-testing. He has, however, not engaged in such enterprises himself (see also Gile 2004:€16).



Kaisa Koskinen

would appear to be constructed on a different basis. In the matrix, the division of the vertical academic axis runs according to the amount of reflexivity and the role of ethics (see Table€ 2). In the matrix, the counterpart of professional TS is found in “critical Translation Studies”. This subfield contains all metatheoretical critical investigations examining the foundations of TS (thus arguably including both this article and many of those cited here), and also a large proportion of ethically motivated critical approaches such as feminist or postcolonial studies (see also Koskinen 2004). In this scheme, it is the moral duty of critical Translation Studies to initiate debates within TS and to spark discussion about the ethical values of research. Admittedly, research identified as LAP by Gile is more prone to moral statements and is thus partially – but not entirely – located in the critical TS camp in the Burawoy matrix. This subfield has been gaining ground in TS throughout the “cultural turn”, and various critical paradigms have been established. A prime example of a “CTS” approach to research can be found in Rosemary Arrojo’s statement of what a theory of translation should be for (in response to Andrew Chesterman’s equally programmatic ESP view that a theory should be able to make predictions) (Chesterman and Arrojo 2000:€159): Instead of trying to make predictions, a theory of translation should attempt to empower translators-to-be and raise their conscience as writers concerning the responsibility they will face in the seminal role they will play in the establishment of all sorts of relationships between cultures.

Chesterman and Arrojo’s article is an example of an attempt to bridge the gap between approaches (for a similar attitude, see also Gile 2006). One can also discern a worry over the fragmentation, incoherence or non-cumulativeness of TS in these accounts, and the desire to create a duality may be interpreted as an attempt to systematise the field. The “LAP” camp may not find the fragmentation equally problematic (cf.€Burawoy 2005:€17–18), which may partially explain the lack of programmatic statements of a joint ethos among the various branches. Similar to sociology, and all humanities, in TS the “overwhelming movement has been in a critical direction” (Burawoy 2005:€7), and this has perhaps added to the anxiety of ESP proponents. As Burawoy’s matrix indicates, an over-emphasis of this particular subfield can indeed be problematic. In its pathological form, critical TS can become dogmatic and lock itself into an antagonistic position with respect to professional TS, thus hampering rather than enhancing the flourishing of the field. Critical TS needs a multifaceted and thriving professional TS. The moral and reflexive task performed by critical Translation Studies evokes subjective and prescriptive statements that may be interpreted as “unprofessional” or “unscientific” by those located in professional TS. Burawoy’s terminology is perhaps slightly misleading, as it seems to imply that only one of the subfields is “professional”. In Table€2, an alternative title of “scientific TS” is also used to indicate that subjective statements, moral stances and intuitive comments which can be and no doubt are fruitfully employed by (TS) scholars are best interpreted in the framework of critical



What matters to Translation Studies 

Table 2.╇ Division of labour in TS (adapted from Burawoy 2005) Academic audience

Extra-academic audience

Instrumental knowledge Knowledge Legitimacy Responsibility Politics Pathology

Professional (scientific) TS theoretical/empirical scientific norms peers professional self-interest self-referentiality

Policy (pragmatic) TS concrete effectiveness clients/patrons interventions servility

Reflexive knowledge Knowledge Legitimacy Responsibility Politics Pathology

Critical TS foundational ethical values critical intellectuals internal debate dogmatism

Public TS communicative relevance designated publics public dialogue faddishness

TS, a subfield not primarily governed by the scientific norms of empirical research but equally subject to the professional norms of academic argumentation. It seems plausible to argue that in Translation Studies both scientific and critical subfields are alive and well (see, however, Gile and Hansen 2004). The recent discussions also indicate that there are conscious efforts towards a dialogic interaction between them, making development towards a pathological form less likely. But how has TS catered for its extra-academic audiences? Answers are harder to find, since these issues may not be documented in academic publications. In Finland, there are some success stories. For example, following her dissertation (2005) on authorized translation (i.e. a contribution to scientific TS), Kaarina Hietanen participated in a working group with the result that new improved legislation was passed and a new, improved authorisation system has been implemented (policy TS). Similarly, in community interpreting the small Finnish research community is actively involved in steering the practice. But are these isolated cases, or do we often have a say in policy decisions related to TS? Or, do we make ourselves heard? It remains to be seen, for example, whether the TS community will have much say in training reforms (in addition to national reforms such as in Finland, there is the on-going Bologna process and the EMT network). It seems that at least in Finland we have not yet been very successful in creating the publics that would take us on board in policy work (see Burawoy 2005:€8). For example, when the Ministry of Education set up a large and influential national task force to reconsider and forecast the needs for all kinds of language training in Finland in 2006, no representative of translator training was invited to join the group (KIEPO 2007). Those setting up the working group simply never came to think about translator training (this did not prevent them from making statements and forecasts about it, though).



Kaisa Koskinen

The public sector, at the municipal, national and supranational levels (see Koskinen 2008: Ch. 2), is facing increased challenges of multilingualism, with its concomitant translation and interpreting needs. There thus is, at least potentially, a growing market for research-based consultation and benchmarking for best practices. Different from sociology however, where connections to the state apparatus have traditionally been strong (in particular in the Nordic countries) several potential clients of policy TS may also be found in the private sector. This subfield could therefore also be called “pragmatic Translation Studies”, to better accommodate for cooperation projects with the actors from the language industry and other “stakeholders”. I suspect many training institutions do have pragmatic networking projects (traineeships, visiting lecturers, real translation and interpreting assignments, etc.), but these are often constructed in order to bring pragmatic knowledge to the students. The idea of policy TS in the Burawoy matrix is to do the opposite, bringing the instrumental knowledge generated by scientific TS to the professional field. In the best of cases, this dialogic knowledgesharing can lead to a mutually rewarding experience where new knowledge is generated in the process, creating a shift from policy TS to public TS. Located along the instrumental horizontal axis, policy TS cherishes concrete and effective interventions, and the pragmatic aims easily override the more critical reflections. In a pathological form, it is prone to a servile attitude. That is why it requires public TS to function as its conscience. But do we have a flourishing subfield of public Translation Studies?

4. Public translation studies According to Michael Burawoy, this present era, which he calls “the third wave of marketisation”, calls forth the age of public sociology. Public sociology, says Burawoy, maintains that “science without politics is blind, that critique without intervention is empty – [it] calls on sociologists to engage directly with society before it disappears altogether” (2007:€357). Burawoy identifies three phases during the hundred years’ history of sociology, with some uncanny resemblances to discussions within TS. First there was the speculative and utopian phase of the founders, strongly imbued with moral concerns. The second phase was of positive science and empirical research. As a reaction to the earlier phase, this positivist science wanted to expel moral questions from the realm of science entirely (Burawoy 2007:€363–364). Burawoy argues that our time of the commodification of everything (Burawoy’s background is in Marxism, hence the vocabulary) is taking offensive against labour rights and social rights and eroding civil society. It calls for a new kind of science, and this third phase is that of public sociology: It calls for a science that seeks to develop alternative values, hence the importance of value discussions, what I have called critical sociology. We no longer strive for a single paradigmatic science but a discipline made up of multiple intersecting



What matters to Translation Studies 

research programs, founded on the values of different publics, but, at the same time, working out theoretical frameworks through engaging their external anomalies and internal contradiction. I call this reflexive science, a science that is not frightened of reflecting on its value foundations nor of articulating them publicly, but science nonetheless. (Burawoy 2007:€365)

We can discern the workings of marketisation and commodification in the fields related to translation as well. Universities are evolving further and further away from the Humboldtian ideals, and their status as a “public good” is becoming questionable. Commodified programmes, increased productivity requirements both in research and degrees awarded, and business-oriented restructuring of universities are all background factors for the present plight of translator training in Finland. As a latecomer to the academic scene of language programmes, it may become a casualty of this marketisation.4 Similarly, the profession has undergone a remarkable shift from a predominantly humanistic occupation into a globalised industry (Abdallah and Koskinen 2007). If we want to be involved in these changes, if we wish our viewpoints to be heard, we cannot be happy with developing one vertical axis of Burawoy’s matrix only. No matter how interesting findings scientific Translation Studies can unearth, no matter how sharply critical Translation Studies reflects on the foundations of TS research programmes and on the state of the art in the professional world, none of this is of much value unless we are able to communicate these beyond our own small cadre of TS scholars. Public Translation Studies, if it was designed along the lines of public sociology, would need to begin by identifying and if necessary creating the relevant publics. In addition to reaching out to the general audiences unidirectionally via the opinion pages of newspapers and popularised books, it could also include more dialogic and engaged forms. For example, actively reaching out for a sustained dialogue with neighbouring disciplines might have spared us some misunderstandings. Over the years, we may have been too complacent in creating our own little nest. Perhaps there have also been attempts at dialogue, with little success, leaving us disheartened. But as a discipline we cannot matter to others unless we communicate with them. Considering the recent changes in the profession, together with the declining working conditions reported by many practitioners, we might also find a responsive audience from among the professionals. Starting from where they are, not where we scholars are (cf.€Burawoy 2005:€9), we can re-enforce scientific Translation Studies by bringing in the added impetus of pressing social issues, and we can “back-translate” the accumulated knowledge to the practitioners. 4. At the time of the revision process (June 2008) the worst turmoil around Finnish translator training (hopefully) is over, and it now seems that most programmes will prevail, albeit within new institutional structures and with somewhat diminished resources. To an extent, there may even be beneficial consequences as we have been forced to rethink our aims and values and to restructure our programmes accordingly.



Kaisa Koskinen

No doubt there has always been a public face to TS. With its roots in the practicalities of translation and translator training, TS has never really been in an ivory tower. But public TS remains largely invisible. The organic dialogic interventions we scholars may engage in seldom get reported in academic fora, nor are they recorded for academic merits. Burawoy argues that in sociology this needs to change. The same might well apply to Translation Studies. I find Burawoy’s matrix useful precisely because its categorisation makes visible also those aspects of the academic division of labour that tend to be easily obscured or labelled as non-academic or extra-curricular. The present instability both in universities and in the professional field make his call for increased public engagement resonate in Translation Studies as well as in sociology. The matrix can be of assistance in assessing whether there is a good balance between the four types of academic labour. Among the questions we should ask are the following: Is the public image of TS within and beyond academia as good as it might be? Is the profession thriving? Are we in touch with recent developments? Are the effects of translation on various groups understood and taken into account in translation policies? Is there enough critical discussion within and beyond TS about the goals and values of research? Answers to these questions indicate that there is room for public Translation Studies. Translation Studies as a discipline cannot focus on scientific progress only. All four fields need to be addressed.

5. Academic activism? Compared to sociology, Translation Studies is a miniature discipline. In the American Sociological Association alone, there are 14,000 members. According to Burawoy, most of them only occupy one quadrant of the matrix at a time (2005:€14). In contrast, the success of policy and public Translation Studies may depend on the ability of TS scholars and translator trainers to occupy multiple locations, either in succession or simultaneously. This is a tall order, and there are obvious limits to what an individual scholar can achieve. It is not without irony that as I am writing this article, there is a draft opinion piece concerning the plight of translator training in Finland on my desk, waiting for my comments. My pursuits for addressing the academic audience are thus actively stalling my contribution to public Translation Studies. However, there is also a positive side to this lack of manpower. There is less risk of compartmentalisation, as individual scholars cannot be put into one box and labelled. Rather than having a fixed identity as professional/scientific TS scholars or public TS scholars, we may identify particular scientific, critical, policy and public momentums in our careers, and, hopefully, an enriching and rewarding interplay between the various subfields we can engage in. In spite of the obvious hindrances, the time may be ripe for a more explicit inclusion of an activist component of TS similar and parallel to activist translation identified and recently widely debated and discussed by academics (e.g.€ Tymoczko 2000; Baker 2007; Gambier 2007). Naming this particular approach “public Translation



What matters to Translation Studies 

Studies” might be used to validate and reinforce this impetus, to bring it into the limelight. Scholars in the field of Translation Studies need to ask themselves two fundamental questions (cf.€Burawoy 2005:€10−11): – Translation Studies for whom? – Translation Studies for what? These questions can and should be answered in a number of ways. Public Translation Studies is not a singular form but plural: the multiplicity reflects not only different publics but also different value commitments. This is where the fragmented and diversified nature of Translation Studies actually comes in handy: because of its heterogeneity, the discipline is better equipped to address multiple publics. Public Translation Studies has no (nor should have) intrinsic normative valence; its only commitment is to the dialogue around issues raised in and by the field it is engaged with (Burawoy 2005:€8). Do these issues matter to Translation Studies?

References Abdallah, K. and Koskinen, K. 2007. “Managing trust: Translating and the network economy.” Meta 52(4): 673–687. Baker, M. 2007. “Reframing conflict in translation.” Social Semiotics 17(2): 151–169. Burawoy, M. 2005. “For public sociology.” American Sociological Review 70: 4–28. Also available at https://webfiles.berkeley.edu/~burawoy/burawoy_pdf/2004PresidentialAddressASR.pdf (viewed 11.12.2007). Burawoy, M. 2007. “Public sociology vs. the market.” Discussion Forum, Socio-Economic Review, vol. 5: 356–367. Also available at https://webfiles.berkeley.edu/~burawoy/burawoy_ pdf/sociology_market.pdf (viewed 14.12.2007). Chesterman, A. 2005. “Causality in translator training.” In Training for the New Millenium. Pedagogies for Translation and Interpreting, Tennent, M. (ed.), 191–208. Amsterdam/Philadelphia: John Benjamins. Chesterman, A. and Arrojo, R. 2000. “Shared ground in Translation Studies.” Forum, Target 12(1): 151–160. Delabastita, D. 2003. “Translation Studies for the 21st Century: Trends and perspectives.” Génesis 3: 7–24. Gambier, Y. 2007. ”Réseaux de traducteurs/interprètes bénévoles”. Meta 52(4): 658–672. Gile, D. 2004. “Translation Research versus Interpreting Research: Kinship, differences and prospects for partnership”. In Translation Research and Interpreting Research: Traditions, Gaps and Synergies, C. Schäffner (ed.), 10–27. Clevedon: Multilingual Matters. Gile, D. 2006. “The liberal arts paradigm and the empirical science paradigm.” (22.1.2006) http://www.est-translationstudies.org/ (viewed 11.12.2007). Gile, D. 2008. “Hypotheses and research questions in empirical TS research.” (9.2.2008). http://www.est-translationstudies.org/ (viewed 23.6.2008). Gile, D. (n.d). “Initiation a la recherche empirique en traductologie.” http://www.cirinandgile. com/ESIT/1initrechempir.ppt (viewed 13.12.2007).



Kaisa Koskinen Gile, D. and Hansen G. 2004. “The editorial process through the looking glass.” In Claims, Changes and Challenges in Translation Studies, G. Hansen, K. Malmkjær and D. Gile (eds), 297–306. Amsterdam/Philadelphia: John Benjamins. Hietanen, K. 2005. Virallinen kääntäjä paljon vartijana. Ammattitoiminnan ja auktorisointimenettelyn yhteensovittamisen haaste (The Licensed Translator Safeguarding the Public Interest. The Challenge of Reconciling Professional Practice and Method of Authorisation), PhD thesis. Tampere: University of Tampere. KIEPO 2007. “The project Finnish Language Education Policies.” http://www.jyu.fi/hum/laitokset/solki/tutkimus/projektit/kiepo/in_english (viewed 14.12.2007). Koskinen, K. 2004. “Shared culture? Reflections on recent trends in Translation Studies.” Target 16(1): 143–156. Koskinen, K. 2008. Translating Institutions. An Ethnographic Study of EU Translation. Manchester: St. Jerome. Snell-Hornby, M. 1988. Translation Studies: An Integrated Approach. Amsterdam/Philadelphia: John Benjamins. Tymoczko, M. 2000. “Translation and political engagement: Activism, social change and the role of translation in geopolitical shifts.” The Translator 6(1): 23–47.

Translation and society

Translators as cultural mediators Wish or reality? A question for Translation Studies David Limon

University of Ljubljana, Slovenia

Translators are referred to with increasing frequency as ‘cultural mediators’ or experts in intercultural communication, but to what extent does translation practice confirm this description? Do most translators see themselves in this way and are they viewed as such by those who employ or commission them? Does the relative status of the translator within a particular society make it more or less likely that he or she is going to function as a genuine cultural mediator, or is this more dependent on institutional attitudes to translation or even on the type of translation being carried out? How does the translation brief (or lack of it) and the way the translation is evaluated influence the translator’s strategy? How significant are cultural differences with regard to text production, as well as the differing status of the two cultures involved? These are the kinds of questions that Translation Studies should be able to answer. They will be discussed here in relation to a specific cultural context and textual genre: the translation from Slovene to English of online texts about Slovenia’s natural heritage. Keywords: cultural mediation, translator strategy, explicitation, status of the translator, prevailing norms

1. Introduction Translation Studies deals not only with the process and product of translation as a linguistic phenomenon, but also with translation as a form of intercultural mediation taking place in a specific social and cultural context. And translators are increasingly portrayed as cultural mediators who possess a high degree of intercultural competence, as well as relevant subject-area expertise and other non-linguistic skills (cf.€Schmitt 1998; Wilss 1999; Katan 2004). My interest in this issue arises from the gap I perceive – at least, in the context in which I work – between what many translation scholars say should be the case (a good example is the concluding statement made by Snell-Hornby (1999:€164)) and the reality. I am also interested in how translators see themselves and



David Limon

how they are viewed by those who employ or commission them, as well as how their work is evaluated. These are, of course, the kinds of questions to which Translation Studies should be able to provide answers, for one of the aims of translation research should be to describe and evaluate the actual state of affairs in the field. I shall focus here on the kind of translation that I both practise and teach: that is, from Slovene into English. Whether at the university or on professional seminars, I work with translators translating from their native language into English. It is now a commonplace that such translation away from one’s first language is undesirable from a theoretical point of view, but it is a practical necessity for many languages which are often labelled as ‘peripheral’, either because they are ‘exotic’ from a European standpoint, such as Japanese, or less widely used, such as Slovene or Danish (cf.€Campbell 1998).

2. Explicitation Where genuine cultural mediation is taking place one would expect to find more translation shifts occurring than in a less interventionist linguistic transfer or transcoding. More specifically, such translations would be marked by the presence of explicitation, which was first described by Blum-Kulka (1986) as part of the process of interpretation performed by the translator on the source text, leading to the greater explicitness of the target language text. This tendency to expand the text in order to make it clearer seems to be inherent in the process of translation and has since been proposed as a possible translation universal (cf.€Baker 1996), although Toury (2004:€20) voices reservations in this regard. Blum-Kulka (op.cit. 298) distinguishes between text-focused and readerfocused translation shifts, the latter being concerned with “shared or non-shared reference networks”, or what we assume readers know and do not know. In the discussion below I shall focus on this aspect of the translation process: in particular, on how the translator deals with specific aspects of the source culture (or cultural space) that are likely to be unfamiliar to target readers. I am frequently struck by the lack of explanation of culturally restricted references – what we might refer to as cultural explicitation – in English translations of Slovene texts, which suggests that the translators are not functioning as cultural mediators. I shall illustrate my discussion with examples relating to natural heritage, specifically to Slovenia’s Karst region.

2.1

Examples: The Karst

There is a widespread assumption in Slovenia that everyone knows what the Karst is and what its features are. Karst is limestone or dolomite terrain, which is very porous, so karst areas have little surface water but a proliferation of underground rivers and thus many caves and related phenomena such as sinkholes. The Karst is also a region, as well as a geological type of landscape: when people in Slovenia say that they are



Translators as cultural mediators

going to “the Karst” it has a specific meaning and associations – just as when people say they are going to “the mountains” or “the coast”. The international scientific community has settled on the term “Karst”, from the German name for the Kras, a region in Slovenia that extends into Italy where it is called the Carso and where the first scientific research of a karst topography was carried out. Those who live in this area will undoubtedly be familiar with the term, as will geologists, but assuming a wider familiarity is questionable. The first three texts (A-C) below are extracts from English translations of Slovene texts available on the Internet when this paper was being prepared (August 2007) and are representative of the online texts informing a non-specialist international public about Slovenia’s Karst region and its remarkable features. These include Lake Cerknica, a disappearing or intermittent lake that in the winter is the largest body of water in Slovenia but in the summer largely disappears through sinkholes, and Rakov Škocjan, a regional park encompassing a valley containing a spectacular gorge with natural bridges at either end. The fourth text (D) was written in English by a non-specialist outsider who maintains a substantial website describing his travels in South-Eastern Europe. Strikingly, the latter is the only text to mention the crucial role played by the interaction of water and limestone in the formation of karstic features, or the fact that Lake Cerknica is a disappearing lake, or to clarify that Rakov Škocjan is a natural park centred around a gorge. The translated texts offer somewhat circular definitions of what the Karst is (text A), introduce karstological terms borrowed from Slovene, such as “karst polje” (text B), or explain what a “karst valley” is through the use of other, technical terms such as “gable end valley” (text C). The translated texts are thus marked by an absence of cultural explicitation or explanation of features belonging to Slovenia’s natural heritage; the translators do not seem to see it as part of their role to intervene in the translation process in order to make the content more accessible to the target reader. The rest of this paper will be taken up with a discussion of why this might be the case. (A) Karst Investigations in Slovenia The Karst presents a distinctive type of earth with special karstic phenomena getting the name after the region Kras in the background of the Trieste Bay. The name has Paleoeuropean origin (karra – stone) and in Antiquity it used to be Carusardius. From 1177 the Slovenian form “Grast” has been known and from 1230 the Croatian “Kras”. In the international terminology the German version – (Karst) – has remained in use, obviously after the place name Kras. http://www.ukom.gov.si/eng/slovenia/background-information/karst/ (B) Cerknica Lake Cerknica Lake is the largest typical karst polje in Slovenia. It is surrounded by the mountains of Slivnica (1022 m) on the north and Javorniki Mountain (1268 m) on the South, which belong to the Dinaric mountain chain. The climate is continental with the mean annual temperature 9,20 C in Cerknica.





David Limon

The annual precipitation is about 1700 mm in Cerknica and more than 2000 mm in the surrounding mountains. The driest month is July. http://www.life-kosec.org/index_en.php (C) Rakov Škocjan Between the Cerknica polje (Cerkniško jezero – lake) and Planinsko polje is located the picturesque karst valley of Rakov Škocjan. The creation of this karst valley is the result of a junction of two types of valleys: ‘the gable end valley’ where the river has its source and the ‘blind valley’ where the river flows underground. In the upper part, the gable end of the valley there are the remnants of the tunnel, which is the Little natural bridge and about 2.5 km down the Rak river is the Great natural Bridge. The Rak river flows from the Zelske cave (as a matter of fact from the sliplands). http://www.burger.si/Jame/RakovSkocjan/RakovSkocjanENG.htm (D) The Karst Southwest of Ljubljana is the intriguing Karst region. The interaction of water and limestone has created oddities such as the disappearing lake at Cerknica and the natural arches at nearby Rakov Škocjan gorge. The most popular attractions are Postojna and Škocjan Caves. The latter is slightly harder to get to but it’s well worth the effort. […] Rakov Škocjan A short distance north-west of Cerknica Lake, yet more geological oddities can be found in the Rakov Škocjan Regional Park (despite the name the park has no connection with Škocjan Cave). The park contains a short gorge with natural bridges at each end (Veliki and Mali Narodni Most). In between is a peaceful valley – it’s hard to reconcile the stillness here with the force required to create the stone arches. Information boards point out the various sites of interest in the park. http://www.balkanology.com/slovenia/article_karst.html

3. Possible reasons for the translator strategy So what explains lack of cultural explicitation in such texts? There are, I believe, a number of factors at play, which are listed below and dealt with in turn: – the kinds of texts that prevail in Slovene-English translation; – the status of translators in Slovenia; – the translation briefs that are given; – how translations are evaluated; – institutional pressure; – the fact that we are dealing with translation away from the native language;



Translators as cultural mediators 

– the education and training that translators have received; – and the differences between the two languages or cultures involved.

3.1

Genre

Most translation from Slovene into English is non-literary (e.g.€reports, presentations, conference papers, technical specifications, user manuals, correspondence, promotional texts) and many of the genres involved are relatively new to the Slovene environment. In earlier research (Limon 2004), I showed how much translating in Slovenia was influenced by the translation of EU legal texts into Slovene. The degree of mediation involved (cf.€Hatim and Mason 1997), i.e. the extent to which translators intervene in the transfer process, is very low with regard to EU texts, many of which are legal or ‘high status’ texts (the transfer of ‘quasi-sacred’ words, perhaps?), and this seemed to be carried over to other texts, such as those written for presentational or promotional purposes, where one would expect a different, more interventionist strategy to be used.

3.2

Translator status

Does the relative status of the translator within a particular society make it more or less likely that he or she is going to function as a genuine cultural mediator? Experience suggests that there is a link between translation strategy and status – the higher the assigned status within a particular socio-cultural context, the more probable it is that the translator will have the confidence to mediate or intervene in the translation process. However, the status of translators in Slovenia is generally low (with the possible exception of literary translators, who are seen as having an important cultural role – although ironically, literary translation is much worse paid than non-literary translation, translators presumably doing it primarily for personal satisfaction): tellingly, in a government classification of professional activities from 1994, translation is categorised alongside general secretarial activities and proofreading (Uradni list Republike Slovenije, no. 34/1994, 16.6.1994). In most professional contexts, translators are seen not as experts in cross-cultural communication or text formation who should have input into texts written for translation, but rather as (skilled) providers of language services. This, along with the generalisation of the strategy applied to EU translating mentioned in the previous section, could well be one of the reasons that translators tend to err on the side of caution, meaning that source text features are closely reflected in the target text. The most common strategy in Slovene non-literary translation could be characterised as being based on prudence and capitulation rather than risk-taking and persistence (cf.€Campbell 1998). Other reasons for the prevalence of this strategy are offered below; however, it is worth noting here that some scholars, such as Pym (2007), suggest that a disposition towards “risk aversion” – along with the



David Limon

tendency to standardise and the tendency to channel interference – may be one of the possible laws of translation behaviour.

3.3

Translation brief

My experience as a practising translator in Slovenia suggests that often commissioners have not really thought about why a particular text is being translated into English, or what the target audience is: it has simply become the norm over the last decade for Slovene official, institutional or organisational websites to appear in both English and Slovene. To obtain any kind of explicit brief demands a certain effort and assertiveness on the part of the translator which is often not repaid in terms of clarification of the translation task. And this assumes, of course, that the translator has direct contact with the client (the dyadic model), whereas if working for one of the agencies that have mushroomed in recent years (and which obtain much of their work through public calls for tender in which price is the main criterion), the translator may have no such contact and may receive no guidelines as to the functional purpose of the target text. In such a translational situation, professional common sense suggests that it is probably best to ‘play safe’ and the default brief becomes ‘do not deviate too much from the original’, which once again discourages a more active mediation. In future, this kind of situation may become easier to avoid if the EN-15038€European Quality Standard for Translation Services begins to be widely applied in Slovenia, as this spells out what is required of all the participants in the translation process; there are promising signs that the more reputable translation companies (such as members of the Eupharia consortium) are already committed to ensuring that quality and not just price becomes part of the translation equation.

3.4

Translation evaluation

For simple practical reasons, i.e. the lack of native speakers of English with a knowledge of Slovene, translations into English tend to be evaluated by native speakers of Slovene, often by the commissioner or the writer. Usually they have less or much less competence in English than the translator. A common judgement is that the translation is a ‘good’ one if it ‘sounds like’ (reads like) the original.€This reinforces the noninterventionist strategy mentioned above and tends to exclude both explicitation and implicitation.

3.5

Institutional pressure

In 2005, I carried out a small research project (Limon 2005) into the educational profile, social and legal status, training and work experience, and views on translation of non-literary translators in Slovenia. This research identified institutional pressure on



Translators as cultural mediators 

translators to conform to prevailing norms, or to pursue what Venuti (1998) calls an “ethics of sameness”, and to adopt a low-mediation approach. Of the 87 translators who were interviewed or completed questionnaires, 45% said they adhere closely to the source language text when they had been criticised in the past for failing to do so. Thus in reply to the question “In what circumstances do you tend to stay close to the original when translating?” the most frequently selected responses were: when the topic is unfamiliar, 75%; when the original is difficult to understand or unclear (”the less clear the text, the more word-for-word the approach” as one translator observed), 72.5%; when the text is “important”, 50%. However, translators avoid risk not only because of possible sanctions, but also because of the “cultural reward system” (cf.€Pym 2005) that structures their work. Simeoni (1998) suggests that “subservience” or extreme conformity to norms and the translator’s endorsement of it is a central feature of the translatorial habitus. Due to the way in which translators internalise normative behaviour, it may be that translation competence is more characterised by “conformity” than other competences involved in cultural communication (Simeoni op.cit. 5). Moreover, as Toury (1998:€17) points out, norms are likely to have “much more binding force” within a subgroup, such as a specific body of professional translators working in a particular field between specific languages, than in a larger more heterogeneous group (e.g.€text producers in general).

3.6

Translating into a non-native language

Few translators in Slovenia translate only into their native language: in the survey mentioned in the last paragraph, only 7% of respondents said they translated exclusively into Slovene, while 24% translate mainly into a non-native language and 53% translate more into a non-native language than into Slovene. This raises the question as to whether cultural explicitation is more or less likely if the translator is working out of his/her native language. It seems at least a reasonable hypothesis that it is harder for those translating away from their mother tongue to identify with the target language reader, compared to translators working into their native language and culture. (Ideally, we would compare here both those translating into and those translating away from their native language, but the problem with the kind of texts we are looking at – as is the case with almost all non-literary translation, with the exception of published books and catalogues – is that the translators are anonymous.) And the less likely one is to place oneself in the non-local reader’s shoes, the smaller is the likelihood that one will recognise the need for local or ‘insider’ knowledge to be spelled out or elaborated.

3.7

Translator education and training

According to the survey mentioned in Section€3.5, the great majority (over 75%) of translators in Slovenia with five years or more experience are language graduates, who



David Limon

will have learnt their foreign language(s) through more traditional methods that emphasised (as they still do) grammatical and lexical accuracy. Moreover, they are unlikely to have been involved in any formal study of non-literary texts: except, perhaps, in essay writing classes, there is little emphasis in language programmes on text level analysis, or on the contrastive study of genre conventions or rhetorical preferences. Within language teaching, partly because of the washback effects of examinations, a great deal of decontextualised lexical and grammatical work takes place, leading to a focus on lexico-grammatical accuracy rather than communicative considerations. And, due to the dynamics of the language classroom, the same can be said of the limited translation work that takes place in this context: methodically working through a text and its translation, sentence by sentence if not word for word, focusing on surface detail, seems to offer satisfaction to both teacher and students – to the former because such a lesson is easy to plan and straightforward to run, to the latter because they feel they are learning something concrete, i.e. pairs of ‘equivalent’ expressions, which they can note down, perhaps with a view to memorising them later. Moreover, training and support provided to professional translators, such as those in government ministries and other public bodies, generally focus on terminological issues and lexical standardisation through the use of translation software (see, for example, the paper by Krstič and Belc (1999) on the translation of EU legislation, which focuses on workflow management, documentation types and, above all, on software translation tools). This tendency to encourage one-to-one associations between source and target translation units leads to the production of what Lörscher (1991; quoted in Fawcett 1997:€143) refers to as “sign-oriented” rather than “sense-oriented” translations – the translation unit being taken as the formal linguistic sign, rather than the sense of the message.

3.8

Cultural differences

In many discourse fields (academic and scientific writing, as well as computing, telecommunication and marketing texts are obvious examples), Anglo-Saxon cultural values are widely seen as universal or neutral: no “cultural filter” (cf.€House 2001) is employed, but rather rhetorical patterns and register values are imported directly into Slovene, influencing a wide range of genres. With translation from English, the default assumption is increasingly that cultural explanation is not required due to the language’s growing global role. What is surprising is that translators may be applying the same assumption in the other direction – translating from Slovene into English – even though the Slovene culture is obviously more peripheral than the Anglo-Saxon one. Hinds’ (1987) distinction between “writer responsible” and “reader responsible” cultures – i.e. whether responsibility for effective communication is seen to lie primarily with the writer or the reader – is pertinent here. In English, there is a tendency towards writer responsibility: if communication fails we do not blame the reader for not making enough effort, but assume that what was said was insufficiently clear or well-organised. In other cultures it is seen as the reader’s responsibility to understand what the



Translators as cultural mediators 

writer intended to say and writers may prefer to offer hints and nuances rather than make direct statements, so that more is implied than stated explicitly. Japanese is the example discussed by Hinds but the same tendency may be seen in Slovene and in other Central European cultures (see, for example, Čmejrková and Daneš 1997). This tendency towards writer responsibility may help to explain the greater propensity of writers in English to clarify or provide more information on local features that may not be known to a wider audience. To take examples from my native county: it is striking that the first ten entries that appear in Google for Hornsea Mere all spell out to the reader what it is, i.e. the largest freshwater lake in Yorkshire, while the entries for Roseberry Topping clarify that it is a distinctive hill (the peculiar shape is due to a geological fault and a mining collapse early in the 20th-century) with fine views across Yorkshire. There seems to be an English tendency to err on the side of caution in this regard rather than to assume knowledge on the reader’s part and this is not perceived by English readers as patronising or ‘reader disrespectful’. By contrast, my translating experience, as well as extensive seminar discussions with both students and professional translators on political, business, academic and humanities texts in the two languages, suggest that Slovene writers are more likely to assume knowledge on the reader’s part and this tendency may well influence Slovene translators, who have been raised and educated within the same culture. An even more speculative question is whether such clarification is less likely in small cultures than large ones: in such a small geographical space as Slovenia, with its population of less than two million, it is frequently said that ‘everyone knows everyone’; by analogy, one might also say ‘everyone knows everywhere’ in the sense that knowledge of natural features other than the most obscure and their location are somehow taken for granted.

4. Conclusions and implications for Translation Studies Much of what is said above is of a speculative nature, generalising from a limited number of cases and small-scale research: clearly there is plenty of scope for further research into the translator habitus, in Slovenia as elsewhere, to enable us to better understand not only the cognitive act of translation but also the translation event (cf.€Toury 1995:€249ff.). General research into translation norms and conformity to them is obviously important in this regard. I am also convinced of the need for more contrastive rhetorical research into pairs of languages such as Slovene-English – or, to be more precise, on the specific cultures underlying the languages – looking at broad textual preferences, preferred rhetorical styles, genre conventions, expectation norms and so on, facilitating the construction of an appropriate cultural filter. To understand differences between translated and non-translated texts in a language we need crosscultural insights into factors such as writer versus reader responsibility, as well as differences with regard to information load, reader friendliness, simplicity and clarity, tolerance for digression, recapitulation and repetition, and so on.



David Limon

To turn for a moment to translator education and training: mastering new genres and styles of discourse in a target language is not part of the translator’s general language competence – indeed we usually lack such competence in our own language if a genre is new to us. Translators thus require specific training in this area: if translation is viewed as a special kind of writing, then the relevant writing skills need to be acquired, whether they are going to be employed interculturally or intraculturally. Translator education should provide the kind of intercultural awareness that makes it easier for translators to view the two cultures between which they are mediating as both ‘insiders’ and ‘outsiders’, making it more likely that they will be able to pursue a more interventionist translation strategy when this is required. This is, of course, particularly difficult in translating into a non-native language, which is actively taught in Slovenia and many other countries in Europe, even though in some others it may be frowned upon (the influential British scholar Newmark (1988:€ 3) typically refers to translating into your first language as “the only way you can translate naturally, accurately and with maximum effectiveness”). Here, a particular emphasis on audience awareness training is required, enabling student translators to see beyond language differences to cultural ones. The aim, of course, is not ‘biculturalism’, which is too nebulous and unrealistic a target, but a focused, genre-based, cross-cultural textual awareness informed by the necessary contrastive research. A much wider and more long-term goal is to raise the profile and status of translators and thus, hopefully, their self-esteem and willingness to be assertive, which is a prerequisite for cultural mediation. Final year students of translation at the University of Ljubljana attend classes on professional aspects of the translator’s work, which can make a contribution in this area. If a translator acts like a cultural mediator or an expert in text formation and can articulate that role, then he or she is more likely to be treated as such by other stakeholders in the translation process. However, the situation in which most translators find themselves may not be conducive to this and there is plenty of anecdotal evidence (much of it heard at translation conferences such as the one that prompted the present paper) that while some translators may well be moving into higher profile roles in international marketing, many more feel isolated and marginalised, which is likely to reinforce the already noted tendency towards risk aversion. Translation Studies really does matter in this regard: descriptive research can give us a greater insight into how translators actually work and the constraints under which they operate, rather than projecting a translatorial role that may in many contexts be little more than wishful thinking. If, however, the role of Translation Studies is to ‘change the world’, as well as to interpret it, then it should be able to contribute towards a blueprint of how the concept of translator as cultural mediator can be realised.



Translators as cultural mediators 

References Baker, M. 1996. “Corpus-based Translation Studies: The challenges that lie ahead.” In Terminology, LSP and Translation. Studies in language engineering in honour of Juan C. Sager, H. Somers (ed.), 175–186. Amsterdam/Philadelphia: John Benjamins. Blum-Kulka, S. 1986. “Shifts of cohesion and coherence in translation.” First published in Interlingual and Intercultural Communication: Discourse and Cognition in Translation and Second Language Acquisition Studies, S. Blum-Kulka and J. House (eds), Tübingen: Gunter Narr. Page references here to The Translation Studies Reader, L. Venuti (ed.), 2004, 298–313. London and New York: Routledge. Campbell, S. 1998. Translation into the Second Language. London and New York: Longman. Čmejrková, S. and Daneš, F. 1997. “Academic writing and cultural identity: The case of Czech academic writing.” In Culture and Styles of Academic Discourse, A. Duszak (ed.), 41–61. Berlin and New York: Mouton de Gruyter. Fawcett, P.D. 1997. Translation and Language: Linguistic Theories Explained. Manchester: St Jerome. Hatim, B. and Mason, I. 1997. The Translator as Communicator. London and New York: Routledge. Hinds, J. 1987. “Reader versus writer responsibility: A new typology.” In Writing across languages: Analysis of L2 text, U. Connor and R.B. Kaplan (eds), 141–152. Reading, Mass: Addison-Wesley. House, J. 2001. “Translation quality assessment: Linguistic description vs social evaluation.” Meta 46 (2): 243–257. Katan, D. 2004. Translating Cultures. An Introduction for Translators, Interpreters and Mediators. 2nd revised edition. Manchester: St Jerome. Krstič, A. and Belc, J. 1999. “Prevejanje pravnih aktov evropske unije.” In Prevajanje in tolmačenje (Uporabno jezikoslovje, Tematska številka, 7–8), I. Kovačič and I. Štrukelj (eds), 191–205. Ljubljana: Društvo za uporabno jezikoslovje. Limon, D.J. 2004. “Translating new genres between Slovene and English: An analytical framework.” Across Languages and Cultures 5 (1): 43–65. Limon, D.J. 2005 “A social profile of the translator in Slovenia.” Unpublished paper given at the conference Translating and Interpreting as a Social Practice, 5–7€May 2005, Universität Graz. Lörscher, W. 1991. Translation Performance, Translation Process and Translation Strategies: A Psycholinguistic Investigation. Tübingen: Gunter Narr. Newmark, P. 1988. A Textbook of Translation. Hemel Hempstead: Prentice Hall. Pym, A. 2005. “Explaining explicitation.” In New Trends in Translation Studies. In Honour of Kinga Klaudy, K. Károly and A. Fóris (eds), 29–34. Budapest: Akademiai Kiado. Pym, A. 2007. “On Toury’s laws of how translators translate.” (draft). Accessed 17.12.2007 at http://www.tinet.org/~apym/on-line/translation/translation.html. Schmitt, P.A. 1998. “Berufsbild.” In Handbuch Translation, M. Snell-Hornby, H.G. Hönig, P. Kussmaul, P.A. Schmitt (eds), 1–5. Tübingen: Stauffenburg. Simeoni, D. 1998. “The pivotal status of the translator’s habitus.” Target 10 (1): 1–39. Snell-Hornby, M. 1999. “Some concluding comments on the responses.” Current Issues in Language and Society 6 (2): 161–164. Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam and Philadelphia: Benjamins.



David Limon Toury, G. 1998. “A handful of paragraphs on ‘Translation’ and ‘Norms’.” In Translation and Norms, C. Schäffner (ed.), 10–32. Clevedon: Multilingual Matters. Toury, G. 2004. “Probabilistic explanations in Translation Studies: Welcome as they are, would they qualify as universals?” In Translation Universals: Do They Exist? A. Mauranen and P. Kujamäki (eds), 15–32. Amsterdam and Philadelphia: John Benjamins. Venuti, L. 1998. The Scandals of Translation: Towards an Ethics of Difference, London: Routledge. Wilss, W. 1999. Translation and Interpreting in the 20th Century: Focus on German. Amsterdam and Philadelphia: John Benjamins.

Censorship in the translations and pseudo-translations of the West Carmen Camus Camus

University of Cantabria, Spain

During the Franco dictatorship (1939–1975), the Western genre (hereafter ‘Western’), both imported American narratives dealing with themes of the so-called Far West and pseudo-translations (imitations by Spanish authors using American generic models) enjoyed enormous popularity with the Spanish reading public. Like all literary production in Franco’s Spain, these works had to go through the state censorship system set up by the regime. This paper examines a sample catalogue of 727 censorship files for the Western from the Administration’s General Archive, and attempts to explain the popularity of the genre in political and economic terms. Overall, 11.2% of the authorization requests were censored, 8% requiring suppressions and 3.2% being denied authorization. Censorship affected imported works and Spanish production to the same extent. Censorship varied over the period studied, being most lenient in the peak period between 1961 and 1965 and stricter after the 1966 Press Law, but especially between 1970 and 1973. The data revealed that no publisher or author was favoured in any special way by the censors. It is concluded that the boom experienced by the Western stemmed from an act of culture planning by the Franco regime, which was implemented through the expertise of publishers in exploiting both the pre-existing popularity of the genre and the advantageous financial conditions offered by the government. Keywords: censorship in Franco’s Spain, translation of popular genres, translating the Western genre

1. Introduction During the Franco era, especially in the early years, Spain was isolated from the rest of the world, with an economy devastated by a civil war and suffering the consequences of World War II. Within this context, censorship was a measure used by the Franco regime to control every minute detail of cultural activity that took place in the country.



Carmen Camus Camus

State censorship was, therefore, a practice that pervaded all Spanish literary production and, as a result, acted as a brake on creative writing. However, in spite of Spain’s dire economic status, popular literature, especially the narrative of the Far West, was a realm of Spanish literary production in which sales during the forties, fifties and sixties soared to levels never achieved either before or after that period. The soaring popularity of the Western in a country where food was scarce and rationed is a paradox, and raises a number of crucial questions. What made it possible for a genre like the popular literature of the West to flourish and gain the favour of the public? Did these texts receive special favour from the officials in charge of the censorship files? Did Franco’s government somehow promote the emergence of this type of narrative? Why were some writers of Republican ideology allowed to write and publish in this kind of genre? Toury (1995) considers translations as products used to fill a void that has emerged in a given culture at a particular historical time. Pseudo-translations – that is, works produced in the country where such a need has been created – are imitation models of original or source culture texts introduced into the host culture: they are “texts which have been presented as translations with no corresponding source texts in other languages ever having existed” (Toury 1995:€40). These texts produced in the target culture are then regarded as translations of putative original texts written in the source culture and as such pertain to Translation Studies due to the cultural position they share with the genuine translations (Toury 1995:€45). This paper examines from a descriptive point of view the effects of Franco’s censorship on the translations and pseudo-translations of the West for the period that spans from 1939 to 1975, and attempts to identify what governmental, political and economic measures made possible the flourishing and splendour of this popular narrative genre.

2. Material and methods To address the above-mentioned questions, we have based our analysis on the censorship files for Western narrative at the Archivo General de la Administración (AGA, Administration’s General Archive), which is where all the files opened during Franco’s regime are kept. A catalogue of 727 censorship files1 was created by selecting an arbitrary starting point in 1939 and incorporating the files for all Westerns contained in every fiftieth AGA box thereafter until 1975. The AGA box, which is the container used to store the censorship files, could hold between one and fifteen files. This selection procedure guarantees that the sample is representative of the genre in the files stored at the Archive for literary production generated in Spain during the period 1. This catalogue forms part of the TRACE Project (TRAnslation and CEnsorship in Franco’s Spain), a joint venture of Translation Studies Researches at the University of León and the University of the Basque Country at Vitoria.



Censorship in the translations and pseudo-translations of the West 

studied (for details see Merino Álvarez 2005). These files not only contained information on the authors’ names, any pseudonyms used and editorial affiliation and characteristics, but also and more importantly included details of the Censorship Board’s reviews and decisions. This information was transferred to a specially created data base in electronic format to facilitate subsequent analysis in relation to socio-political data and official documentation on state censorship. The catalogue compiled for this study covers all types of narrative dealing with the Western, from comic-style picture stories to full-length novels, and includes both pseudo-translations and the translations of the imported narrative of prestigious American authors as well as pulp fiction writers. The choice of the Western was based on the findings in a previous study from the TRACE project (Rabadán 2000:€ 258), which showed that this genre accounted for 87.5% of the total files found for Kiosk fiction between 1955 and 1981.

3. Censorship under the Franco regime Censorship in Franco’s regime falls into three broad categories which varied during the period: – state censorship which was affected by law reforms and ministerial changes; – religious censorship which was affected by the changing relations between the state and the church; – and economic censorship which was affected by both state and publishers’ policies. Although the three types of censorship are closely linked, this paper will focus on state and economic censorship.€The Franco Regime established a closely knit censorship system to control all cultural activity in the country during the dictatorship.€This meant that before reaching the public every single cultural item had to be submitted to the censorship board to obtain the official pass. The criteria used by this board had an aura of mystery, as no one knew for certain what bounds could not be trespassed. According to Abellán (1980:€88) these unwritten and unknown criteria fell into four broad categories: sexual morality, political opinions, improper use of language, and religion. In this political scenario, the emergence and popularity of the Spanish Western merits closer examination. There were a number of factors which conditioned the promotion of this Western narrative in Spain. In the first place, there was the need to fill the void created by the stampede to leave Spain by the nation’s talented writers during and just after the civil war. In addition, the regime’s zeal to promote “true National Identity” led to a rejection of the translations of foreign works, which were considered dangerous because they were seen as vehicles of a nonSpanish or even worse, anti-Spanish spirit (Camus 2008:€160). In response to this



Carmen Camus Camus

situation, a Law to promote a Spanish narrative was passed in 1946. In the preamble to the Law we can read: On the greatest possible diffusion of Spanish books, both at home and abroad. The unfavourable position and state of abandon of Spanish books in foreign markets, together with the prices – inaccessible to modest budgets – reached on the home market, which is clearly detrimental to the diffusion of any type of culture, require immediate and effective action to end this precarious situation and set the foundations for the future expansion for which Spanish books are destined, principally in Spanish-America, because of the universality of our language and the catholicity of our spirit (LEY de 18 de Diciembre de 1946. BOE de 19 de Diciembre de 1946, nº 353, pp.€8829–8830. JEFATURA DEL ESTADO).2

This Law, which would allow the regime to control the product more easily, established a number of measures including State control of the price of books, a reduction in the cost of paper and the granting of tax concessions to publishing companies. As translations were anathema to the regime, publishers – with the support of the regime – sought to promote narrative models that were already accepted by the public. Ironically, the most popular genre at that time was the Western. To fill the gap left by the exiled intellectuals, publishers recruited an army of ready-made writers from diverse professions and even employed Republican writers for their new enterprise. To conceal from the public eye the lack of men of letters or the fact that Republican authors were producing those works, writers were asked to sign their works by using pseudonyms, and in the case of the Western, they were asked to choose names with reminiscences of the prairie (Camus 2008:€ 157). The new bunch of writers produced their works by imitating the models which had been imported through translations. These pseudo-translations could be closely controlled not only by the State system but also by the publishers, who exerted economic censorship on both the writers and their writings. These fictitious Westerns, therefore, represent an example of what Toury refers to as “an act of culture planning”, understood as “any attempt made by an individual, or a small group, to incur changes in the cultural repertoire, and the ensuing behaviour, of a much larger group” (Toury 2005). The shadow of this Book Promotion Law stretched out as far as 1966 when the censorship system was modified by a Press Law which replaced the 1938 Censorship Law. Although the new Law officially abolished compulsory censorship, actual censorship practice continued much as before, but now the onus fell on the publisher. Publishers could choose either to present galley-proofs for authorization in what was termed voluntary consultation, or follow the newly opened path of Deposito Previo ‘Prior Deposit’. This entailed applying for a registration number, which was extremely difficult to obtain, and publishing directly after depositing six copies of the finished 2. All the English translations of the Spanish texts are mine except where the original English text is available.



Censorship in the translations and pseudo-translations of the West 

book at the Ministry of Information and Tourism. However, with the new procedure, publishers ran the risk of having their works confiscated, and, therefore, placed all their investment at risk. So, in fact, most publishers continued to present their works for voluntary consultation, but since the general belief was that the new system was more lenient, publishers became more adventurous and, as a result, the number of suppressions increased.

4. Results Figure 1 illustrates the evolution of authorization requests for the genre during the period for each year. As the figures do not follow a stable pattern, the graph shows mean values calculated for each five year period from 1941 to 1974, which gives a clearer picture of the changes over time. The first file for the genre was found in 1943. There was a sharp but sustained increase from 1945 to 1957, followed by a further rise to reach a peak in the period 1961 to 1965 and then a steady decline to the end of the 60 50 5-year average 40 34.2 29.8

30

25.8 20.6

15.6

20 10

6.2

0

1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975

1.8

Book Promotion Law 1946

Arias Salgado 1951

Fraga Iribame 1962

State-Church Concordat 1943

II Eucharistic II Vatican Congress 1952 Concilium 1962-5

Sánchez Bella 1969

Pío Cabanillas 1974

Press Law 1966

Book Law 1975

Figure 1.╇ Evolution of authorization requests for the Western in relation to relevant political and historical events



Carmen Camus Camus

regime.3 The initial sharp increase came after the Book Promotion Law was passed in 1946. As we have seen, the spirit of this Law was to promote the publication of truly Spanish books, that is, books written in Spain in tune with the values imposed by Franco’s regime, and to this end the Law embraced financial measures to encourage publishers. The period of splendour coincided with the years immediately after the second Vatican Council and with Fraga Iribarne as the head of the Ministry of Information and Tourism. The second Vatican Council represented the end of the close bond between the regime and the Catholic Church, which now lost much of its influence on the censorship process. In addition, the regime wanted to offer the world a new reformist image and this task was handed to the incoming minister Fraga Iribarne. Fraga introduced new legal regulations on censorship and, with the slogan of apertura “opening”, brought a breath of fresh air to the press and publishers. The start of the decline of the Western coincided with the new Press Law of 1966 (BOE 18/3/1966) and continued with the hard-liner Sánchez Bella in the Ministry. The arrival of Sánchez Bella meant a regression in the freedom of the media and a return to the all-powerful mode of censorship of previous periods. Table 1 shows the distribution of the three types of resolutions emitted by the censors for the whole sample and for imported works and pseudotranslations. The vast majority (88.8%) were authorized directly, with 8% authorized after suppressions had been made and only 3.2% actually denied authorization. This genre, therefore, can be considered to have been treated relatively leniently by the censorship system with 11.2% suffering some kind of censorship.€This holds true both for the translations of novels imported from America and for the pseudotranslations written by Spanish authors. Table 2 presents the raw data for the incidence of censorship on the Western during Franco’s dictatorship for individual years. For those works authorized after suppressions had been made, the highest figures coincide with the period after the Press law was passed in 1966, with 37 of the 58 instances (64%) corresponding to this nineyear period, when in fact the number of authorization requests for the genre began to Table 1.╇ Comparison of censorship resolutions in imported and Spanish texts

Imported works Pseudotranslation Total

Authorization requests

Authorized directly

â•⁄ 88 639 727

â•⁄ 77 (87.5) 569 (89) 646 (88.8)

Authorized with Authorization suppressions denied â•⁄ 7 (8) 51 (8) 58 (8)

â•⁄ 4 (4.5) 19 (3) 23 (3.2)

* Figures in brackets are percentages

3. The figures for 1975 were excluded from this calculation because the exceptionally high figure for this year is due to the introduction of the International Standard Book Number (ISBN), whereby publishers had to resubmit previously published works in order to obtain the ISBN, and this measure inflated the quantity of works passing through the system.



Censorship in the translations and pseudo-translations of the West 

Table 2.╇ Incidence of censorship on Western narrative during the Dictatorship in the study sample (N = 727) Year

Authorized with Suppressions

Authorization Denied

1945 1946 1949 1950 1951 1953 1954 1955 1956 1957 1959 1961 1964 1965 1967 1968 1969 1970 1971 1972 1973 1974 1975 Total

1 1 – – – 2 2 1 1 4 1 3 3 2 7 6 4 4 7 5 2 1 1 58

1 – 1 1 1 1 3 – 1 – 1 – – – – – 2 2 8 1 – – – 23

decline. Interestingly, in the period with the highest number of authorization requests, from 1961 to 1965, only 8 resolutions with suppressions were found in the study sample. For the texts that were denied authorization, again, the highest figures coincide with the stricter application of censorship after the 1966 Press Law: 13 (57%) out of a total of 23 for the whole period. In Figure 2, these censorship data are related to periods coinciding with the main historical events. It can be seen that censorship was relatively lenient in the years of increasing popularity of the Western up to 1965. The lowest figures within this period (4.5%) coincided with the first years of Fraga Iribarne at the head of the Ministry of Information and Tourism and are consistent with the new air of openness with which he started his term of office. The data also confirm the stricter censorship after the 1966 Press Law, when the incidence of censorship rose above the mean (11.2%) for the

Carmen Camus Camus 35

Average for whole period

30.8%

30 25 Percent



20 15 10

16.6% 9.5% 7.5%

4.5%

5 0

1945-52

1953-61

2.5%

1962-65

1966-69

1970-73

1974-75

Period With suppressions Book Promotion Law 1946

Arias Salgado 1951

Fraga Iribame 1962

State-Church Concordat 1943

II Eucharistic II Vatican Congress 1952 Concilium 1962-5

Denied Sánchez Bella 1969

Pío Cabanillas 1974

Press Law 1966

Book Law 1975

Figure 2.╇ Incidence of censorship according to periods related to relevant political events

whole period to 16.6%. This rise was due to the misguided belief that this Law represented the end of censorship, which led to authors reducing self-censorship and publishers relaxing their censorship norms. However, with this Law, not only was there no decrease in censorship but, in fact, it reached an unprecedented peak (30.8%) under the hard line of Sánchez Bella, whose brief three-year rule saw the percentage of requests authorized with suppressions or completely denied soar to almost three times the average for the period under study. We shall now look at how censorship affected publishers and individual authors. Table€3 reflects the authorization requests presented by the main publishers, together with the figures both for authorization with suppressions and for those works that were banned. In terms of requests, Bruguera, which accounts for 290 (40%) of the 727 censorship files in the study sample, is the outstanding publishing house for the genre, followed at a distance by Rollán, Toray and Exclusivas Ferma. A similar pattern can be seen for authorization with suppressions, with Bruguera at the head followed by Rollán and Toray, and these three publishers accounting for almost 70% of this type of resolution. In contrast, for requests that were denied authorisation, the main publisher affected was Producciones, a publisher under the same management as Exclusivas Ferma. Together



Censorship in the translations and pseudo-translations of the West 

Table 3.╇ Incidence of censorship according to main publishers Publisher Bruguera Rollán Toray Exclusivas Ferma/Producciones Plaza Castellana E. Cies Pocket Books Dollar Cid Editorial Eisa/ Ediciones Perma Books Favencia Edhasa Caralt Alonso Editorial Molino Petronio Afha Internacional Editorial Marco Hispano Americana Ediciones Símbolo Tesoro Buru Lan Editorial Téllez Others Total

Total Requests 290 108 â•⁄ 56 â•⁄ 45 â•⁄ 23 â•⁄ 18 â•⁄ 15 â•⁄ 15 â•⁄ 14 â•⁄ 13 â•⁄ 13 â•⁄ 12 â•⁄ 10 â•⁄â•⁄ 9 â•⁄â•⁄ 3 â•⁄â•⁄ 3 â•⁄â•⁄ 3 â•⁄â•⁄ 3 â•⁄â•⁄ 2 â•⁄â•⁄ 2 â•⁄â•⁄ 2 â•⁄â•⁄ 2 â•⁄â•⁄ 2 â•⁄â•⁄ 1 â•⁄â•⁄ 1 â•⁄ 62 727

Authorized with Authorization Suppressions Denied 30 â•⁄ 7 â•⁄ 3 â•⁄ 2 â•⁄ 0 â•⁄ 1 â•⁄ 1 â•⁄ 0 â•⁄ 0 â•⁄ 1 â•⁄ 3 â•⁄ 0 â•⁄ 0 â•⁄ 1 â•⁄ 2 â•⁄ 1 â•⁄ 1 â•⁄ 0 â•⁄ 1 â•⁄ 1 â•⁄ 0 â•⁄ 1 â•⁄ 1 â•⁄ 1 â•⁄ 0 â•⁄ 0 58

â•⁄ 4 â•⁄ 2 â•⁄ 0 â•⁄ 6 â•⁄ 0 â•⁄ 0 â•⁄ 1 â•⁄ 0 â•⁄ 0 â•⁄ 0 â•⁄ 0 â•⁄ 2 â•⁄ 0 â•⁄ 1 â•⁄ 0 â•⁄ 0 â•⁄ 0 â•⁄ 3 â•⁄ 1 â•⁄ 0 â•⁄ 2 â•⁄ 0 â•⁄ 0 â•⁄ 0 â•⁄ 1 â•⁄ 0 23

they had a total of 6 requests denied. In spite of the large number of requests, Bruguera only made 4 requests that failed to receive authorization from the Censorship Board. Figure 3 summarises the previous results and compares the negative censorship resolutions for the four main publishers. Of these, Toray, with only 5% of its requests needing modifications prior to publication, and Rollán (8.3%) were treated leniently. The 11.7% for Bruguera is virtually identical to the mean of 11.2% for the whole study sample (see Table€1). In spite of the outstanding contribution of this publisher to the Western, Bruguera does not appear to have been favoured in any special way by the censors. On the other hand, the fact that this publisher was not harshly treated reflects the



Carmen Camus Camus 40 35

11.8%

11.7% 4

30 25 20 15

11 30 15.6%

10

6

5 0

2 Bruguera

Exclusivas Ferma/Producciones With suppressions

8.3% 2 7 Rollán

16

5.0% 3 Toray

Others

Authorization Denied

Figure 3.╇ Negative censorship resolutions for the four main publishing houses

effectiveness of the internal editorial censorship imposed on the wide range of writers who submitted works to this company. Thus, from our data it was only Producciones and Exclusivas Ferma that suffered particularly harshly at the hands of the Censorship Board. As we have seen, the most significant publisher for the Western was Bruguera. Of the major authors who published their work in this genre with Bruguera, the most prolific production was by Lafuente Estefanía, accounting for 85 of the 290 files found for Bruguera in the sample. Lafuente Estefanía was, in fact, a brand name under which the father, Marcial, and his two sons submitted their works to Bruguera for publication. However, for this type of narrative, to publish under your own name was exceptional since, as seen in Section€3, writers were encouraged to hide their identity with a Western-flavoured pseudonym (Camus 2008:€157) Other important writers for this publisher were Oliveros Tovar (25), Prado Duque (19), González Ledesma (18), Olcina Esteve (15), and García Lecha (14). Another five authors had between 6 and 10 requests made by Bruguera, with the remaining 75 censorship files for this company corresponding to numerous other writers. Figure 4 reflects how censorship affected the four leading writers publishing with Bruguera. Although in absolute terms the numbers are very similar for the four authors, who all had three or four works censored in some way, the data show that Marcial Lafuente Estefanía, a writer with a Republican background, was treated leniently: the four works of his which either required modification or failed to obtain authorization represent only 4.7% of his total of 85 titles. In contrast, González Ledesma, another



Censorship in the translations and pseudo-translations of the West 4.7% 1 3

90 80 70 60 50 40 30 20 10 0

81

22.2% 1 3 14 González Ledesma, Francisco (Silver Kane)

Lafuente Estefanía, Marcial Antonio

Authorized

16.0% 1 3

11.1% 1 2

21

24

Oliveros Tovar, Miguel (Keith Luger)

Prado Duque, Fidel (W. Martyn)

With suppressions

Denied

Figure 4.╇ Incidence of censorship on the four main writers for the Western publishing with Bruguera

Republican writer, was controlled very strictly by the censorship system, since the 22.2% for his production is twice the average for the sample. However, Oliveros Tovar (16.0%) was also adversely affected by the censors, whereas the percentage for Prado Duque (11.1%) coincides with the average of 11.2% for the whole sample. It appears, therefore, that it was not only political reasons that influenced the censors’ decision.

5. Censorship and sexual morality We shall now look at a few instances of the censorship inflicted on the Western. In general, in this genre, censors targeted improper language, scenes depicting violence and especially explicit and even implicit erotic references. In Example 1, the censor has suppressed part of the text describing a passionate scene in which the sexuality was considered too explicit and contravened the moral standards set by the regime, which was at that time strongly influenced by the Catholic Church. However, the censor has also included a marginal note: lo que está acotado en negro no se tacha, i.e. “what is marked in black is not to be suppressed”, since this expresses proposal of marriage, and this was in line with the regime’s morality. This part of the text reads: – ¿Quieres casarte conmigo? – le oyó murmurar como en sueños, junto a su oído. Loca de felicidad le echó los brazos al cuello. “Will you marry me?” she heard him whisper as in a dream into her ear. Madly happy, she threw her arms round his neck.





Carmen Camus Camus

Example 1.╇ File No. 21/1957. Tex Taylor (Mario Calero Montejano): El Ranger

Example 2.╇ File No. 3479/1967. Louis Rock (Luis Rodríguez Aroca): Una Cuenta con la Horca



Censorship in the translations and pseudo-translations of the West 

Example 2, which consists of two related passages from a novel whose censorship file is from 1967, illustrates what could be described as the typical trend for the post-Press Law period: the censor simply suppresses or modifies specific items in the text in order to attenuate meaning. Here, first totalmente desnuda ‘completely naked’ is suppressed, but then only totalmente is removed from the compound totalmente desvestida ‘completely undressed’, presumably because the state of undress without the adverb would only evoke a state of partial undress. In the third context, the censor recommends replacing completamente desnuda ‘completely naked’ with semi-desvestida ‘half-undressed’. However, with these changes, the narrative loses not only its force but also a good deal of its coherence as the girl has been used to lure the members of a gang into a trap.€The English translation reflects the censored version with the changes indicated in square brackets: Pete and Flash had heard the fight from a distance and rushed to the aid of their companion. Linda was waiting for them in the centre of the cabin, [completely naked,] with her arms open and a smile on her face. “How do I look, boys?” The sight of the woman, much younger and more beautiful than any they had ever seen, [completely] undressed and apparently offering herself to them, made Flash and Pete groan with a mixture of admiration and desire. Almost the same thing happened to them as to their partner and boss. Blinded by Linda’s beauty, they forgot for a moment about Joe and Bill. They crossed the threshold ready to take the girl in their arms, but no sooner had they done so than they saw Joe lying unconscious on the floor and heard the sound of a threatening voice behind them. “Hands up or I’ll shoot.” *** “Don’t you think it would be enough to make them careless,” he asked not daring to look at the young girl, “if you were suddenly to appear before them half undressed [completely naked]?” Linda went very red and struggled to fight off the tears that…

Example 3 shows both the galley proofs with the censor’s indications (top) and the corresponding published page (bottom). The original text reads as follows, with the words highlighted by the censor in bold: She was motionless, waiting for him to be through, or listening to herself; then restlessness stirred her and came through her arms. She raised her head. “It’s not fair,” she murmured. She drew a breath, her face shadowed and moving reluctantly inward. “It’s never fair.” The weight of her hands increased on his shoulders and when she reached his mouth he felt a second reaction against him and a halfhearted impulse to turn away. He held on. She seemed heavy to him, wanting him and not wanting him, undecided, liking it but not liking it, afraid. But her lips stayed with him and the flimsy little strings that held him back began to break and the rest of it came on, the blind thing, the unkind thing, the thing that had no conscience; the growling, bitter unbearable thing that stiffened his legs and closed



Carmen Camus Camus

Example 3.╇ File No. 3862/1954. Ernest Haycox: Abriendo Camino (The Earthbreakers) Galley proof (top); Published page (bottom)



Censorship in the translations and pseudo-translations of the West 

his arms around her and made him scheme to carry her beyond herself. She ceased to be heavy; she was a willow bending gracefully to him; she was a willow no more, she was angry with the same anger he had, she was motion in his arms, mouth feeding at him with strong appetite. She cried in her throat and flung her head aside. “Let me stop.” She came back in violent aggressiveness, fingers biting through his coat. She struck her head on his chin. “Make me stop.” He let go and watched her struggle with herself. She stared at him, she came nearer and ran her fingers across his forehead.

As can be seen from the suppressions in this extract, it was not so much simply passion that trespassed the bounds of the moral code but more especially, passion felt and exhibited by a woman, since this was not in keeping with the submissive docile role of women promoted by the regime and implemented through the censorship system. In a handwritten annotation to the typed report included in the censorship file, the censor responsible for reviewing the suppressions states: Estimo debe suprimirse el párrafo íntegro señalado y no las palabras entresacadas, como han hecho en la galerada “I consider that the whole paragraph indicated should be suppressed and not only the highlighted words, as they have done in the galley proofs”. This shows that the instructions given by the first censor had not been followed correctly, and reveals the close scrutiny that the texts were submitted to under the censoring procedure.

6. Conclusions This study has shown the usefulness of Translation Studies in revealing the influence of the Franco regime’s planning on culture. From this perspective, we can conclude that the enormous boom experienced by the Western in the period from 1946 to 1966, especially in the form of pseudo-translations, occurred as a result of the government’s desire to increase production and exports of autochthonous works. The analysis of the data indicates that behind this “innocuous” popularity of the Western there was a subtly structured State policy aimed at promoting the regime’s idea of the “new” Spain. The cultural planning was fostered in the form of the 1946 Book Promotion Law, which provided economic support to publishers by granting them special rates to buy paper and tax reductions to export their books, especially to Latin American countries. Publishers, eager to boost their sales, recruited people from outside the literary field to write, imitating models, especially of the popular Far West literature, thus helping to fill the void left by the exodus of Spanish intellectuals. These writers were asked to use pseudonyms to conceal their true identity from the reading public, and this use of a mask both covered up the lack of genuine authors and allowed writers with a dissident ideology to earn a living. Writing this kind of narrative was transformed into a kind of production system where writers were given some guidelines from editors as to what they could and what



Carmen Camus Camus

they should not include in their texts with regard to politics, religion and morality. What these publishers did was to establish an author production system to write these pseudo-translations. The autochthonous Western narrative not only served to fill the void left by the exodus of Spanish writers but also promoted the cultural change sought by the dictatorship.€Nevertheless, the data presented here show that for the Western, the influence of censorship was the same for both translated works and pseudo-translations, and that neither publishers nor individual writers were favoured by the censorship system. In general, the hand of the censors was relatively lenient on the genre, probably due to the control exercised by publishers and self-censorship of authors.

References Abellán, M.L. 1980. Censura y creación literaria en España (1939–1975). Barcelona: Ediciones Península. Camus, C. 2008. “Pseudonyms, pseudotranslation and self-censorship in the narrative of the West during the Franco dictatorship.” In Translation & Censorship: From the 18th Century to the Present Day, T. Seruya and M. Lin Moniz (eds), 147–162. Cambridge: Cambridge Scholars Press. Merino Álvarez, R. 2005. “From catalogue to corpus in descriptive Translation Studies. Translations censored under Franco. The TRACE Project.” RCEI, Revista Canaria de Estudios Ingleses: 85–104. [http://www.ehu.es/trace/publicaciones/2005aRMA_RCEI.pdf.] Rabadán, R. 2000. “Modelos importados, modelos adaptados: Pseudotraducciones de narrativa popular inglés-español 1955–1981.” In Traducción y censura inglés-español: 1939–1985, R.Rabadán (ed.), 255–277. León: Universidad de León. Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins. Toury, G. 2005. “Enhancing cultural changes by means of fictitious translations.” In Translation and Cultural Change, E. Hung (ed.), 7–13. Amsterdam/Philadelphia: John Benjamins.

A world without God Slovene Bambi Nike K. Pokorn

University of Ljubljana, Slovenia

This article presents a small part of a larger research project focussing on the specifics of translated juvenile fiction in a communist country. The reflection of ideology and values of the target society is demonstrated on the Slovene translation of Felix Salten’s Bambi (1950), one of the most popular books for children in the Socialist Republic of Slovenia in the period from the 1950s to 1980s. The analysis of the translation revealed an omission of the religious ending, i.e. an ideological intervention that was in line with similar interventions found in other translations of children’s literature created in socialist times. Based on the interviews, translation analyses and historical evidence, it is argued here that the consistent displacement of religious elements in juvenile translations in the socialist Slovenia is a result of self-censorship and that, despite the absence of any kind of official censorship, the translator of Bambi and other translators of children’s literature were lead to act in an almost uniform way by a specific translator’s habitus. Keywords: socialist translation, self-censorship, children’s literature

1. Introduction The analysis of Slovene Bambi is a part of a larger research project focussing on a reconstruction of the translatorial field, in particular on the field of translated juvenile fiction, in the communist period in Slovenia; that is, the period which extended from 1945 (the end of the Second World War, when Slovenia became one of the six republics within the Socialist Federal Republic of Yugoslavia) to 1992 (when Slovenia became an independent state). Translations of children’s literature were chosen because in many cultures children’s literature is one of the genres most susceptible to considerable change and manipulation through translation (see e.g.€ Puurtinen 1992; Du-Nour 1995; Fernandez Lopez 2000; Malmkjaer 2003 and 2004; Desmidt 2006; Oittinen 2006). This acceptance of modifications most probably stems from the conviction that



Nike K. Pokorn

children’s literature, including the works that are not openly didactic, should not be harmful to the development of children into ideal citizens or individuals – and, since the concept of an ideal adult is not a stable term, translations of children’s literature are often very clear reflections of the ideology of a particular TL culture in a particular time. It is also argued here that translators of children’s literature acquire a specific habitus (as defined by Bourdieu 1990:€91; Simeoni 1998; Gouanvic 2005) that is tuned to the specific demands of translated juvenile fiction.

2. The role of Translation Studies Although the article talks about ideology in Slovene culture, it should be stressed that the concept of culture is not regarded as a monolithic entity, constituted by one ideology, but is seen as inherently unstable, relativistic and always shifting in meaning and shape (cf.€ Greenblatt 1992). This perpetually unstable and control-resisting culture, however, is not regarded as immune to hegemonic discourses: societies and schools can still be discriminatory, certain values and practices are still defined as cultural capital required for success in dominant institutions, and certain ideologies can still try to present themselves as stable and eternal.€The invaluable role of Translation Studies is therefore seen in the examination of how the types of hegemonic power find expression in translation practice; moreover, the role of TS is seen not only in the celebration of hybridity, but above all in the struggle against hegemonic discourses in order to reconstruct a more democratic order. This role of TS is not a new one: already decades ago prominent TS scholars (e.g.€Lefevere 1990; Hermans 1985) showed us that translations can not only tell us about the self-image of a particular culture and the changes that self-image undergoes, but also about the strength of a poetics and ideology pertaining to this perpetually changing culture. It was shown that the critical study of translations could reveal the way different cultures interact and, even more importantly, the ways texts had been manipulated through translation (for a definition of manipulation in translation, see Malmkjær 2004). Since the dominant poetics, ideologies, and other hegemonic discourses constantly reinvent the ways they impose themselves in various levels of society, the study of ideological shifts in translation remains a perpetual and much needed goal of TS research. By showing how manipulative shifts take place in translations, translatological research provides an insight into the processes that shape our lives, and “teach us a few things not just about the world of literature, but also about the world we live in” (Lefevere 1990:€ 27). Consequently, such findings are not only intriguing for the TS scholars, but prove to be extremely valuable and interesting for the general public as well (see also Lefevere 1992:€51). Moreover, the revelations of hidden ways of expression of hegemonic discourses does not seem to be the only goal of this kind of TS research, although we must agree with Susan Bassnett and André Lefevere that there is a



A world without God 

certain pleasure in the mere realisation that even if we may not like what we see, “at least we shall not be kept in the dark” (Lefevere and Bassnett 1990:€13). The motor that keeps such research going is a conviction that greater insight into the mechanism of the society and greater understanding of the processes that shape our lives can lead us to greater resistance to new forms of hegemonic discourses. Such a goal inevitably surpasses by far the interests of the relatively small TS research community and joins forces with much larger and continual efforts to uphold and increase the level of democratic order in every society.

3. Methods The research project described above integrates results from translation analysis, interviews, and historical and archival investigation. Thus, first, the most exhaustive Slovene electronic online bibliographic source (COBISS, www.cobiss.si) and the printed Slovene bibliography for the period from 1945 to 1963 were checked for all translations of children’s literature. Of these, those translations that were reprinted or retranslated in later periods were selected in order to focus attention only on those works that proved successful. Those translations were then compared to the originals, reprints and later translations, and, finally, biographies of translators and editors were studied, and interviews were made with selected translators and editors with an aim to identify and explain any possible ideological interventions (i.e. the intentional choices, motivated by a set of well-articulated principles and beliefs, see Malmkjær 2004 for the distinction between censorship and error) used in the socialist period that differ from the interventions used in subsequent, post-socialist period from 1992 onwards. This paper shows reflection of ideology and values of the target society on the Slovene translation of Felix Salten’s Bambi (1923), a book that used to be a part of the cultural capital of Slovene children from 1950s throughout to 1980s.

4. Salten’s Bambi The author, Felix Salten (1869–1945), was born in a Jewish family as Siegmund Salzmann in Budapest, and moved as an infant with his parents to Vienna. Despite little formal education, Salten became a member of the Jung Wien movement and wrote letters, poems, essays, and short stories that were published in newspapers and periodicals in both Vienna and Berlin. In 1923, he wrote his best-known novel Bambi: A Life in the Woods (Bambi: Eine Lebensgeschichte aus dem Walde). In the late twenties his reputation grew, so that from 1927 to 1933 he even became the president of the Austrian writers’ association. When Germany annexed Austria, Salten managed to escape to Zürich, Switzerland, where he lived until his death in 1945.

 Nike K. Pokorn

In 1929, only five years after its creation, Salten’s Bambi was translated into English by Whittaker Chambers (1901–1961), a prominent American journalist and later a communist spy, who at that time needed to supplement his income while working at a communist newspaper (WhittakerChambers Org). His translation immediately became a Book-of-the-Month Club hit. Soon after that, Thomas Mann brought Bambi to the attention of Walt Disney, and in 1933 Salten sold the film rights to Disney studios which spent five years refashioning Salten’s novel into an immensely successful animated film that premiered in Britain and the United States in August 1942. Salten had the idea of writing Bambi while on vacation in the Alps after being charmed by the wildlife there, so it is no wonder that readers from Alpine Slovenia found it so appealing. Bambi (the name is based on the Italian word bambino) is the name of the main character, a male roe deer beginning life as a fawn, then an adolescent spike, and finally a buck. It is a classic coming-of-age novel, following a deer from birth to adulthood, realistically depicting the dangers and harshness of nature and the cruelty of human hunters. The following excerpt is taken from a longer description of a hunt during which the game is being driven into the open: “This way!” called someone with a gasping voice. Bambi obeyed involuntarily and found an opening at once. Someone moved feebly in front of him. It was Friend Hare’s wife who had called. “Can you help me a little?” she said. Bambi looked at her and shuddered. Her hind leg dangled lifelessly in the snow, dyeing it red and melting it with warm, oozing blood. “Can you help me a little?” she repeated. She spoke as if she were well and whole, almost as if she were happy. “I don’t know what can have happened to me,” she went on. “There’s really no sense to it, but I can’t seem to walk…” In the middle of her words she rolled over on her side and died. Bambi was seized with horror again and ran. (Salten [1928] 2002:€97–98)

The young Bambi thus discovers the beauties and atrocities of nature; he observes the cruelty of human beings, who represent a constant threat to all the inhabitants of the woods. The hunters also kill his mother and take his friend Gobo away. He is forced to survive on his own, and finally manages to win a fight and earn the friendship and partnership of his long-term friend, Gobo’s sister Felina (Faline in the English translation). Throughout his life he is being watched over by the greatest roe deer in the woods, his father, the solitary stag, called the old Prince, der alte Fürst. The Old Stag is his mentor in many ways: he teaches him how to avoid hunters, how to mislead them, and, above all, he introduces his young disciple to the awareness of the importance of solitude and leads him towards the “final enlightenment”, which is revealed in the final pages of the book. This final truth is connected with the false status of human beings in nature. All the animals consider human beings to be god-like creatures. For example, this is how a dog describes him: The dog stared around him. “Who are you?” he yelped. “What do you want? What do you know about it? What are you talking about? Everything belongs to Him,



A world without God 

just as I do. But I, I love Him. I worship Him. I serve Him. Do you think you can oppose Him, poor creatures like you? He is all-powerful. He’s above all of you. Everything we have comes from Him! Everything that lives or grows comes from Him!” The dog was quivering with exaltation. (Salten [1928] 2002:€180 – 181)

However, in the last pages of the novel, the Old Stag leads the grownup Bambi to a human corpse, a dead poacher, and reveals to him the final wisdom that the belief in the superiority of human beings is false: “Bambi”, fuhr der Alte fort, “erinnerst du dich an das, was Gobo gesagt hat, an das, was der Hund gesagt hat, an das, was sie alle glauben... erinnerst du dich?” Bambi vermochte nicht zu antworten. “Siehst du wohl, Bambi”, sprach der Alte weiter, “siehst du nun, daß Er daliegt, wie einer von uns? Höre, Bambi, Er ist nicht allmächtig, wie sie sagen. Er ist es nicht, von dem alles kommt, was da wächst und lebt, Er ist nicht über uns! Neben uns ist Er und ist wie wir selber, und Er kennt wie wir die Angst, die Not und das Leid. Er kann überwältigt werden gleich uns, und dann liegt Er hilflos am Boden, so wie wir anderen, so wie du Ihn jetzt vor dir siehst.” Eine Stille war. “Verstehst du mich, Bambi?” fragte der Alte. Bambi erwiderte flüsternd: “Ich glaube…” Der Alte gebot: “So sprich!” Bambi erglühte und sprach bebend: “Ein anderer ist über uns allen… über uns und über Ihm.” “Dann kann ich gehen”, sagte der Alte. (Salten [1923] 1930:€246–247) “Bambi,” the old stag said, “do you remember what Gobo said and what the dog said, what they all think, do you remember?” Bambi could not answer. “Do you see, Bambi,” the old stag went on, “do you see how He’s lying there dead, like one of us? Listen, Bambi, He isn’t all-powerful as they say. Everything that lives and grows doesn’t come from Him. He isn’t above us. He’s just the same as we are. He has the same fears, the same needs and suffers in the same way. He can be killed like us, and then He lies helpless here on the ground like all the rest of us, as you see Him now.”



There was silence. “Do you understand me, Bambi?” asked the old stag. “I think so,” Bambi said in a whisper. “Then speak,” the old stag commanded.

Bambi was inspired, and said trembling, “There is Another who is over us all, over us and over Him.” “Now I can go,” said the old stag. (Salten [1928] 2002:€187–188)



Nike K. Pokorn

5. Slovene Bambi There is only one Slovene translation of the entire text by Felix Salten. It was first published in 1953 in the series Knjižica Sinjega Galeba and then reprinted in 1970 in the influential book series Biseri, published in 6,000 copies by the state-owned publishing house, Mladinska knjiga, in Ljubljana. This publishing house was founded immediately after the war in 1945 with the aim of focusing on publication of juvenile fiction, but soon also started covering other genres. The translator of both editions was Kristina Brenk. Kristina Brenk was born in 1911 in Horjul and was educated in a nunnery where, during her seven-year stay, she learned German, French, Esperanto and stenography. When she was 17, she left the nunnery for good only 48 hours before consecration. In Ljubljana she started studying at the teachers’ training college and received a degree in education in 1938 at the age of 27. During the Second World War, she joined the Slovene resistance movement, and was consequently imprisoned in 1943. After the war, in 1949, married and with children, she started working as the editor for children’s literature at the newly established publishing house Mladinska knjiga, where she remained the editor until 1973 when she retired. She never became a member of the Communist Party, although she was repeatedly asked to do so (Hostnik 2007). Kristina Brenk not only edited books for young readers, she also wrote books for children, promoted Slovene illustrators, helped many intellectuals that were ostracized by the society of that time by giving them jobs as translators (e.g., Ludvik Mrzel), and, last but not least, translated many of the works herself. Now, age 97, she lives as an honorary citizen in Ljubljana. Both editions of Slovene Bambi had a foreword by John Galsworthy and Kristina Brenk. The later edition does not change the translated text; the only difference between the two editions is that the second edition was published with illustrations by the Slovene illustrator Ančka Gošnik-Godec. Brenk’s introduction typically (see Lathey 2006) provides an insight into a particular image of the child reader and emphasizes the didactic aspect of the work: Salten’s books will be a pleasure to everyone who is seriously and sincerely concerned about the growth and formation of the young generation, in particular to teachers and parents when they look for presents for their children or pupils, wanting to offer a book that would encourage them in life with a right thought. (Brenk 1970:€150. This and all the subsequent translations from Slovene are by the author of the article.)

The Slovene translation, which is otherwise very faithful and does not omit any of the instances of cruelty on behalf of other animals or human beings, changes only the final, but significant passage: “Bambi,” nadaljuje Starec, “ali se spominjaš, kaj je dejal Gobo, ali se spominjaš, kaj je dejal pes in kar vi vsi verjamete? Ali se spominjaš?”



A world without God 

Bambi ne more odgovoriti. “Poglej, Bambi,” nadaljuje Starec, “poglej, da leži Človek prav tako na tleh kakor kdo izmed naših. Poslušaj, Bambi! Človek ni vsemogočen, kakor pravijo. Iz Človeka ne izhaja vse, kar raste in živi. Človek ni nad nami. Človek je poleg nas in je tak, kakršni smo mi, kajti Človek pozna prav tako kakor mi, strah, stisko in trpljenje. Človek je lahko premagan kakor mi in tedaj leži, prav tako kakor mi, nebogljen na tleh, leži tako, kakor ga zdajle vidiš pred seboj.” Dolgo molčita. “Ali me razumeš?” povpraša Starec. Šepetaje odgovori Bambi: “Razumem...” Starec ukaže: “Tedaj govôri!” Bambi se razvname in strastno odgovori: “Človek ni nad nami… Človek ni vsemogočen. ...” “Potem lahko odidem,” reče Starec. (Salten [1953] 1970:€145 – 146, emphasis added) “Bambi,” the old one says, “do you remember what Gobo said and what the dog said, what you all think? Do you remember?” Bambi cannot answer. “Do you see, Bambi,” the old stag goes on, “do you see how Man is lying there, like one of us? Listen, Bambi! Man isn’t all-powerful as they say. Everything that lives and grows doesn’t come from Man. Man isn’t above us. Man is next to us and he’s just the same as we are; Man has the same fears, the same needs and suffers in the same way. Man can be conquered like us, and then He lies helpless here on the ground like all the rest of us, as you see Him now.” There was a long silence. “Do you understand me?” asks the Old Stag. Bambi replies, whispering, “I understand…” The Old One commands: “Then speak!” Bambi passionately replies: “Man is not above us… Man is not almighty…” “Then I can go,” the Old One says.

Not only does the Slovene translation attenuate the religious overtones present in the original by replacing the personal pronoun with the noun človek ‘man’, the change of this final paragraph is particularly significant because it changes the final wisdom of the Old Stag. In the original and in the English translation, the final wisdom is that nature with all the animals and human beings is subordinate to a transcendental being that surpasses them all (“Ein anderer ist über uns allen … über uns und über Ihm.” (Salten [1923] 1930:€247); “There is Another who is over us all, over us and over Him.” (Bambi [1928] 2002:€187–188)), whereas in the Slovene version the final wisdom is only that human beings are as relative as other beings and are not superior, but equal to them. There is no mention of the existence of any other transcendental being; this position is in accordance with the then dominant communist ideological position that regarded Judeo-Christian belief as something obsolete and negative, as “the opium of the people” (“Die Religion... ist das Opium des Volkes”) as described by Karl Marx in the



Nike K. Pokorn

introduction to his work Contribution to Critique of Hegel’s Philosophy of Right in 1843. Because Marx argued that the “abolition of religion as the illusory happiness of the people is the demand for their [of the people] real happiness” (ibid.), religion consequently represented something that might have had a negative influence on the developing mind of socialist children.

6. Conclusions The omission of the religious ending in this translation of Bambi corresponds to the findings in other analysed translations in the corpus which consists of children’s literature proper and the literature originally written for adults but later adopted by the juvenile audience: up to now, selected tales of the Brothers Grimm, Robinson Crusoe by Daniel Defoe (1952, 1974), Uncle Tom’s Cabin by Harriet Beecher-Stowe (1954), The Wind in the Willows by Kenneth Grahame (1961), Treasure Island by Robert Louis Stevenson (1950, 1956, 1983), Pippi Longstocking by Astrid Lindgren (1962), Heidi by Johanna Spyri (1954, 1982) and Winnetou by Karl May (1952/53, 1962, 1983) have been analysed. Although these works were created by different translators, had different editors and were published by different publishing houses, all of them revealed the application of similar translatorial strategies directed against the same kind of elements; in all of the works that were translated in the socialist period (i.e. from 1945 to 1992), all religious elements (in particular those of Judeo-Christian origin) in translations for children were omitted or replaced by other non-Christian items. The religious elements, however, were generally not omitted or replaced in literature for the adults (e.g.€in Robinson Crusoe, published in 1974). In an interview with Kristina Brenk on 8€November 2006, she confirmed that there was no censor employed at the publishing house and that there was no censor’s office at the state level either. The general manager of the publishing house, however, had to sign his approval of every new publication, but it was possible (although dangerous) to avoid that as well. She, as the editor, did not change or “cleanse” translations either; in fact, she claims that she trusted the people she asked to do translations for her. She does not remember any political interventions on her work, the only exception being when Mladinska knjiga published Biblical Stories and the general manager of the publishing house called her to his office. She said that she was able to explain that Biblical stories belong to the world’s cultural heritage and therefore merit being presented to children as well. She knew, however, that certain works and certain passages might cause problems for her and for translators. When directly asked why she had changed things in her translation of Pippi Longstocking (where she replaced Christmas with New Year’s Eve) and Bambi, she replied that she did not remember that she had changed anything. This non-awareness should not be attributed to her advanced age only. The archival material and other historical evidence support her words because there is no evidence that might suggest that there existed in Slovenia a censor’s office



A world without God 

like the one Gaby Thomson-Wohlgemuth (2006a, 2006b) describes in East Germany. The displacement of the religious element in Bambi should therefore be attributed not to any kind of post-censorship or punitive, repressive censorship, but to a form of preventive or prior censorship, i.e. to self-censorship by the translator (for a more detailed discussion of various meanings of the term, see Merkle 2002; Wolf 2002; Gambier 2002). The absence of any official censorship could be that, in Bourdieu’s terms (cf.€1980:€91), the Yugoslav and Slovene socialist society of the time seemed to be politically stable enough that the translatorial habitus ensured the perpetuation of the dominant discourse without the use of coercive measures. Translation is a complex activity for which there must be multiple sources of explanation and there can be no guarantee that any single cause can explain all the facts (see Pym 1998:€158). In view of the fact that the conjunction of a set of conditions is normally needed to supply a sufficient condition (see Brownlie 2003:€112; also Sturge 2002:€167), there is also a possibility that Brenk’s decision to delete any religious presence at the end of the novel was not due to the fact that she had no other choice, as Lefevere put it (1992:€13), but was also the result of her personal interpretation of the text or of her personal ideology. Indeed, the biographical data show that Brenk had an ambivalent attitude towards Christian religion: on one hand, she detested the oppressive atmosphere in the nunnery she had experienced as a young girl, on the other hand she managed to promote the publication of Biblical stories for children in a time that was openly opposed to religious education and when the Theological faculty, for example, was excluded from the only university in the Republic. However, the fact that all translators of children’s literature omitted or replaced religious elements, in particular those belonging to the Judeo-Christian tradition, from the translations of children’s literature in the period when Slovenia belonged to socialist Yugoslavia (e.g.€the Slovene translation of Heidi, the book that was edited by Kristina Brenk, consistently replaces the original “pastor” to “teacher”, see Kocijančič Pokorn 2005), suggests that this conformity should primarily be attributed to specific social conditions. Furthermore, the particular “communist” translation strategy was the result of the self-censorship by the translators who, consciously or subconsciously, internalised the communist attitude towards religion. The Slovene translation of Bambi thus shows not only that the translator’s habitus inevitably reflects the social conditions within which the translator acquired a particular set of dispositions which incline the translator to act and react in certain ways, but also that it is the elaborate result of a personalized social and cultural history (cf.€ Simeoni 1998). Regardless of whether the translators of children’s literature in communist times accommodated to the established order to avoid negative consequences or sanctions, or because they supported the communist attitude towards religion, the results show that the translator’s habitus led Kristina Brenk and other translators of children’s literature in communist Slovenia to always act in a similar way. Despite the absence of an official state censorship, the Yugoslav and Slovene socialist society of the time assured sufficient indirect pressure to achieve social conformity in



Nike K. Pokorn

the field of translation of children’s literature. Moreover, Kristina Brenk seems to indicate that at least some translators of children’s literature in socialist times did not regard their translatorial habitus as limitations or some imposed rules; on the contrary, they became their second nature.

References Primary sources Beecher Stowe, H. 1954. Koča strica Toma ali življenje črncev v suženjskih državah Amerike. Transl. by Olga Grahor, adapted by Kristina Brenkova. Ljubljana: Mladinska knjiga. Brenk, K. [1953] 1970. “Beseda ob slovenskem Bambiju.” In Bambi, F. Salten, 149–150. Ljubljana: Mladinska knjiga. Defoe, D. 1952. Življenje in čudovite prigode pomorčaka Robinzona Kruzoa. Transl. Vinko Gabrski. Ljubljana: Mladinska knjiga. Defoe, D. 1974. Robinson Crusoe. Transl. Mira Mihelič. Ljubljana: Cankarjeva založba. Grahame, K. 1961. Veter v vrbju. Transl. Jože Dolenc. Ljubljana: Mladinska knjiga. Lindgren, A. 1962. Pika Nogavička. Transl. Kristina Brenkova. Ljubljana: Mladinska knjiga. May, K. 1952 – 53. Winnetou. Transl. by Oskar Hudales. Maribor: Obzorja. May, K. 1962 [1967, 1968]. Vinetou. Transl. and adapted by (P. S.) Ludvik Mrzel. Ljubljana: Mladinska knjiga. May, K. 1983. Vinetou. Trans. by Jože Dolenc. Ljubljana: Mladinska knjiga. Salten, F. [1928] 2002. Bambi: A Life in the Woods. Transl. by Whittaker Chambers. New York: Simon & Schuster. Salten, F. [1953] 1970. Bambi. Transl. Kristina Brenkova. Ljubljana: Mladinska knjiga. Salten, F. 1931. Bambi: Eine Lebensgeschichte aus dem Walde. Berlin, Vienna, Leipzig: Paul Zsolnay Verlag. Spyri, J. 1954 [1982]. Heidi. Transl. Meta Sever. Ljubljana: Mladinska knjiga. Stevenson, R.L. 1950 [1956, 1983]. Otok zakladov. Transl. Pavel Holeček. Ljubljana: Mladinska knjiga.

Secondary sources Biography Base. 2004. http://www.biographybase.com/biography/Salten_Felix.html (accessed December 23, 2008) Bourdieu, P. 1980. Le sens pratique. Paris: Minuit. Bourdieu, P. 1990. In Other Words: Essays Toward a Reflexive Sociology. Stanford: Stanford University Press. Brownlie, S. 2003. “Investigating explanations of translational phenomena: A case for multiple causality.” Target 15 (1): 111–152. Desmidt, I. 2006. “A Prototypical Approach within Descriptive Translation Studies? Colliding Norms in Translated Children’s Literature.” In Children’s Literature in Translation: Challenges and Strategies, J. Van Coillie and W. P. Verschueren (eds), 79–96. Manchester/New York: St Jerome.



A world without God  Du-Nour, M. 1995. “Retranslation of Children’s Books as Evidence of Changes of Norms.” Target 7 (2): 327–346. Encyclopædia Britannica, 2008. s.v. “Felix Salten” and “Chambers, Whittaker”. Chicago: Encyclopædia Britannica. Fernandez Lopez, M. 2000. “Translation Studies in contemporary children’s literature: comparison of intercultural ideological factors.” Children’s literature association quarterly: 29–37. Gambier, Y. 200. “Les censures dans la traduction audiovisuelle.” TTR: traduction, terminologie, rédaction 15 (2): 203–221. Gouanvic, J.-M. 2005. “A Bourdieusian Theory of Translation, or the Coincidence of Practical Instances: Field, ‘Habitus’, Capital and ‘Illusio’.” The Translator 11 (2): 147–166. Greenblatt, S. 1992. Marvelous Possesions: The Wonder of the New World. Oxford: Clarendon Press. Hermans, T. (ed.). 1985. The Manipulation of Literature: Studies in Literary Translation. London, Sydney: Croom Helm. Hostnik, M. 2007. “Nisem bila samo firbčna, bila sem tudi zelo izbirčna.” Interview with Kristina Brenk. Dnevnikov objektiv, Dnevnik (19€May). Kocijančič Pokorn, N. 2005. “Ste brali Heidi? Primer (post)socialistične cenzure prevodov.” In Beyond equivalence [Graz translation studies 9], N. Kocijančič Pokorn, E. Prunč and A. Riccardi (eds), 57–72. Graz: Institut für Theoretische und Angewandte Translationswisenschaft. Lathey, G. 2006. “The Translator Revealed: Didacticism, Cultural Mediation and Visions of the Child Reader in Translators’ Prefaces.” In Children’s Literature in Translation: Challenges and Strategies, J. Van Coillie and W. P. Verschueren (eds), 1 – 18. Manchester/New York: St Jerome. Lefevere, A. and Bassnett, S. 1990. “Introduction: Proust’s Grandmother and the Thousand and One Nights: The ‘Cultural Turn’ in Translation Studies.” In Translation: History & Culture, A. Lefevere and S. Bassnett (eds), 1–13. London: Pinter. Lefevere, A. 1990. “Translation: Its Genealogy in the West.” In Translation: History & Culture, A. Lefevere and S. Bassnett (eds), 14–28. London: Pinter. Lefevere, A. 1992. Translation, Rewriting, and the Manipulation of Literary Fame. London/New York: Routledge. Malmkjaer, K. 2003. “What happened to God and the angels? An exercise in translational stylistics.” Target 15 (1): 37–58. Malmkjaer, K. 2004. “Censorship or error: Mary Howitt and a problem in descriptive TS.” In Claims, Changes and Challenges in Translation Studies, G. Hansen, K. Malmkjaer and D. Gile (eds), 141–156. Amsterdam, Philadelphia: John Benjamins. Marx, K. 1843. “Introduction to A Contribution to the Critique of Hegel’s Philosophy of Right. Karl Marx in Deutsch-Französische Jahrbücher, February, 1844.” (http://www.marxists.org/ archive/marx/works/1843/critique-hpr/intro.htm), (accessed October 22, 2008). Merkle, D. 2002. “Presentation.” TTR: traduction, terminologie, rédaction 15 (2): 9–18. Oittinen, R. 2006. “No Innocent Act: On the Ethics of Translating for Children.” In Children’s Literature in Translation: Challenges and Strategies, J. Van Coillie and W. P. Verschueren (eds), 35 – 45. Manchester/New York: St Jerome. Puurtinen, T. 1992. “Dynamic style as a parameter of acceptability in translated children’s books.” In Translation Studies: An Interdiscipline, K. Kaindl, F. Pöchchacker and M. Snell-Hornby (eds), 83–90. Amsterdam/ Philadelphia: John Benjamins. Pym, A. 1998. Method in Translation History. Manchester: St. Jerome. Simeoni, D. 1998. “The Pivotal Status of the Translator’s Habitus.” Target 10 (1): 1 – 39. Sturge, K. 2002. “Censorship of Translated Fiction in Nazi Germany.” TTR: traduction, terminologie, rédaction 15 (2): 153–169.



Nike K. Pokorn Thomson-Wohlgemuth, G. 2006a. “Translation from the point of view of the East German censorship files.” In Sociocultural Aspects of Translating and Interpreting, A. Pym, M. Shlesinger and Z. Jettmarová (eds), 53 – 64. Amsterdam/ Philadelphia: John Benjamins. Thomson-Wohlgemuth, G. 2006b. “Flying High: Translation if Children’s Literature in East Germany.” In Children’s Literature in Translation: Challenges and Strategies, J. Van Coillie and W. P. Verschueren (eds), 47 – 59. Manchester/New York: St Jerome. WhittakerChambers Org. www.whittakerchambers.org/translations.html (accessed December 23, 2008) Wolf, M. 2002. “Censorship as Cultural Blockage: Banned Literature in the Late Habsburg Monarchy.” TTR: traduction, terminologie, rédaction 15 (2): 45–61.

Manipulating the matricial norms A comparison of the English, Swedish and French translations of La caverna de las ideas by José Carlos Somoza Yvonne Lindqvist

Stockholm University, Sweden

The paper examines the “crime fiction boom” and the cultural spaces from which translated crime fiction originates within the Swedish cultural system. This introduction contextualizes the Swedish translation through comparison with the English and French translations of the Spanish crime novel La caverna de las ideas. The study draws on the polysystem theory approach (Even-Zohar 1990; Toury 1995, 1998) and the cultural sociology of Pierre Bourdieu (Bourdieu 1986, 1992a; Gouanvic 1997, 1999) in putting forward the hypothesis that the meta-textual elements in the novel will be translated differently, depending on the cultural system within which the translators perform their task. It is shown that the English translator manipulates the matricial norms of the source text to a higher extent than do the Swedish and French translators. Moreover, the paper examines some of the international consecration processes within the global literary space (Casanova 1999), which made it possible for the Somoza crime novel to reach the Swedish literary system by means of translation. Keywords: literary translation, polysystem, literary field, crime fiction, metatext, Spanish, English, French, Swedish

1. Introduction Much has been said recently in Sweden about the “boom” in Swedish crime fiction. Highbrow literary critics and authors fear the “invasion” of translated crime fiction to the detriment of more serious literature (Högström 2007; Brunner 2007; Håkansson 2007). They accuse the Swedish publishing houses of acting short-sightedly in pursuit of rapid financial profits neglecting their responsibility to invest in cultural capital.€This paper initially examines whether this “crime fiction boom” in Sweden does really exist.



Yvonne Lindqvist

It strives to reconstruct the Swedish literary field of crime fiction in contextualizing the Swedish translation of the Spanish crime novel La caverna de las ideas. A compairison to the English and French translations of the same novel is also carried out. The hypothesis put forward in this paper formulates the claim that meta-textual elements in the novel will be translated differently depending on the overall makeup of the cultural system within which the translators perform their task. The British, the Swedish and the French cultural systems differ substanÂ�tially in terms of their overall translation policy, and hence with regard to the position of translated literature within these cultures (Venuti 1996:€11–13). On the one hand, the Swedish cultural and literary systems are extremely open systems, where translated literature has a confirmed central position. ApproxiÂ�mately 50% of all published fiction in Sweden consists of translations (Lindqvist 2005:€ 29). By comparison, the British systems on the other hand, are definitely the most closed systems in this respect, with a rate of only 2€percent for published translated literature, while the French systems hold a middle position, at approximately 12% (Venuti 1996:€12).

2. The “boom” in Swedish crime fiction? A quick statistical overview of the publication of crime fiction in Sweden during the last 15 years reported in periods of five years clearly shows an increase in terms of the total number of publications: Judging from the statistics in Table€1 it is true that there has been an augmentaÂ�tion in crime fiction publication rates during the last 15 years, but considering the total numbers for the 5-year periods the increase is not sufficiently dramatic to justify the fear expressed by the literary critics. From the first period to the last, the increase consists of 111 publications, i.e. approximately 20 publications a year and thus less than 2 novels a month – not exactly a “boom” to fear. The augmentation of the crime novels written in the Swedish language from 28% to 44% is rather surprising, however. Therefore, there should be no fear with regard to translations in the field, or an “invasion” of crime fiction into other literary fields. There is no compensational mechanism between translated and originally produced crime fiction and between crime fiction and more “serious” literature. On the contrary, the total number for published fiction in Table 1.╇ Crime fiction publication rates in Sweden during the period 1990–2004 reported in five-year periods (Wopenka 2004:€22–24) Origin/Year

Swedish

Translations

Total

% Swedish novels

1990 – 1994 1995 – 1999 2000 – 2004

191 224 350

488 459 440

679 683 790

28 33 44



Manipulating the matricial norms 

Sweden has increased recently, probably due largely to new printing techniques and the general tax reduction on books introduced in 2002 (Nationalbibliografin 2007). Instead, the development of the crime fiction genre in Sweden reveals the relational character of literary systems or fields (Bourdieu 1986; Lindqvist 2002). Translated literature dominates at a particular moment in a specific literary field, serving as a supplier of repertoire and enhances, over time, the production of original literature (cf.€Gouanvic 1997:€33–45). It is probable that this development will occur if there has existed a prior strong tradition within the specific genre in the literary field in question. This is the case in the Swedish literary field of crime fiction, where the first “golden age” came about in the fifties with authors such as Vic Suneson, Stieg Trenter and Maria Lang. The seventies saw the rise of the second “golden age”, exemplified by authors as Olov Svedelid and Sven Sörmark (Broberg 2005:€8) – these authors having mainly written for the national market. The most important writers during the seventies were, however, the successful duo Maj Sjöwall and Per Wahlöö. They opened up the international market for Swedish authors of crime fiction. Sjöwall/Wahlöö won great success on European markets after receiving, in 1971, the Edgar Allen Poe Award presented each year by the Mystery Writers of America. Together they created the archetypical character for the Swedish police crime novel, namely Martin Beck, who suffers from the awareness that something is rotten in Swedish welfare society. The Sjöwall/Wahlöö novels thus initiated the strong tradition of social realism within Swedish crime fiction. We are currently witnessing the third “golden age” of Swedish crime fiction. But this genre’s unquestionable “boom” is taking place abroad. Of the approxiÂ�mately 70 living Swedish crime fiction authors today, 53 have been translated into an average of 7 languages (Widerberg 2004:€4). The leading authors in this respect are Henning Mankell (35 languages), Maj Sjöwall (34), Liza Marklund (26), and Håkan Nesser (25).

3. The source text and the target texts of La caverna de las ideas La caverna de las ideas by the Spanish author José Carlos Somoza, is the source text for the comparison of three translations from different types of cultural and literary systems (cf.€Table€2). In the year 2000, Somoza was short-listed for the “Premio Nadal”, one of the most important literary prizes in Spain. In 2002, The Athenian Murders was presented with the Macallan Gold Dagger Award by the UK Crime Writers Association. The Athenian Murders was the first of the novels by Somoza to be translated into English, French and Swedish. This study is based on the paperback editions from each of these cultures. These paperback editions appeared only one year after the publication of the hard-cover editions, which is an indication of the international success of this novel.



Yvonne Lindqvist

Table 2.╇ The source text of the study, La caverna de las ideas, the titles of the British, Swedish and French target texts, the names of the translators and the publishing houses and the years of publication of the hard-cover and paperback editions1 La caverna de las ideas José Carlos Somoza 2000. Alfaguara Paperback 2006 Biblioteca Debolsillo, second edition. Nr.€578 The Athenian Murders Translated by Sonia Soto

2002 Abacus Publishing House

Paperback 2003

Abacus1

Idéernas grotta Translated by Karin Sjöstrand

2004 Leopard

Paperback 2005

Månpocket

La caverne des idées Translated by Marianne Millon

2002 Actes Sud

Paperback 2003

Babel nr 604

The title, La Caverna de las ideas, was translated literally into Swedish as Idéernas grotta and was first published in 2004 by the Leopard publishing house – a relatively new publishing house located in Stockholm. The explicitly stated policy of Leopard is to mirror the dramatic changes in the Swedish society by publishÂ�ing social debate books and third world literature. Curiously this publishÂ�ing house is owned and financed by the highest selling crime novelist in Sweden ever, Henning Mankell. The activities of the Leopard publishing house are important to the Swedish book market, since their publishing policy favours languages other than English as source languages for translations. English is the main source language for translated fiction in Sweden. In 2004, as much as 74% of all translated fiction was translated from English. In the same year, the percentage for Spanish as source language amounted to 1% (Johansson 2005). In 2004, La caverna de las ideas was one of the few novels translated into Swedish from Spanish. In the same year, this novel was selected for the Swedish annual crime fiction award Flintyxan at the Gothenburg Book Fair.

1.

Reprinted 2003 four times and twice 2004 & 2005.



Manipulating the matricial norms 

4. The novel La caverna de las ideas The novel, La caverna de las ideas, largely fulfils the criteria of post-modern narration as defined by Umberto Eco (2004:€186), since it is constructed by means of meta-narrativity, dialoguicity, double coding and inter-textual irony. The novel is, in fact, a meta-translation, i.e. a translated text commenting on its own genesis. An anonymous translator translates an ancient manuscript (already interpreted by a certain savant, named Montalo), and he rewrites it before our eyes. The various (meta) levels of the novel are signalled in the very first line of the narration, in the title of the first chapter, which begins with the Roman numeral I and an asterisk (except in the English translation) directing the attention of the reader to the bottom of the page where a translator’s footnote states: 2 Source text Spanish I* *Faltan las cinco primeras líneas. Montalo, en su edición del texto original, afirma que el papiro había sido desgarrado en este punto. Comienzo mi traducción de La caverna de las ideas en la primera frase del texto de Montalo, que es el único del que disponemos. (N. del T.) Translations English I3 1. The first five lines are missing. In his edition of the original text, Montalo noted that the papyrus is torn here. I’m beginning my translation of The Athenian Murders with the first sentence of Montalo’s text, which is the only version we have. (Translator’s Note.) Swedish I* *De första fem raderna saknas. Montalo hävdar i sin utgåva av originÂ�alÂ�Â�texten att papyrusen var avriven där. Jag börjar min översättning av Idéernas grotta med den första meningen i Montalos text, som är den enda version vi har. Ö.a. French I* Les cinq premières lignes sont manquantes. Dans son édition du texte original, Montalo affirme que le papyrus a été déchiré à cet endroit. Je commence ma traduction de La Caverne des idées à la première phrase du texte de Montalo, qui est le seul que nous ayons à notre disposition. (N.d.T.) The subtle narratological construction of the novel thus includes a translator dialogically interacting with the reader by means of an advanced system of footnotes – a 2. The reader of this article is encouraged to imagine the text of the page in between the footnote and the Roman numeral. 3. This footnote is part of the layout in the English translation of La caverna de la ideas not of this article.



Yvonne Lindqvist

fictional translator literally occupying and invading the space of the real-life English, Swedish and French translators. On the one hand, there is the framework plot where the translator is the main character. On the other hand, a crime plot itself is unfolding in the manuscript on which the translator is working. The fictional translator mediates not only the translation to the reader, but also his thoughts about his translation decisions and about his own life in general.€He conducts a dialogue with himself, with the former savant translator, with a friend whom he frequently meets and also with the events in the manuscript. Eventually, the levels of the narrative become intermixed. Strange and inexplicable events happen in the surroundings of the translator. It then begins to dawn on him that the manuscript contains a secret message – a message in which someone else is also showing great interest – since the translator is kidnapped and locked up in a dark cell. The condition for his release is that he reveals the secret message hidden in the ancient manuscript. However, there is also the crime plot enclosed in the translator’s manuscript. A murder has been committed. The place is Athens. A young pupil at Plato’s Academy has been found dead. Initially it had been assumed that a pack of wolves had been responsible for his death, but the famous “Decipherer of Enigmas,” Heracles Pontor, was sure that the young man had been murdered.4 In order to avoid a scandal at the Academy, the murderer had to be found immediately. The “Decipherer” was thus contracted to solve the mystery.

5. Comparing the translations On comparison, the three translations of La caverna de las ideas may be assessed as overall adequate translations (Toury 1995:€53–69). No substantial cuts or omissions have been performed in the translations. It would have been surprising had this not been so, since the novel is a high-prestige crime novel, which has received several national and international literary awards. This double consecration in the global literary space (Casanova 2002; 2005:€ 5), might in fact have been a prerequisite for the novel to have been translated into Swedish. The British award had been of decisive importance in the decision of the Swedish literary agents to translate the novel into Swedish, since the British literary market forms the most important “stock” for translations into Swedish.5 The matricial norms of high prestige texts in general, whatever their genre, are usually less manipulated by translators than those of low prestige texts. If translators work within an open literary system, this kind of manipulation is rare. In a system 4. The choice of the name Heracles Pontor by the author is a clear inter-textual irony, since the name is reminiscent of another famous decipherer of enigmas – Hercule Poirot. 5. La caverna de las ideas had also been short listed, but not awarded, a very prestigious prize in Spain before being translated to English and French.



Manipulating the matricial norms 

which is more closed in this respect, it is probably more common (Even-Zohar 1990:€ 44–50; Lindqvist 2002:€ 30Â�-38). Although somewhat simplifying the issue, I intend to show that the English translation, performed by Sonia Soto, is relatively the most manipulated translation when compared to the Swedish and French translations. This is – I will argue – due to the position of translated literature within each literary system and to the prestige accorded to translations and translators within the respective literary systems.

5.1

The translation of the title of the novel

To begin the comparison of the translations, it can be observed that there are diverse patterns of strategy for translating the title of the novel, where the English translation is very different from the other translations.6 In the case of the English translation in the UK, the title has been changed (cf.€Table€2). The change provides the British reader with a clear genre declaration. However, recalling the four traits of post-modern narration mentioned preÂ�viously, the change of the title in the English translation deprives the target reader of the dialoguicity and double coding in the original title evoking Plato’s allegory of the Cave, which is so central to the construction of the novel. The translators to Swedish and French have, on the contrary, performed a literal translation of the original title.

5.2

The name of the translator

One clue to the prestige of the translator within a literary system is the placing of the name of the particular translator in the translated novel. Comparing the three translations in this respect, the pattern is not as easy to observe as is the case with the translation of the titles. With regard to the placement of the name of the translator, in the most open system, – the Swedish – and in the most closed – the British – similar strategies are found. The name of the translator is placed on the inner title page, immediately below the name of the author but in smaller characters. In the case of the French translation, however, the name of the translator is exposed on the cover of the novel, in relatively small characters below the title. Judging from the Internet site of the publishing house Actes Sud this is a general policy regarding translations also applied by Babel, which indicates the relatively high prestige of the literary translator within the French literary system. On the other hand, when comparing the treatment of the quotation from Plato, preceding or opening the narrative in the novel, only the Swedish translation provides the name of the previous translator of Plato. 6. I am aware of the fact that the translator is usually not responsible for the choice of the publishing house editor, who usually is the person deciding the title of the novel, at least in Sweden, but this choice gives a general indication of the respect for the source text.



Yvonne Lindqvist

5.3

Three examples of manipulation of matricial norms in the English translation

Turning to a consideration of the various meta-levels in the narration of the fictional translator, I wish to discuss three examples of manipulation on the matricial level of the source text in the English translation, where there is no correspondence in the translated texts stemming from the more open cultural systems, and no correspondence at all in the source text. These kinds of manipulations are the signature of the English translator. I have marked these in bold characters in the examples below. Returning to the novel, the frequent occurrence of philosophical reasoning in the story is based on the philosophy of Plato. Very briefly, according to Plato, “(t)he world that appears to our senses is in some way defective and filled with error, but there is a more real and perfect realm, populated by entities (called “forms” or “ideas”) that are eternal, changeless, and in some sense paradigmatic for the structure and character of our world.” (Stanford Encyclopedia of Philosophy 2007). The fictive translator in the novel sets out to search for this perfect realm suspecting that the source text is strongly eidetic. According to the translator’s note that closes the first chapter, eidesis is an ancient Greek literary technique, which he defines as follows: Example 1 Source text Spanish […] Consiste en repetir metáforas o palabras que, aisladas por un lector perspicaz, formen una idea o una imagen idependiente del texto original.€Arginuso de Corinto, por ejemplo, ocultó mediante eidesis una completísima descripción de una joven a la que amaba en un largo poema aparentemente dedicado a las flores del campo. Y Épafo de Macedonia… Translations English […] It consists in repeating, in any text, metaphors or words that, when identified by a perspective reader, make up an idea or image that’s independent of the original text. Arginusus of Corinth, for example, used eidesis to hide a detailed description of a young woman he loved in a long poem apparently about wild flowers. And Epaphus of Macedonia inserted his will by means of eidesis into an epic tale describing the death of the hero Patroclus. And Euphronius of – ’ Swedish […] Den innebär att man upprepar metaforer och ord som, då de uppfattas av en klarsynt läsare, formar en idé eller en bild oberoende av originalÂ�texten. Argnusos från Korinth, till exempel, dolde med eidetik en detaljerad be-



Manipulating the matricial norms 

skrivning av en ung kvinna som han älskade i en lång dikt som skenbart tillägnades ängsblommor. Och Epafos från Makedonien… French […] Elle consiste à répéter des métaphores ou des mots qui, isolés par un lecteur averti, forment une idée ou une image indépendante du texte originel. Arginuse de Corinthe, par exemple, dissimula sous une eidesis une description très détaillée d’une jeune fille qu’il aimait dans un long poème apparemment consacré aux fleurs des champs. Et Epaphe de Macédoine… Thus, the first example of manipulation of the matricial norm consists of the metatextual explication of the literary technique known as eidesis, where the British translator has added an explanatory example in her English text, which had no correspondence in the source text, nor in either the Swedish or French translations. The addition might possibly be indicative of the extensive reading of the real translator or her deep knowledge of ancient Greek literature. But in the analysis, I have regarded it as indicative of a lack of respect of the source text.7 Another interesting point is that by means of this translation strategy, the real translator’s voice becomes mingled up with the voice of the fictional translator, thereby inserting an additional dimension to the meta-fictional levels of the novel. Further, the fictional translator in example number two questions the prior erudite scholar’s neglect of the literary technique in the old manuscript. Example 2 Source text Spanish *Sorprende que Montalo, en su erudita edición del original, ni siquiera haga referencia a la fuerte eidesis que revela el texto, al menos a lo largo de todo este primer capítulo. Sin embargo, tambien es posible que desconozca tan curioso recurso literario. A modo de ejemplo para el lector curioso, y tambien por relatar con sinceridad cómo he vencido a descubrir la imagen oculta en este capítulo […] Translations English 4. I find it surprising that, in his scholarly edition of the original, Montalo should make no mention of the powerful eidesis present in the text, at least throughout the first chapter. But maybe he didn’t know about this strange literary device. It’s not unusual to find translators, even among the most erudite, 7. One might argue that this addition to the English translation of the text is relatively insignificant. After all it only concerns one sentence. But I wish to claim that the introduction of the proper name of the author and the hero in the example does significantly change the sense of the source text, even though the additions might well be judiciously chosen.



Yvonne Lindqvist

who are not familiar with a literary technique which may, in any case, have been used by only a handful of Greek writers – in some ways the most celebrated ones – and whose main feature is precisely that it is only noticed by those who know about it. By way of example for the curious reader, and also to be honest about how I came to discover the image hidden in this chapter […] Swedish *Det är förvånande att Montalo i sin lärda utgåva av originalet inte ens nämner den starka eidetik som finns i texten, i alla fall i detta första kapitel. Men det är också möjligt att han inte känner till ett så egendomligt litterärt grepp.€Som exempel för den nyfikne läsaren, och för att uppriktigt beskriva hur jag kom att upptäcka den dolda bilden i det här kapitlet […] French *Il est surprenant que, dans son édition érudite de l’original, Montalo ne fasse même pas référence à la forte eidesis que révèle le texte, du moins tout au long de ce premier chapitre. Mais il est également possible qu’il ignore l’existence d’un procédé littéraire aussi étonnant. A titre d’exemple pour l’édification du lecteur curieux et aussi pour expliquer avec sincérité comment j’en suis venu à découvrir l’image cachée dans ce chapitre […] The second example of manipulation of the matricial norm in the English translation is more textually extensive than the first. In this instance, the translator has inserted four additional lines into the source text. Again she has become lost in a digressive explanation whereby no respect is shown toward the matricial norms of the source text. And she includes herself amongst the initiated group of translators, firmly fighting off the fictional translator to reclaim her natural space in the footnote at the bottom of the page. The third and last example is excerpted from a passage in the novel, where the fictional translator begins to mix up the different meta-levels. The fictional translator expresses his surprise that he is talking to a character in the manuscript he is working on. Example 3 Source text Spanish *Debo haberme vuelto loco.¡He estado dialogando con un personaje! De repente me pareció que se dirigía a mí, y le contesté con mis notas. Quizá todo sea achacable al tiempo que llevo encerradoen esta celda, sin hablar con nadie. […] A Crántor no le preocupa ser creíble: se complace, incluso, en revelar el artificio verbal que lo rodea, como cuando hizo hincapié en las palabras eidéticas. (N. del T.)



Manipulating the matricial norms 

Translations English 61. I must be insane€– I’ve been talking to a character in a book! I suddenly felt he was addressing me, and I answered him in my notes. It must be because I’ve been locked in this cell all the time, with no one to talk to. […] And he doesn’t care whether he’s believable or not; he even enjoys drawing attention to the verbal artifice that surrounds him, as when he stresses the eidetic words. The unexpected move to the second person (‘he smiled at you’) is a clever way of drawing in the reader. It’s drawn me in anyway. (T.’ s N.) Swedish * Jag måste ha blivit galen. Jag har fört samtal med en person i en bok! Plötsligt verkade det som om han vände sig till mig, och jag svarade med mina fotnoter. Kanske allt kan skyllas på att jag varit instängd i den här cellen så länge, utan att tala med någon. […] Krantor bryr sig inte om ifall han är trovärdig: han tycker till och med om att avslöja den verbala konstfärdigheten som finns runt honom, som när han lade tonvikt på de eidetiska orden. Ö.a. French * Je dois être devenu fou. J’ai dialogué avec un personnage! Soudain il m’a semblé qu’il s’adressait à moi, et je lui ai répondu dans mes notes. Tout est peut-être imputable au temps depuis lequel je suis enfermé dans cette cellule, sans parler à personne […] Crantor ne se soucie pas d’être crédible: il se plaît même à révéler l’artifice verbal qui l’entoure, comme lorsqu’il a insisté opiniâtrement sur les mots eidétiques. (N.d.T.) The additional explanation inserted at the end of the translator’s note in the English translation emphasises the voice of the real translator, explaining a literary device in the manuscript on which the fictional translator is working. It is clear here that the translator has not respected the meaning of the source text when she has manipulated her translation. She seems to work on her own agenda retrieving footnote space and explaining the source text to the target readers.

6. Conclusions The results of the comparison of the limited selection of examples in this paper suggest that this line of research might shed light on translation behaviour since they do not contradict the initial hypothesis of the study. Thus the translator working in the relatively more closed cultural system regarding translation policy such as the British has a pronounced tendency to insert information into the target text, which has no correspondence in the source text, nor in the French and Swedish translations.



Yvonne Lindqvist

However, the French and Swedish translators have behaved in a surprisingly similar manner in their translations, performing more adequate translations, allowing to a greater extent their readers to perceive the possibilities of post-modern narration. But on at least one meta-fictional level the Swedish and the French transÂ�lations do differ. On the page preceding the narration in the French translation an additional editorial note has been inserted: Toutes les notes du traducteur sont de l’auteur. [The translator’s notes are written by the author]8

This note in the French translation reduces the possibilities for the French readers to fully enjoy the play of the different meta-levels in the novel. From this comparison, then, the Swedish translation is on the whole to be regarded as the most adequate translation. In my view, working in accordance with such an adequate translation strategy is a prerequisite for being accepted as a translator in the Swedish field of high prestige crime fiction. It is an integral part of the translation capital of this field (Lindqvist 2006:€72–74). To conclude, these kinds of studies combining observations of translations on both the cultural macro and micro level might in fact provide a deeper insight into underÂ�standing the translator habitus (cf.€Bourdieu 1992b: 51; Simeoni 1998; Sela-Sheffy 2005). That should be of concern to Translation Studies, since the mapping of the translator habitus presents a fundamental clue to translation behaviour. However, further research examining other translators and translations is needed to confirm the hypotheses of the study.

References Bourdieu. P. 1986. Kultursociologiska texter. Stockholm: Salamander. Bourdieu. P. 1992a. Les règles de l’art. Genèse et structure du champ littéraire. Paris: Seuil. Bourdieu. P. 1992b. Texter om de intellektuella. Stockholm: Symposion. Broberg. J. 2005. “Allt går igen – nästan. En lägesrapport med tillbakablickar.” Jury. 1:5–7. Brunner. E. 2007. “Nu lägger de stora författarna av.” Dagens nyheter. Kultur.15€juli. Casanova. P. 1999. La république mondiale des lettres. Paris. Éditions du Seuil. Casanova. P. 2002.€“Consécration et accumulation de capitale littéraire. La traduction comme échange inégal.” Actes de la recherche en sciences sociales. 144: 7–20. Casanova. P. 2005. “Literature as a world.” New Left Review. 31: 1–10. Eco, U. 2004. “Intertextuell ironi och läsnivåer.” Tankar om litteratur. Översättning Barbro Andersson. (Sulla letteratura, 2002 R.C.S Libri S.p.A. Milan Bompiani). Stockholm. Brombergs Förlag. 185–205. Even-Zohar. I. 1990. “Polysystem studies.” Poetics Today. 11:1. Tel Aviv. The Porter Institute for Poetics and Semiotics. 8. My translation.



Manipulating the matricial norms  Gouanvic. J.-M. 1997. “Pour une sociologie de la traduction. Le cas de la littérature américaine traduite en France après la Seconde Guerre mondiale (1945–1960).” In Translation as Intercultural Communication, M. Snell-Hornby, Z. Jettmarová and K. Kaindl (eds), 33–45. Amsterdam/Philadelphia: John Benjamins. Gouanvic. J.-M. 1999. Sociologie de la traduction: la science-fiction américaine dans l’espace culturel français des années 1950. Arras: Artois Presses Université. Håkansson. G. 2007. “Vem kan man lita på?” Dagens nyheter. Kultur. 11€augusti. Högström. J. 2007. “Författare oroas av romanens kris.” Dagens nyheter. Kultur. 30€juli. Johansson. L. 2005. “Engelskan ohotad etta i bokvärlden.” Dagens Nyheter. Kultur. 4€augusti. Lindqvist. Y. 2002. Översättning som social praktik. Toni Morrison och HarlequinÂ�serien Passion på svenska. Acta universitatis Stockholmiensis. Stockholm Studies in Scandinavian Philology. 26. Stockholm: Almqvist & Wiksell International. Lindqvist. Y. 2006. “Consecration mechanisms – The reconstruction of the Swedish field of high prestige literary translation during the 1980s and 1990s.” In Übersetzen – Translating – Traduire: Towards a ”Social Turn”? M. Wolf (ed.), 65–75. LIT: Münster-Hamburg-BerlinWien-London. 65–75. Nationalbibliografin.2007.http://www.kb.se/soka/bibliografier/nationalbibliografin/nationalbiografin/statistik/ (accessed 18€September 2007). Standford Encyclopedia of Philosophy. 2007.http://plato.stanford.edu/entries/plato-metaphysics/#9, [accessed 12€December 2007). Toury. G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins. Toury. G. 1998. “A handful of paragraphs on ‘Translation’ and ‘Norms’.” Current Issues in Language & Society. 5. 1 & 2. 10–32. Sela-Sheffy. R. 2005. “How to be a (recognized) translator: Rethinking habitus, norms, and the field of Translation.” Target 17. 1–26. Simeoni. D. 1998. “The pivotal status of the translator’s habitus.” Target. 10:1. 1–39. Venuti. L. 1995. The Translator’s Invisibility. A History of Translation. London and New York: Routledge. Widerberg. Bl. 2004. “En riktig deckarboom.” Jury. nr.€3. 2004. 4. Wopenka. J. 2004. “Den svenska deckarboomen – i den utsträckning den finns.” Jury. 4. 22–24.

Knowledge in Translation Studies and translation practice Intellectual capital in modern society Hanna Risku, Angela Dickinson and Richard Pircher1 Danube University Krems, Austria

This paper seeks to show that translation practice and Translation Studies matter on both an organisational and a social level by demonstrating that the knowledge they generate forms an important part of the intellectual capital in the knowledge society and a key factor in value creation in organisations. It looks at methods available for managing knowledge and seeks to identify those instruments that can play a role in a translation setting. By virtue of their suitability to the forms of knowledge used in translation, we hypothesize that the appropriate use of such methods and instruments can facilitate the transformation of translation knowledge into intellectual capital, add value to products and services and thus make a valuable contribution to the sustainability of a knowledge society. The challenge now facing translation practice is to ensure that translators assume their role in knowledge management endeavours, implement appropriate methods for generating and sharing knowledge and so turn their knowledge into true intellectual capital. Keywords: intercultural communication, knowledge management, intellectual capital, knowledge work, personalisation, codification, translation, Translation Studies

1. Introduction Knowledge management (KM) has gained increasing importance in the business world over the last decade, although the concept of knowledge work is in fact far older. It dates back to the 1950s when the term knowledge worker was coined by Peter Drucker, in his 1959 book Landmarks of Tomorrow. The importance of intellectual capital and rise in significance of KM has again brought knowledge work to the 1. Hanna Risku and Angela Dickinson: Danube University Krems, Austria; Richard Pircher: University of Applied Sciences bfi Vienna, Austria.



Hanna Risku, Angela Dickinson and Richard Pircher

forefront of management research, where it is used to refer to people whose work primarily involves the development or use of knowledge. Numerous examples are given in literature of the types of professions that can constitute knowledge workers, including, but not limited to, researchers, engineers, product developers, analysts and teachers. Although it obviously merits being regarded as a knowledge profession, the field of professional translation is rarely included in general discussions on knowledge work. For organisational KM – the systematic support of knowledge creation and sharing in an organisation rather than the managing of personal knowledge by individuals – to be applied in translation departments and companies, it needs to be seen as an organisational driver or integrated approach to achieving organisational goals that places particular focus on knowledge as the new factor of production (Sammer et al.€2003). This means asking questions like: How can we best share our existing knowledge with those who need it? What new knowledge will we need in the future? What is the best way to identify or generate this new knowledge? Organisational KM centres on the recognition that knowledge forms a key corporate asset, and organisations have good reason to try to manage their knowledge or intellectual capital.€Aside from the relevance of organisational KM for in-house translation departments or translation agencies, this paper aims to show that translation practice and Translation Studies matter on an organisational and a larger social level by demonstrating that: 1. The knowledge generated in Translation Studies forms an important part of the intellectual capital in the knowledge society, and 2. The knowledge involved and embedded in professional translation forms a key factor in value creation in organisations. It also seeks to introduce some of the KM methods and instruments that can play a role in translation. By virtue of their suitability to the different forms of knowledge used in translation, we hypothesize that the appropriate use of such methods and instruments can facilitate the transformation of translation knowledge into intellectual capital and thus make a valuable contribution to the sustainability of a knowledge society.

2. Knowledge and intellectual capital Intellectual capital is knowledge that can be exploited for some useful purpose and includes a company’s collective knowledge, experience and competences. It can also include artefacts and intangible resources, such as the capabilities and interaction of employees, formal and informal communities, customers, partners and other stakeholders. Edvinsson and Malone (1997:€52) define intellectual capital as knowledge that is of value to an organisation and made up of human capital, structural capital and customer capital, while others see it as a combination of human, structural and



Knowledge in Translation Studies and translation practice 

relational capital (see Reinhardt et al.€2001:€795–796 for a discussion of the different concepts of intellectual capital). We also prefer the term relational rather than customer capital, because it takes not only customers, but also other stakeholders into account. In translation practice, this might include long-term relationships with freelancers and subject matter experts as well as stable roles in professional networks (such as LinkedIn or EST) since these also form part of a translator’s, translation department’s or translation agency’s intellectual capital (see also Koskinen and Abdallah (2007), who refer to this category as social capital). Human capital is found in the minds of individuals (knowledge, competencies, experience, know-how, etc.). In translation, this includes aspects like language skills and the ability to select a specific translation strategy. Structural capital is what remains in the company after the employees go home (processes, information systems, databases, patents, etc.). Examples in translation include translation memories, glossaries, translations in other languages or checklists. Finally, relational capital is the value of an organisation’s relationships with its customers and stakeholders. This involves more than simply business relationships with financial dependencies, and is very dependent on the most valuable relational capital of all – trust (Risku and Pircher 2006; Risku 2007; Koskinen and Abdallah (2007)). There are numerous definitions of what is actually meant by knowledge, but in view of recent studies and developments in Cognitive Science presented in Risku (2000, 2002 and 2004), we strongly support the Situated Cognition approach in which knowledge is seen as the dynamic, interactive and socially generated activity patterns and processes of an individual or, more generally, a cognitive system. Consequently, knowledge is not seen as the stable, linguistically transparent mental rules and concepts assumed in prior scientific approaches (such as the symbol manipulation, problem-solving and information processing hypotheses). Although the KM field remains dominated by applied business and management goals, a constant topic of discussion is also the definition of knowledge itself (as the main object of study and intervention). Probst et al.€(1999:€46) describe it as “the entirety of experience and abilities used by individuals to solve problems. This includes theoretical insight as well as practical day-to-day rules and instructions for action”. This definition is only acceptable in a Situated Cognition approach if we add that (1) knowledge serves not just to solve specific, consciously perceived or clearly definable problems, but also to provide orientation in our mental and physical environments, and (2) rules and instructions are implied in cognitive systems more as connections and patterns than explicit, fixed commands. Roehl (2002:€19–21) describes knowledge as the product of contexts, i.e. the links, correlations, ties and relations that allow experience to develop and become effective, and maintains that it is this contextuality which characterises knowledge. Knowledge is the background we use to interpret and construct reality and represents the complex system of more or less vague assumptions, images and stories of ourselves and the world we carry around with us in our heads.



Hanna Risku, Angela Dickinson and Richard Pircher

Knowledge also takes different forms, with the most common differentiation made between explicit and tacit knowledge. Since they have already been described and used in a Translation Studies context before, our description of these two forms of knowledge has been kept deliberately brief. Instead, we seek to examine how they can be transformed into intellectual capital, i.e. how their development and management can be supported and optimised in translation departments, companies and society as a whole. Explicit knowledge is that which can be codified, stored, transmitted and shared (although it is not stored in the brain in the form of codes, symbols or rules, but rather as activation patterns and connections, with the transformation remaining to be carried out). It is formal and systematic and can be expressed easily in words and numbers. It is the knowledge of rationality and is easily communicated in the form of hard data, formulae, codified procedures or principles (Nonaka and Takeuchi 1995:€8). Examples in translation include language and grammar rules, glossary and resource links, translation memories, reference material or documented information on the subject matter. Polanyi (1966:€4) describes tacit knowledge as a combination of bodily experience and practice (“We can know more than we can tell.”). It is highly personal, hard to formalize and share with others and is not easily visible and expressible (Nonaka and Takeuchi 1995:€8). Tacit knowledge plays an important role in translation since translators grasp the source, target and client situation and work with the meanings behind the words to also address the target audience’s tacit knowledge. Examples in translation include feeling for language and intuitive decisions to translate a sentence in a particular way (ignoring dictionary definitions) based on empathy for the context or knowledge of the target culture. The important realisation for the application of our analysis is that explicit and tacit knowledge need totally different support to grow and flourish. The following provides a brief look at the development of applied KM to determine the kinds of methods and instruments available for managing explicit and tacit knowledge.

3. Approaches to Knowledge Management As the KM movement developed, two different strategies emerged, namely codification and personalisation (Hansen et al.€1999), with the former best suited to managing explicit knowledge and the latter also able to tackle tacit knowledge. The codification approach centres on the managing of information as the explicit data relevant for a system or situation (Sveiby 1996). It regards knowledge as identifiable objects that can be stored and managed in information systems and deals primarily with explicit knowledge, since this is generally more readily accessible and more easily codifiable. According to Hansen et al.€ (1999:€ 108), codification strategies are based on “reuse economics”, i.e. focus on providing high-quality, reliable information systems by developing ways of codifying, storing and reusing existing knowledge via a “people-to-documents” approach. In other words, knowledge is “extracted from the



Knowledge in Translation Studies and translation practice 

person who developed it, made independent of that person, and reused for various purposes” (Hansen et al.€1999:€108). Here we find a clear parallel to at least three lines of translation research: (1) the comparative study of linguistic and cultural systems, which aims to document the explicit regularities and conventions of language use (e.g.€comparative linguistics, multilingual terminology, semantic networks), (2) the applied development of translation tools that depend on the reuse of previously translated linguistic material (e.g.€translation memories), and (3) the study of corpora, the linguistic material gathered in databases. Common to these lines of research is their concentration on the study of products not processes and quests to find and document systematic regularities that can – to some extent – be decontextualized and depersonalised. Since the documentation of explicit translation knowledge generates valuable information resources for translation, we can conclude that the systematic and strategic application of codification-based KM methods in translation and Translation Studies generates intellectual capital.€However, as the history of KM shows, codification can only gain practical value by also adopting the second KM strategy, personalisation. The personalisation approach looks at human issues, i.e. managing and mobilizing people to develop, share and use knowledge. Knowledge is regarded as a process, a complex set of dynamic skills, know-how and experience that is constantly changing. According to Hansen et al.€(1999:€108), personalisation strategies aim to provide creative, analytically rigorous advice by channelling individual expertise and focus on: […] dialogue between individuals, not knowledge objects in a database. Knowledge that has not been codified – and probably couldn’t be – is transferred in brainstorming sessions and one-to-one conversations.

In Translation Studies, this is the realm of approaches emphasizing and studying individual translator competence, analysing unique real-world, real-time translation in action, exploring the possibilities for translators to assume an expert role, describing the situated, cognitive processes of translation and developing experience- and projectbased translation teaching methods. They are unified not only by their object – translators (not languages or translations) – but also by their emphasis on the importance of the individual translator in a specific context of translation. Recent KM trends show that the importance of the human and cultural aspects (personalisation) now seems to outweigh that of an IT-based knowledge strategy (codification). The problem with the codification approach is that databases and documentation systems often end up as outdated, often much-hated data cemeteries: Organisations introduce an information technology tool in the vain hope it will solve the task of managing their knowledge. However, in the stress of their daily work, people are rarely motivated to feed such systems with knowledge acquired, while users quickly realize that the imported data is rarely of direct relevance to a current, specific problem and context. For example, Siemens Austria chose not to try to make its (employees’) knowledge and experience available in an information system, since the dynamics and



Hanna Risku, Angela Dickinson and Richard Pircher

growth of the domain knowledge would overload any database, and the specific experience of its employees did not seem to be transferable into a static documentation system (see also Füricht-Fiegl 2006). A good example in the translation field is translation memories where quality issues generally prevent acceptance of even a 100% match without at least a quick check. Instead, Siemens Austria set up communication platforms to create knowledge networks in which staff with relevant experience can be contacted to discuss specific issues or problems. In other words, their solution was not to try to document knowledge, but instead to document the contact data of their individual experts and enable them to easily locate each other and interact.

4. Parallels between KM, translation practice and Translation Studies Translators are an excellent example of knowledge experts. They make extensive use of their knowledge and skills in their daily work, are life-long learners and perennial language students, proficient at managing their personal knowledge. They allocate significant portions of their time to knowledge-oriented activities such as research, managing information, maintaining glossaries, networking, etc. (Dickinson 2002:€10). In the process, they make extensive use of information and communication technologies, such as the internet, computer-aided translation and project management tools, translation memories and collaboration possibilities such as online communities and newsgroups. All these activities serve to generate intellectual capital as a resource for the future. Translators are natural communicators, and as a survey of over 300 translators by Risku and Dickinson (2009) shows, the embracing of the online translation communities that have emerged in recent years indicates they – at least subconsciously – recognise the benefits of knowledge exchange in such networks. Despite trends towards teamwork and collaboration, translation remains a solitary profession, and many members regard these communities as virtual workplaces where they can meet and communicate with colleagues. Furthermore, continuing developments in communication and translation technologies enable translators to improve their knowledge activities and increase productivity by providing them with easier access to each other and enabling them to work together, share information, learn from each other and expand their collaboration and cooperation possibilities. By doing so, translators improve their own skills and knowledge, adding value to the services they provide.

5. Knowledge in translation practice The new professional profile of the translator emerging to meet the changing requirements in today’s business world is that of a highly skilled problem solver, working in teams, applying their knowledge to dealing with complex intercultural situations and



Knowledge in Translation Studies and translation practice 

thereby generating intellectual capital.€ Meeting with experts to gather information, coordinating teams, researching subject matter, creating, managing and distributing documents and glossaries, working with translation management systems, networking with colleagues around the globe and providing consulting services to customers are now very much part of the day-to-day work of many translators (Risku 2007). These trends are confirmed in the results of a survey carried out by the Canadian Translation Industry Sectoral Committee in 1999 in which respondents identified the following skills, knowledge and personal characteristics as necessary for a career in professional translation: Professional skills: excellent understanding of the source language; solid skills in writing and syntax in the target language; mastery in at least one field of specialization and the ability to work comfortably in some others; mastery of computer tools. Personal abilities: well-rounded cultural and intellectual curiosity; ability to work within a team; adaptability, availability, motivation, good judgment, good communication skills, efficiency and flexibility; concern for the client. In view of the professionalisation of translation practice, the skills described in the above survey could be further expanded to include aspects such as sufficient knowledge of the history of translation as a profession, the different possible roles of the translator, as well as knowledge of the use of different translation strategies in different situations. As intercultural communication experts, translators require extensive knowledge – both tacit and explicit – to carry out their roles and continually refer to this knowledge throughout the translation process. For the purposes of our analysis, and to illustrate the relevance of KM for translators, we classified the types of knowledge used and required by translators into five general categories based on a study of relevant literature (cf.€ Chesterman 2005; Dickinson 2002; Risku 2003; Samuelsson-Brown 1998; Schäffner 2000; Sofer 1999): 1. Language, linguistic and translation knowledge; 2. Country and cultural knowledge; 3. General and subject matter knowledge; 4. Client and business knowledge; 5. Information technology and computer skills. We then looked at each category individually to investigate the extent to which the different types of knowledge required in the translation process are codifiable or not and identify appropriate KM tools and instruments (see Table€1). At this stage, our categorisation is based solely on a combination of literature review and experience in the fields of KM and Translation Studies. A subsequent stage will seek empirical validation through further field studies. Codifiable aspects of language, linguistic and translation knowledge include grammar rules, terminology, register guidelines, translation methods and strategies. Appropriate KM tools include glossaries, translation memories, databases, textbooks,

 Hanna Risku, Angela Dickinson and Richard Pircher

style guides, etc. Non-codifiable aspects include tacit understanding of context and variations in meaning or the instinctive decision to translate a passage in a particular way. A different set of tools and instruments is needed to develop and manage this kind of knowledge. Here, KM possibilities include brainstorming with clients, subject matter experts or colleagues, membership of translation-related communities (online and offline) and associations, participation in expert groups or communities of practice, networking, conferences, etc. As far as country and cultural knowledge is concerned, codifiable aspects include regulatory guidelines and documentable linguistic differences (e.g.€the differences between British and American English). Knowledge management tools would include databases, country information websites, literature, the local press and media. Noncodifiable country and cultural knowledge is again more tacit and includes understanding of the source and target cultures, attitudes and context. This kind of knowledge can be developed using conventional tools to encourage interpersonal exchange such as local translator associations where national and non-national translators living in one country come together. Embassies and Chambers of Commerce are also a good source of country information. But there are also a number of less familiar tools that can be of use such as mentoring (an excellent way of finding your feet and developing local and professional knowledge), storytelling (learning from friends, neighbours and colleagues) or learning journeys (visiting or living in a country to immerse yourself in the culture). Codifiable aspects of general and subject matter knowledge include reference material, journals and industry guidelines, which can be stored and found in databases, knowledge portals, specialized journals, etc. Examples of non-codifiable knowledge would include education in the wider sense, general knowledge, world experience, curiosity and a constant willingness to learn. This can be nurtured and developed using instruments such as brainstorming, continuing learning, participation in conferences, seminars and courses as well as through subject-related blogs, wikis, etc. When it comes to client and business knowledge, there are a number of aspects, such as specific terminologies, glossaries, reference material, in-house style guides, contact information etc., which can be codified in various tools like terminology management systems, translation memories, customer relationship management systems or knowledge portals. However, as far as the non-codifiable aspects are concerned (e.g.€ interaction strategies, long-term client-translator relationships, trust), we must again turn to “less conventional” tools such as networking, collaboration tools, yellow pages, coaching and regular contact or meetings. Finally, codifiable IT and computer knowledge can be stored in user manuals, handbooks, online help, etc. while translators can develop their non-codifiable knowledge (such as an ability to solve complex problems) through continuing training, participation in online forums and mailing lists or taking part in conferences, seminars and training courses. Table€ 1 summarises the codifiability of the different types of knowledge required in the translation process and lists the appropriate KM methods and tools identified.



Knowledge in Translation Studies and translation practice 

Table 1.╇ Codifiability of Relevant Knowledge and Knowledge Management Instruments Type of Knowledge

Codifiable Aspects

KM Instruments for Codifiable Aspects

Non-Codifiable Aspects

KM Instruments for Non-Codifiable Aspects

Language, linguistic, text skills, translation competence

Grammar, terminology, professional and regional conventions, register and writing conventions, translation methodologies and strategies, project management

Glossaries, translation memories, databases, style guides, terminology guidelines, newsletters, handbooks, ontologies

Tacit understanding of context, variations in meaning, experience of source/ target languages, understanding of the purpose of the translation, instinctive, cultural or creative strategies

Mailing lists, online communities, translator associations, translation courses and collaboration tools, Communities of Practice, conferences, seminars

Country and cultural knowledge

Economic, legal, and regulatory requirements, conventional linguistic differences

Databases, websites, literature, media

Tacit knowledge of the source and target cultures, underlying knowledge of complex contexts (attitude, culture, history, etc.)

Translator associations and communities, Chambers of Commerce, embassies, mentoring, storytelling, learning journeys

General and subject matter knowledge

Reference material, journals, industry guidelines

Databases, publications, knowledge portals, expert systems, knowledge and topic maps

“Education”, general knowledge, constant willingness to learn, world experience, curiosity

Yellow pages, blogs. wikis, brainstorming, conferences, seminars

Client and business knowledge

Terminology, glossaries, contact, reference material, stylistic guidelines, industry information

CRM, PM tools, style guides, terminologies, translation memories, knowledge portals

Interaction strategies, long-term client-translator relationships, trust, marketing skills

Collaboration tools, yellow pages, coaching, networking, brainstorming

Information technology and computer skills

Communication, desktop publishing and translation tools, formatting guidelines

User manuals, handbooks

Ability to solve complex problems

Courses, online forums, communities, mailing lists, conferences, seminars



Hanna Risku, Angela Dickinson and Richard Pircher

6. Relevance of Translation Studies for business and society As shown above, translation practice and Translation Studies clearly generate intellectual capital for an organisation and subsequently apply the intellectual capital to add value to products and services. But what relevance does Translation Studies have for society in general? Why would Translation Studies matter? Does Translation Studies generate intellectual capital for society and if so, what kind? Defining the type of knowledge generated in Translation Studies classifies the whole discipline as a representative of a certain scientific affiliation: is it a natural science, a technical and engineering science, or does it belong to the humanities or cultural and social sciences? (For a current discussion of scientific disciplines see Liessmann 2006.) If Translation Studies is considered to be about determining the underlying rules of nature, it could be seen as a natural science aimed at uncovering the universal principles of intercultural communication and language use. Measuring brain activity patterns during translation is a good example of such translation research. Combined with a general objectivistic, positivist view of science, the natural scientific view of Translation Studies would be a showcase of the picture and quality of science still dominant in the debate on the value of fundamental science: the knowledge it has to offer is fundamental in nature and thus has to be seen to be of value in its own right. Any requests or demands on the part of society to provide applied knowledge for use in specific current problems would be irrelevant for the science as such. The use of the knowledge acquired would lie outside the scope or responsibility of the discipline and its representatives. If Translation Studies seeks and (is thus able) to offer knowledge of how to bridge linguistic and cultural communication barriers, it can be classified as a technical and engineering science. Here, it is able to help solve a core problem of modern society, and thus show relevance as a key competence and an invaluable source of intellectual capital. If Translation Studies asks the questions of “why” and “where” the people, processes and products involved in translation come from in the course of human culture and society, and “where they are going to”, it should be classified as a humanities or cultural and social studies discipline. Translation Studies critically questions the underlying sense and meanings behind the choices made and developments shown. For society, Translation Studies offers frameworks of understanding and orientation and, at the same time, plays an emancipating role: laying open the possibility of different choices, decisions and interpretations and their links to different consequences. As Maria Tymoczko argued in her plenary talk at the 2007 EST conference in Ljubljana, the diversity of the scientific approaches is a value as such: diversity contributes to the sustainability of the discipline. Translation Studies as a scientific discipline will ideally illustrate the strengths of a good combination of a critical, cultural and social studies perspective and an engineering approach: it participates in the discussion on value, contributes to sensemaking, enables us to understand the consequences of different choices and offers practical knowledge of how to deal professionally with translation issues or bridge the culture gaps. Indeed, this article is an attempt to do exactly that.



Knowledge in Translation Studies and translation practice 

7. Conclusion Although translators are often primarily seen as language professionals, their knowledge and skills extend far beyond their language pairs. Translation is an analyticalsynthetical, research-intensive process that requires extensive background knowledge (both tacit and explicit) of the source and target languages and cultures, as well as the subject matter of the text, purpose of the translation, requirements of the target audience, potential roles of the translator and translation methods and strategies suitable for different cultures and communication situations. In times of global cooperation and conflict, intercultural communication helps smooth the way for dialogue and successful value creation. To overcome cultural and communication barriers, societies today need access to professionals with the appropriate knowledge and competencies. As intercultural communication experts and knowledge professionals, translators are in an excellent position to make a unique contribution to the value creation process. The challenges now facing translation practice in this regard are to ensure that they assume their rightful role as an integral part of KM endeavours, implement appropriate KM methods for generating and sharing the corresponding types of knowledge and, by doing so, turn their knowledge into true intellectual capital. With this paper, we have demonstrated that translation practice and Translation Studies both generate intellectual capital and leverage this by adding value not only to an organisation’s products and services but also to society and business as a whole. Consequently, Translation Studies does matter! To develop this intellectual capital systematically in the future, a detailed study of translation departments and agencies that actually use KM in some form will be required to determine what really works for our industry. As Brian Mossop (2000) already requested back in 1998, we need large-scale observation of workplace procedures. Here, Translation Studies and teaching can be of major assistance by showing the tools that are available beyond translation memories.

References Canadian Translation Industry Sectoral Committee 1999. Survey of the Canadian Translation Industry, September 1999. [Online.] www.uottawa.ca/associations/csict/princi-e.htm (24.02.2002). Chesterman, A. 2005. “The memetics of knowledge.” In Knowledge Systems and Translation, H. V. Dam, J. Engberg and H. Gerzymisch-Arbogast (eds), 17–30. Berlin: Mouton de Gruyter. Dickinson, A. 2002. Translating in Cyberspace. Virtual Knowledge Communities for Freelance Translators. Master Thesis: Danube University Krems. Drucker, P. F. 1957. Landmarks of Tomorrow. New York: Harper. Edvinsson, L. and Malone, M.S. 1997. Intellectual Capital: Realizing Your Company’s True Value by Finding its Hidden Brainpower. New York: HarperCollins. Füricht-Fiegl, G. 2006. “Kommunikationskreislauf: Von der Entstehung von Bedeutung bis zur Wirkungsforschung.” Timnews 1/2006:€8.



Hanna Risku, Angela Dickinson and Richard Pircher Hansen, M. T., Nohria, N. and Tierney, T. 1999. “What’s your strategy for managing knowledge?” Harvard Business Review 77, No. 2: 106–11. Koskinen, K. and Abdallah, K. 2007. “Managing trust: Translating and the network economy.” Meta 52(4): 673–687. Liessmann, K.P. 2006. Theorie der Unbildung. Die Irrtümer der Wissensgesellschaft. Vienna: Zsolnay. Mossop, B. 2000. “The workplace procedures of professional translators.” In Translation in Context, A. Chesterman, N. Gallardo San Salvador and Y. Gambier (eds), 39–48. Amsterdam/ Philadelphia: John Benjamins. Nonaka, I. and Takeuchi, H. 1995. The Knowledge Creating Company. New York: Oxford University Press. Polanyi, M. 1966. The Tacit Dimension. New York: Doubleday. Reinhardt, R., Bornemann, M., Pawlowsky, P. and Schneider, U. 2001. “Intellectual capital and knowledge management: Perspectives on measuring knowledge.” In Handbook of Organizational Learning and Knowledge, M. Dierkes, A. Berthoinantal, J. Child and I. Nonaka (eds), 794–820. New York: Oxford University Press. Risku, H. and Dickinson, A. 2009. “Translators as Networkers; The Role of Virtual Communities.” Special Issue of Hermes: “Translation Studies: Focus on the Translator.” H. V. Dam and K. Korning Zethsen (eds), 49–70. Risku, H. 2000. “Situated translation and situated cognition: Ungleiche Schwestern.” In Translationswissenschaft. Festschrift für Mary Snell-Hornby zum 60. Geburtstag, M. Kadric, K. Kaindl and F. Pöchhacker (eds), 81–92. Tübingen: Stauffenburg. Risku, H. 2002. “Situatedness in Translation Studies.” Cognitive Systems Research 3:3: 523−533. Risku, H. 2003. “Migrating from translation to technical communication and usability.” In Translation Studies: Claims, Changes and Challenges. Selected papers from the EST Congress, Copenhagen 2001, G. Hansen, K. Malmkjær and D. Gile (eds), 181–196. Amsterdam/Philadelphia: John Benjamins. Risku, H. 2007. “The role of technology in translation management.” In Doubts and Directions in Translation Studies. Selected contributions from the EST Congress, Lisbon 2004, Y. Gambier, M. Shlesinger and R. Stolze (eds), 85–98.: Amsterdam/Philadelphia: John Benjamins. Risku, H. 2009. Translationsmanagement. Interkulturelle Fachkommunikation im InformationsÂ� zeitalter (= Translationswissenschaft 1) 2., überarbeitete Auflage. Tübingen: Narr. Risku, H. and Pircher, R. 2006. “Translatory cooperation: Roles, skills and coordination in intercultural text design.” In Übersetzen – Translating – Traduire: Towards a “Social Turn”? M. Wolf (ed.), 253–264. Münster: LIT. Roehl, H. 2002. Organisationen des Wissens. Anleitung zur Gestaltung. Stuttgart: Klett-Cotta. Sammer, M. 2003. An Illustrated Guide to Knowledge Management. Graz: Wissensmanagement Forum. Samuelsson-Brown, G. 1998. A Practical Guide for Translators. Clevedon: Multilingual Matters. Schäffner, C. 2000. “Running before walking? Designing a translation programme at undergraduate level.” In Developing Translation Competence, C. Schäffner and B. Adab (eds), 143–156. Amsterdam/Philadelphia: John Benjamins. Sofer, M. 1999. The Translator’s Handbook. Rockville: Schreiber Publishing. Sveiby, K.-E. 1996/2001. What is Knowledge Management? www.sveiby.com/Portals/0/articles/ KnowledgeManagement.html (accessed 16.10.2007)

Language issues

Is Translation Studies going Anglo-Saxon? Critical comments on the globalization of a discipline Mary Snell-Hornby

University of Vienna, Austria

The paper focusses on the dominant role of English in the academic discourse of Translation Studies and the effects this has on communication within the scientific community: the danger is increasing that the language English is not only used as a means of communication (e.g.€with publications and conferences exclusively in English), but actually forms part of the object of discussion, dealing with facets of languages contrasted to English, and using English examples, whereby non-native speakers are disadvantaged, other approaches are ignored, and “pseudo-English” discourse is increasingly tolerated. The paper discusses the subtle difficulties involved in English academic discourse and presents a number of possible solutions, concentrating on the concept of passive multilingualism and the introduction of “bridge languages” as a means of gaining access to languages other than English or the active working language of the individual scholar. Keywords: English as lingua franca, bridge languages, academic discourse

Introduction Some time ago I was approached by an editor from the department “English Language and Linguistics” of a major Anglo-American publishing firm and asked to assess a manuscript for publication in the field “Applied Linguistics”. The title and content of the manuscript covered a broad area of topics meanwhile familiar in Translation Studies, which about twenty years ago celebrated its emancipation as a discipline in its own right as against its former status as a branch of Comparative Literature on the one hand and Linguistics on the other. There are of course publishers who do meanwhile list Translation Studies separately in their catalogues and have introduced programmes which emphasize its disciplinary profile, particularly those publishing outside English-speaking countries, but over the last few years there have been some disquieting signs of uncertainty, involving not only a tendency “back to linguistics”



Mary Snell-Hornby

(cf.€Snell-Hornby 2006:€150–159), but a trend, as seen above, which positions Translation Studies solely within the range of English.

1. Globalizing Translation Studies? This is of course closely connected with the development of English as the global lingua franca, an excessively debated topic we need not elaborate here, except to say that the prognoses and reservations expressed towards the end of the 1990s have been more than confirmed and today may even seem like conservative understatements. Globalization needs and favours quick and easy communication: in certain limited situations this is in fact essential (as in international flight control) and for some purposes (as in standard everyday information, Internet chat, small talk etc.) basic elements of the English language code may seem an easy solution, recognizable in what has become known as Global English. The situation is however radically different with sophisticated academic discourse dealing with complex acts of communication across potentially all languages and cultures as in Translation Studies. For our area of research and practice concerning all facets of linguistic and cultural transfer, the dominance of English involves specific problems: the danger is increasing that the language English, beyond being a means of communication, is itself becoming part of the object of discussion – with conferences and publications largely, sometimes even exclusively, in English and run by Departments of English across the world, concerning facets of languages contrasted to English and using English examples. A number of journals which describe themselves as international only accept contributions in English: somewhere in the guidelines you may find a clause saying that if English is not the author’s mother tongue the text should be checked by a native speaker, whether of the British or American variety. In this day and age that sounds rather naive: such native speakers are now vastly outnumbered by those speaking English in one of its innumerable varieties or as a second or official language (English has been famously described by Henry Widdowson as “world property”, cf.€Widdowson 1994), and given the degree of permissive tolerance shown towards the way English is used online, in print and on air, one wonders whether native speakers in this role are relevant at all. As scholars they do however have a definite and unfair advantage over the speakers of other languages, and this is seen most clearly in conferences with English as the only conference language (also those where other languages are explicitly tolerated as “deviations”). This has at least two negative consequences: firstly, an increasing number of contributions being delivered by nonEnglish-speaking scholars in an English which is exceedingly difficult to follow, and secondly an increasing number of colleagues without English as an active working language who hesitate to take part in the debate at all. This inevitably means that



Is Translation Studies going Anglo-Saxon 

Anglo-American (or more specifically US-UK1) discourse has a distinct advantage over that in other languages and cultures, which quite frequently remains unnoticed in the English-speaking scientific community, thus defeating the very purpose of Translation Studies as international and cross-cultural communication. Within a European context this means that a multilingual colleague, let’s say with Bosnian as mother tongue, German as language of education at school, with Russian as second working language, plus basic knowledge in French and some knowledge of English (as a fifth language) is virtually excluded from the discussion, although he/she has a greater language proficiency and hence more potential for Translation Studies than, let’s say, a foreign student of English or a native speaker of English with proficiency in only one foreign language.

2. The limits of my world There is a famous quotation by the Austrian philosopher Ludwig Wittgenstein (who lived in England for many years and even published in English, though he is not as well-known there as he would deserve to be), which runs as follows: “Die Grenzen meiner Sprache bedeuten die Grenzen meiner Welt”2: The limits of my language are the limits of my world. From today’s perspective one might add the ironic comment that even our globalized world is not English, though at a superficial glance and in some limited areas it might seem so. And here we need to make the distinction suggested by Kaisa Koskinen (2004) between the discourse systems of international institutions on the one hand (as in the European Union and United Nations, also to a somewhat lesser extent in commerce and technology), where a global language or procedural languages may be necessary and make financial sense, and translation as cross-cultural communication on the other. In the first area we are concerned particularly with pragmatic texts, standardized terminology and specialized translation (and conference interpreting), needing subject area expertise, and it has led to the hybrid texts as defined by Schäffner and Adab (1997) in the sense of “Eurospeak” or analyzed by Mohammed Didaoui (1996) in the United Nations context. In the latter area we are also concerned with operative and expressive texts (and dialogue interpreting), needing cultural expertise and sensitivity, and it is here that Wittgenstein’s quote can be most meaningful, though for Translation Studies both areas are central for the debate. What is important for our topic today is however that the academic discourse of Translation Studies cannot be classed as a discourse system in the above sense, and 1. This is in the title of the paper, in accordance with the definition in the Collins Dictionary of the English Language, but not without irony, labelled as “Anglo-Saxon”. 2. This is a popularized version of a sentence from Wittgenstein’s Tractatus logico-philosophicus (Wittgenstein 1989, published in English by Routledge & Kegan Paul in 1922), which runs as follows: “Dass die Welt meine Welt ist, das zeigt sich darin, dass die Grenzen der Sprache (der Sprache, die allein ich verstehe), die Grenzen meiner Welt bedeuten.“ (Wittgenstein 1989:€136)

 Mary Snell-Hornby

I seriously doubt the suitability of any one natural language to monopolize global communication in our discipline. Sophisticated scholarly discourse is invariably rooted in its own epistemological, methodological and rhetorical tradition which goes far beyond the mere language code. Michael Clyne’s studies of English and German academic writing (1991) give ample evidence of this: the one typically linear, inductive, symmetrical and reader-oriented, the other digressive, deductive, assymetrical and author-oriented. Further observations show that English scholarly discourse makes more use of concrete examples, intertextuality (particularly quotations) and metaphor, whereas German (along with other European traditions) favours an abstract, nominal style with carefully defined terminology. When the English language code is used in the latter stylistic tradition in an attempt to appear scholarly, the result is frequently interpreted as convoluted jargon. Added to this is the fact that at this level English is far from being as easy as is commonly thought: its immensely rich and subtle lexis, the problems of tense, aspect, focus, complementation, cohesion, collocation and concordance can result in interference errors that make many a non-native academic treatise not only heavy going but even barely comprehensible. A far better solution would in my opinion be the acceptance of more languages in the Translation Studies debate, and of these the scholar would require passive or reading knowledge. There is surely nothing sensational about this. It should be self-evident that anyone involved in Translation Studies needs a basic interest in foreign languages as the means of expression of cultures different from their own. Never has the learning of foreign languages been made so easy as in recent years, and never has travel been more widespread, enabling people to experience the foreign worlds beyond that of their own language. And it is a truism that for a practising translator and interpreter, language proficiency is the first of several prerequisites. To a certain extent the same goes for the translation scholar, whereby the often put question “How many languages do you speak?” might be replaced by the question “How many languages do you understand?” (i.e. beyond the language pair with which he/she is professionally engaged). The key term is here “passive multilingualism”, as proposed in the early 1990s and developed for the Romance languages in 1998 by the EuroCom research group (see Vankúšová, in this volume). Such language competence might well provide a practicable solution for broadening the scope of Translation Studies – and enable qualified colleagues to publish and present conference papers in their mother tongue or active working languages (where these do not include English).

3. Possible solutions? Bearing these thoughts in mind I would now like to present a few basic hypotheses on the increasing “Anglicization” of Translation Studies and add some alternative solutions: 1. Though English has been described as “world property”, an awareness might well be developed that Translation Studies is not the property of English. Ideally,



Is Translation Studies going Anglo-Saxon 

2.

3.

4.

5.

6.

7.

translation scholars should have some degree of proficiency in three or more languages. In the scholarly debate, multilingualism, especially passive multilingualism, should be encouraged. The same principle would apply to the content of academic journals and conference programmes, which could encourage the use of languages other than English. (This does not necessarily refer to administrative matters such as publisher’s guidelines or information on organization etc. – though here the English should at least be reasonably correct.) In academic publications, quotations and references from other languages could first be presented in their original form and then (if necessary) in translation. In this way the different approaches, schools and backgrounds are made visible. In conference presentations technical aids can be put to better use to further understanding of contributions in other languages. For example, in a paper presented in French or Spanish, PowerPoint, which encourages a certain degree of redundancy, could be used to highlight key words in English or show lines of argumentation through diagrams. Hand-outs could serve a similar purpose for lengthy passages of text. A special course in the problems of academic English might be compiled for those many scholars who are clearly unaware of them. People should not underrate the complexities of English or overrate their own language competence. Special courses in “Writing English” – especially papers for publication – are highly recommendable. International conferences might be organized without English as a conference language. (My own experience of such conferences has been extremely positive: new perspectives are revealed, other problems are addressed, and unexpected conclusions reached, cf.€Snell-Hornby 2006:€1–2). Finally, a concrete step would be the introduction of bridge languages to further and encourage passive multilingualism.

4. The potential of bridge languages What is a bridge language? Put very simply, this is one of a family of languages (such as Germanic, Romance or Slavonic for the European context) with prototypical basic features facilitating access to other languages of the same family, especially in the form of passive listening and reading skills – and it is not to be confused with dominant languages of the family such as French or Russian which have been widely taught for historical or political reasons. The concept is explained by Martina Vankúšová (in this volume), who has done research into the potential of Slovak as a bridge language for the Slavonic languages and cultures, which are still hardly visible in the context of the European Union, despite the latter’s avowed principle of democratic multilingualism. The concept of the bridge language need of course by no means be limited to Europe – it could equally be applied to the Dravidian languages of Southern India or the

 Mary Snell-Hornby

regional languages of the Philippines. Within the context of Translation Studies it would be a valuable means to counteract the fixation on the norms of English, thus providing a forum for multilingual conferences and publications and encouraging a form of communication which respects cultural differences and does not depend on a single, over-simplified lingua franca. “The limits of my language are the limits of my world.” This wise statement might provide an answer to those who may be wondering why it is necessary to complicate the issue where the use of English has seemingly made everything so convenient and easy. The global language English has permeated our lives, even as translation scholars: it makes sense to publish in English, for the publisher because it promises wide circulation and for the scholar because his/her work can then hope for world-wide reception. What is widely overlooked are the communication problems caused by idiosyncratic usage and the ensuing misunderstandings involved in a lingua franca of this kind, but in particular the stultifying effect of immensely complex cultural and linguistic material being monopolized by a single language. If we see language as a communicative and epistemological system which plays a crucial part in constructing the group identity of those who speak it (cf.€Snell-Hornby, 2008: 74), we become aware of the importance individual languages have in a discipline like Translation Studies. It is a feature of Global English that as “world property” it does not belong to any such group and does not arouse feelings of identity, and that is what makes it fair game for casual use – or in sophisticated academic discourse even abuse. It might be the function of bridge languages to help overcome this, to further insight into other cultures, heighten respect for the integrity of other groups of speakers and extend the limits of the translation scholar’s world, particularly where this involves the still “undiscovered country” of our discipline.

References Clyne, M. 1991. “Zu kulturellen Unterschieden in der Produktion und Wahrnehmung englischer und deutscher wissenschaftlicher Texte.” Info DaF 18 (4): 376–383. Didaoui, M. 1996. Communication Interferences in a Multilingual Environment. The Role of Translators. Vienna: unpubl. Doctoral thesis. Koskinen, K. 2004. “Shared culture? Reflections on recent trends in Translation Studies.” Target 16 (1): 143–156. Schäffner, C. and Adab, B. 1997. “Translation as intercultural communication – Contact as conflict.” In: Translation as Intercultural Communication. Selected Papers from the EST Congress – Prague 1995, M. Snell-Hornby, Z. Jettmarová and K. Kaindl (eds), 325–337. (Benjamins Translation Library 20), Amsterdam/Philadelphia: John Benjamins. Snell-Hornby, M. 2006. The Turns of Translation Studies. New paradigms or shifting viewpoints? (Benjamins Translation Library 66). Amsterdam/Philadelphia: John Benjamins.



Is Translation Studies going Anglo-Saxon  Snell-Hornby, M. (2008). Translationswissenschaft in Wendezeiten. Ausgewählte Beiträge zwischen 1989 und 2007, M. Kadric and Jürgen F. Schopp (eds) (Studien zur Translation 20), Tübingen: Stauffenburg. Widdowson, H. 1994. “The ownership of English.” The TESOL Quarterly 28: 377–389. Wittgenstein, L. 1989. Logisch-philosophische Abhandlung. Tractatus logico- philosophicus. Kritische Edition, B. McGuinness and J. Schulte (eds). Frankfurt: Suhrkamp.

Slowakisch: Brückensprache zur slawischen Welt? Möglichkeiten und Grenzen einer kleinen EU-Sprache Martina Vankúšová

University of Vienna, Austria

The paper discusses the significance of diversified multilingualism, for Europeans in general and for translators and interpreters of EU-translation services in particular. Officially, the European Union is multilingual but it has long since been part of “the empire of English” as defined by Snell-Hornby (2006). Translations from the simplified Euro-English into other EU-languages make up for over 60% of the total number of translations, and they generally raise no difficulties, but difficulties may well occur when an EU-commissioner, for example, chooses a proverb in his mother tongue as the slogan for an EU-campaign. How can such translation challenges be approached? Can a translator be expected to have a command of all official EU-languages? A possible solution is offered by the multilingual intercomprehension method EuroCom based on developing good linguistic competence in a bridge language as a means of gaining access to other related languages. For Romance languages, this is French and for Germanic languages it is English. They are are also procedural languages of EU-institutions. With Slavonic languages, the situation is more difficult and the ensuing problems are presented and discussed in the paper. Keywords: multilingualism in Europe, bridge languages, Slavonic languages

1. Einleitung Im Jahre 2004, zur Zeit der großen EU-Erweiterung, gab es am Wiener Flughafen Schwechat großflächige Plakatwände zu sehen, mit denen man unter dem Slogan EUROPA BRINGT FRIEDEN für die Friedenseinsätze der Europäischen Union werben wollte. Das helle EU-Blau der ersten zwei Buchstaben sollte klar machen, dass nicht der gesamte Kontinent sondern nur die EU gemeint war. Das Schwarz der

 Martina Vankúšová

übrigen Buchstaben wurde dadurch umso deutlicher und man las diesen Satz folglich als ROPA BRINGT FRIEDEN. Dies ergab für die meisten potentiellen Adressaten wenig Sinn, wohl aber für die Polen, Tschechen und Slowaken, d.h. für immerhin rund 50 Millionen EU-Bürger. Im Polnischen, Tschechischen und Slowakischen heißt ROPA nämlich ERDÖL. Werbung kennt bekanntlich keine Grenzen. Die Hauptsache ist, dass die Botschaft ankommt. Daher machte ich mir als Slowakin über die Mischung aus Slowakisch und Deutsch überhaupt keine Gedanken. Im Gegenteil. Angesichts der damaligen weltweiten FriedenskundÂ�gebungen und Proteste gegen den Irak-Krieg mit dem allgegenwärtigen Slogan OIL FOR BLOOD dachte ich zuerst, dass nun sogar schon am Flughafen Wien gegen den Krieg protestiert wird. Umso erstaunter war ich, als ich näher hinschaute und die den Frieden bringenden Soldaten auf den Panzern mit dem Verweis auf die Webseite der Europäischen Union sah. Heutzutage ist es gang und gäbe, dass große internationale Konzerne nach dem Motto Global handeln, lokal denken gezielte Werbemaßnahmen veranstalten und dabei penibel darauf achten, dass ihre Produkte und ihr Image bei den potentiellen anderssprachigen Kunden nicht durch ungeschickt gewählte Slogans Schaden erleiden. Auch für die Europäische Union mit den weltweit größten Sprachdiensten dürfte eine derartige Imagepflege keine unzumutbare Herausforderung sein.

2. English only? Nach der EU-Erweiterung im Mai 2004 sind auch alle neun neuen Amtssprachen ArbeitsÂ�sprachen der EU geworden. Neu ist seitdem aber auch die Bezeichnung für die bisherigen inoffiziellen Arbeitssprachen: Englisch, Französisch und Deutsch werden seitdem als Verfahrenssprachen (engl. procedural languages) bezeichnet, wobei Englisch in den diversen Übersetzungsstatistiken eindeutig führt. Dem hohen Prozentsatz der slawischsprachigen EU-Bürger wurde auch 2007, nach dem Beitritt Bulgariens, jedoch nicht Rechnung getragen. Es werden Relaissprachen benutzt. Für die neuen Mitgliedstaaten bedeutet dies, dass vor allem Übersetzer in die und aus den Verfahrenssprachen Englisch, Französisch und Deutsch gefragt sind. Laut Reinhard Hoheisel aus der deutschen Sektion der Generaldirektion Übersetzung (DGT) der Europäischen Kommission sind die Chancen, mit einer exotischen SprachkomÂ�biÂ�nation angestellt zu werden, selbst für die Übersetzer mit der Muttersprache Deutsch eher gering, da auch die Übersetzungen ins Deutsche zu 75% aus dem Englischen und zu 15% aus dem Französischen angefertigt werden (Hoheisel, persönliche Mitteilung, Mai 2007). In seiner simplifizierten Euro-Englisch-Form ist Englisch eine leichte Sprache, und in der Regel bereitet das Übersetzen aus diesem Englisch in andere EU-Sprachen keine Schwierigkeiten. Diese treten aber sehr wohl auf, sobald beispielweise ein EU-Kommissar aus einem neuen EU-Land für die von ihm initiierte Kampagne als Motto ausgerechnet ein Sprichwort in seiner Landessprache wählt.



Slowakisch: Brückensprache zur slawischen Welt? 

Unter dem Motto Koľko jazykov vieš, toľkokrát si človekom verabschiedete die Europäische Kommission am 22.11.2005 ihre allererste Mitteilung zum Thema Mehrsprachigkeit, in der die Kommission die Mitgliedstaaten auffordert, das Lehren, Lernen und die Verwendung möglichst vieler Sprachen zu fördern. Aus diesem Anlass wurde auf der Website EUROPA ein neues Sprachenportal in allen Amtssprachen eingerichtet, wo unter anderem auch die erwähnte Mitteilung noch längere Zeit abrufbar war. Derzeit (Juli 2008) kann man unter den Pressemittelungen des zuständigen slowakischen EU-Kommissars Ján Figeľ1 nur noch eine ZusammenÂ�fassung und das Sprichwortmotto finden. Dieses Sprichwort2 würde wörtlich ins Deutsche übersetzt etwa folgendermaßen lauten: Wie viele Sprachen du kannst, so viele Male bist du Mensch. Damit ist gemeint, dass in jeder Sprache ein Weltbild verborgen ist, und dass man, je mehr Sprachen man “kann”, desto mehr verschiedene Weltbilder gewinnt, oder – um WittgenÂ�stein zu bemühen – desto mehr Grenzen überwindet (vgl. SnellHornby, in diesem Band). Die Mitteilung wurde auf Englisch verfasst und das Sprichwortmotto auf Slowakisch, mit der Übersetzung ins Englische und dem Hinweis in Klammern, dass es sich hierbei um ein slowakisches Sprichwort handle, rechts oben angeführt. In den, in die einzelnen EU-Sprachen übersetzten Fassungen der Mitteilung zur Mehrsprachigkeit wurde dieses Sprichwortmotto auf Slowakisch beibehalten, sowie auch der Hinweis in Klammern. Hier die einzelnen Übersetzungslösungen: EN FR DE DA ES NL IT SW PT FI EL CS ET HU LT LV MT

The more languages you know, the more of a person you are. Plus tu connais de langues, plus tu es humain. Je mehr Sprachen du sprichst, desto mehr bist du Mensch. Hver gang du lærer et nyt sprog, bliver du et nyt menneske. Cuantas más lenguas conozcas, más persona eres. Hoe meer talen je kent, hoe rijker je bent. Quante lingue conosci, tante persone sei. Varje gång du lär dig ett nytt språk blir du en ny människa. Aprende línguas e serás alguém. Jokaisen oppimasi kielen myötä kasvat ihmisenä. ΄Οσο πιο πόλλες γλώσσες ξέρεις, τόσο πιο άνθρωπος είσaι. Kolik jazyků umíš, tolikrát jsi člověkem. Mida enam keeli oskad, seda täisväärtuslikum oled inimesena. Ahány nyelven tudsz, annyi ember vagy. Kiek kalbų moki, tiek kartų esi žmogus. Jo vairāk valodu proti, jo pilnīgāka personība esi. Aktar ma taf ilsna, aktar inti bniedem sħiħ.

1. Nach dem EU-Beitritt Rumäniens im Januar 2007 wurde Leonard Orban zum EU-Kommissar für Mehrsprachigkeit bestellt. 2. Das diskutierte Sprichwortmotto dürfte seine Herkunft im Lateinischen haben, in der Slowakei ist es jedoch allgemein als eine Volksweisheit bekannt.

 Martina Vankúšová

PL SK SL

Im więcej znasz języków, tym większym jesteś człowiekiem. Koľko jazykov vieš, toľkokrát si človekom. Več jezikov znaš, več veljaš.

Gewiss könnte man der Übersetzung dieses Sprichwortes einen selbständigen Beitrag widmen und die einzelnen Übersetzungslösungen diskutieren. An dieser Stelle soll jedoch nur festgestellt werden, dass der Hinweis “Slowakisches Sprichwort” kaum beachtet wurde. Die meisten Übersetzer gingen offensichtlich von der englischen oder französischen Übersetzung aus. Die Ergebnisse sind folglich recht vage. Es ist zwar allgemein bekannt, dass Übersetzer in den Übersetzungsdiensten der EU unter starkem Zeitdruck arbeiten, und dass viele “politisch unheikle” Texte auch “außer Haus” vergeben werden – vielleicht auch die hier diskutierte Mitteilung. Warum aber sollte man eigentlich ein slowakisches Sprichwort über das Englische ins Slowenische oder Polnische übersetzen? Und was konnten sich die slowenischsprachigen Adressaten beim Lesen der slowenischen Fassung denken, in der nach dem slowakischen Koľko jazykov vieš, toľkokrát si človekom das slowenische Več jezikov znaš, več veljaš folgte? Auf Slowenisch hätte das Sprichwort nämlich fast klanggleich mit Tschechisch und Slowakisch Kolikor jezikov znaš, tolikokrat si človek heißen können. Die deutsche Übersetzung ist vor ihrem geschichtlichen Hintergrund wahrscheinlich die fragwürdigste; und über die Schwächen des portugiesischen Wortlauts kann nur schwer hinwegtrösten, dass man die Slowakei vielleicht wieder einmal mit Slowenien durcheinander bringen wird. Hinzu kommt, dass in der portugiesischen Übersetzung im Sprichwort die Buchstaben mit slowakischen diakritischen Zeichen jeweils durch ein Kästchen ersetzt wurden. Gerade für Übersetzer und Dolmetscher, die in den internationalen Organisationen arbeiten, ist es wichtig, neben bestimmten Rechtschreib- und Ausspracheregeln die gängigsten Techniken von mehrsprachiger Textbearbeitung zu kennen, werden sie ihnen doch zumindest bei Eigennamen immer wieder begegnen. Kann man jedoch von Übersetzerinnen und Übersetzern verlangen, dass sie alle 23 EU-Amtssprachen beherrschen?

3. Eurocomprehension Die inzwischen international mehrfach ausgezeichnete Mehrsprachigkeitsmethode EuroCom bietet einen möglichen Ansatz, indem sie deutlich macht, dass die meisten europäischen Sprachen keine echten Fremdsprachen sind. EuroCom steht für Eurocomprehension, ein Akronym für die Interkomprehension in den drei großen Sprachgruppen Europas, der romanischen, der germanischen und der slawischen, an denen fast alle Europäer mit einer Erst- oder Zweitsprache partizipieren. Die Forschergruppe EuroCom wurde 1998 in Hagen gegründet und entwickelte, zunächst ausgehend von der Frankfurter Romanistik, eine Sprachvermittlungsmethode zum Erreichen rezeptiver Kompetenzen in allen romanischen Sprachen. Mittlerweile



Slowakisch: Brückensprache zur slawischen Welt? 

umfasst die Forschergruppe Wissenschaftler an zwölf Universitäten in sechs Ländern. Die europäische Mehrsprachigkeit betrachten sie als Fähigkeit, mit unterschiedlichen Kompetenzgraden zu arbeiteten, um bei Bedarf ein verwandtes Idiom zu aktivieren und Kompetenzen zu erweitern. Diese Lernmethode wurde bereits in den 90er Jahren im Rahmen des Hildesheimer Drittsprachmodells unter der Leitung von Reiner Arntz mit Erfolg auf die romanischen und germanischen Sprachen angewandt, und zwar für Portugiesisch und Niederländisch, sowie später auch für Dänisch. Die EuroCom-Lernmethode basiert auf einer möglichst gut entwickelten sprachlichen Kompetenz in einer Brückensprache, die den Weg zu den verwandten Idiomen öffnen kann. Für die romanischen Sprachen ist es Französisch, für die germanischen Sprachen Englisch. Die Forschergruppe für slawische Sprachen EuroComSlav wurde 2001 unter der Leitung von Lew Zybatow gegründet. Als Brückensprache wurde das Russische gewählt.

4. EuroCom für slawische Sprachen Global gesehen nimmt die Zahl der Sprachen ab. Im slawischen Bereich vollzieht sich jedoch eine Art Sprachenvermehrung. So werden in der Wieser Enzyklopädie des europäischen Ostens (Okuka 2002) an die 40 slawische Sprachen aufgezählt. Die slawischen Sprachen haben ihre Ausgangssprache im Urslawischen und weisen heute noch sehr viele Gemeinsamkeiten auf allen Sprachebenen auf, am deutlichsten im sogenannten panslawischen Wortschatz, der allen slawischen Sprachen gemeinsam ist. Wer eine slawische Sprache beherrscht, partizipiert daher an dem Erbvokabular aller übrigen slawischen Sprachen. Im Gegensatz zu den romanischen Sprachen, für die in EuroComRom von etwa 500 panromanischen Wörtern ausgegangen wird, liegt die Zahl der panslawischen Vokabeln wesentlich höher: bei etwa 1500 Wörtern. Laut den Angaben der Webseite EuroComSlav wurden die Panslawismen aus elektronischen Korpora aktueller Printmedien in den einzelnen slawischen Sprachen exzerpiert. Insgesamt werden 438 Vokabeln in 9 slawischen Sprachen angeführt. Die Eintragungen für Tschechisch und/oder Slowakisch sind jedoch in 206 Fällen (somit in fast 50%) mangelhaft bzw. ganz falsch, wie z.B. die Infinitiv-Endung “-ti” bei den tschechischen Verben, die im heutigen Tschechischen nicht mehr im Gebrauch ist und mir eigentlich nur aus Volksliedern bekannt ist. Dies lässt zumindest die aktuellen Printmedien als Quelle in Frage stellen. Beim internationalen Wortschatz handelt es sich um Wörter überwiegend griechischen und lateinischen Ursprungs sowie um Eigennamen und geographische Namen, die in fast allen europäischen Sprachen vorkommen. Für EuroComSlav werden exemplarisch 300 InternaÂ�tionalismen angeführt, die zu den 3000 in der Alltagskonversation am häufigsten gebrauchten Wörtern gehören. 71% der angeführten Wörter kommen in allen slawischen Sprachen vor. Die Vorbereitung des Lehrwerks für die slawischen Sprachen war ursprünglich für 2004 geplant – außer den allgemeinen Informationen sind im Internet aber nur die

 Martina Vankúšová

Listen von Panslawismen und Internationalismen zugänglich. Im Juli 2007 kam eine Liste von LautentÂ�sprechungen hinzu und ein Hinweis darauf, dass die Multiplikatorenwebseite für die slawischen Sprachen im Rahmen eines weiteren EU-Projektes vollständig entwickelt werden sollte.

5. Brückensprache Russisch? Was die Brückensprachen anbetrifft, kommt im Prinzip jede Ausgangssprache als Brückensprache in Frage. Dennoch soll laut Angaben der EuroComSlav-Webseite Russisch eine besonders privilegierte Brückensprache für die anderen slawischen Sprachen sein. Russisch sei die am weitesten verbreitete slawische Fremdsprache und werde an allen universitären slawischen Seminaren und teilweise auch an Schulen gelehrt. Ein weiterer Vorzug, den die Ausgangssprache Russisch biete, bestehe darin, dass über Russisch der Zugang zu den kyrillisch geschriebenen slawischen Sprachen garantiert sei. Die lateinischen Alphabete der übrigen slawischen Sprachen seien aufgrund der Kenntnis des Deutschen ohnehin zugänglich. Außerdem sei Russisch eine UNO-Sprache, habe die meisten Sprecher unter den Slawen und in Zukunft werde es als Lingua franca zwischen den Völkern der Staaten der ehemaligen Sowjetunion an Bedeutung zunehmen, so dass es als einzige, wirklich tragfähige und erfolgversprechende Alternative erscheine. Die geographische Lage Russlands am Rande des slawischsprachigen Gebietes und der dadurch begrenzte Kontakt mit den anderen slawischen Sprachen sprechen jedoch eher gegen Russisch. Für einen Deutschsprachigen, der sich dadurch eine andere slawische Sprache erschließen möchte, ist es – bildlich gesprochen – eine viel zu lange Brücke. An deren Ende angelangt, wird er womöglich feststellen, dass er zu den anderen slawischen Sprachen doch durchs Wasser zurück waten muss. Russisch ist im deutschsprachigen Raum zwar die meist verbreitete slawische Fremdsprache, es liegt jedoch in diversen Statistiken weit hinter Englisch oder Französisch, die unter anderem auch aus diesem Grund als Brückensprachen für die jeweilige Sprachfamilie gewählt wurden. Um sich den Zugang zu den kyrillisch geschrieben slawischen Sprachen zu verschaffen, kann man das kyrillische Alphabet innerhalb von wenigen Stunden lernen. Russisch ist zwar eine UNO-Sprache und hat die meisten Sprecher unter den Slawen, es ist jedoch keine EU-Sprache und die Anzahl der EU-Bürger mit der Muttersprache Russisch spricht bei einem EU-Projekt auch nicht unbedingt für Russisch. Im ehemaligen Ostblock hatte Russisch die Funktion einer Art Lingua franca. Dass es aber in der nächsten Zukunft unter den Völkern der Staaten der ehemaligen Sowjetunion als Lingua franca wesentlich an Bedeutung zunehmen wird, wage ich im Anbetracht der jüngeren Geschichte zu bezweifeln.



Slowakisch: Brückensprache zur slawischen Welt? 

6. Brückensprache Slowakisch? Unter den slawischen Sprachen gibt es keine mit Französisch oder mit Englisch vergleichbare, sowohl historisch, als auch wirtschaftlich und kulturell verankerte Lingua franca. 40 Jahre Ostblock reichten für Russisch bei weitem nicht aus. Eben aus diesem Grund sollte man die Möglichkeit nutzen und für das EU-finanzierte EuroComSlavProjekt eine linguistisch geeignete slawische Sprache wählen. Dies wurde auch von etlichen Slawisten vorgeschlagen, worüber ich im Folgenden berichten möchte. Im Sommersemester 2001 wurde am Institut für Slawistik der Universität Wien unter der Leitung von Juliane Besters-Dilger eine Untersuchung zur spontanen Interkomprehension in den slawischen Sprachen durchgeführt, unter anderen mit folgendem Ergebnis: Wenn es ein vorrangiges Ziel ist, die fünf slawischen EU-Sprachen passiv in Westeuropa bekannt zu machen, ist aus sprachwissenschaftlicher Sicht das Russische als Ausgangssprache nicht geeignet: die kyrillische Schrift lässt es als ‚schwierige Sprache‘ erscheinen, es ist kaum argumentierbar, dass man die lateinisch verschrifteten Sprachen über Vermittlung einer kyrillisch verschrifteten Sprache erlernen soll und infolge vieler phonologischer, morphologischer und syntaktischer Besonderheiten ist Russisch nicht die slawische Sprache, die den besten Zugang zu allen anderen bietet; das wäre nach übereinstimmender Ansicht von Slawisten, die zu diesem Thema befragt wurden, das Slowakische. (Besters-Dilger 2002:€392)

Auch der Grazer Slawist Heinrich Pfandl empfiehlt primär Slowakisch, aber auch Slowenisch oder Tschechisch als “typische” slawische Sprachen, um sich den Zugang zu den anderen slawischen Sprachen zu erschließen: […] für das Argument, dass die Kenntnis einer slawischen Sprache den Zugang zu sämtlichen anderen slawischen Sprachen ermögliche, eignet sich Russisch denkbar schlecht. Russisch ist nicht nur territorial eine periphere slawische Sprache. (...) Sucht also jemand eine “typische” slawische Sprache, um sich den Zugang zu den slawischen Sprachen zu erschließen, so ist er etwa mit dem Slowakischen, ja selbst mit dem Slowenischen oder dem Tschechischen, besser beraten. (Pfandl 1995:€96)

Laut dem slowakischen Sprachwissenschaftler und einem der Autoren des Slawischen SprachÂ�atlanten Anton Habovštiak dominiert in der slowakischen Schriftsprache der westslawische lexikalische Charakter, zugleich ist jedoch auch die Bindung des Slowakischen an die südslawischen und ostslawischen Sprachen belegbar, so dass Habovštiak (1993) die zentrale Stellung des Slowakischen inmitten der slawischen Sprachen bestätigt sieht. Es ist unumstritten, dass alle neuen EU-Sprachen vom Status als Amtssprachen der EU profitieren. Neben einer gewissen internationalen Wahrnehmung ist es vor allem die Kommunikation innerhalb der Europäischen Union, die fast alle Gebiete der

 Martina Vankúšová

menschlichen Tätigkeit berührt und somit für den Wortschatz einzelner EU-Sprachen, in die übersetzt wird, eine nachhaltige Bereicherung bringt. Einige der neuen Mitgliedsländer sahen sich im Bezug auf den Beitritt zur EU sogar gezwungen, überhaupt eine eigene Terminologie zu schaffen, oder, wie im Falle der Slowakei, die bestehende Terminologie auszubauen und zu modernisieren. Im Hinblick auf das Slowakische kann man so auch dem Zerfall der Tschechoslowakei durchaus etwas Positives abgewinnen – in einem EU-Mitgliedsland Tschechoslowakei wäre die Lage des Slowakischen als “Zweitamtssprache” längst nicht so vorteilhaft. Wie hoch kann aber die Bereitschaft sein, eine neue Sprache zu erlernen, die wie das Slowakische 5 Millionen, oder wie das Slowenische lediglich 2 Millionen Sprecher hat? Auch wenn mitunter mit Slowakisch als “Esperanto der slawischen Sprachen” geworben wird und jeder Slowake früher oder später diese Eigenschaft seiner Muttersprache zu schätzen und einzusetzen lernt, sind die Sprachkurse für Slowakisch nicht unbedingt stark nachgefragt und an Universitäten im deutschsprachigen Raum wird es genau so selten angeboten wie Slowenisch. Eine slawische Verfahrenssprache wird es in der Europäischen Union wohl nicht so bald geben, aber gerade für Übersetzer und Dolmetscher, die in den EU-Sprachdiensten arbeiten, wären Kenntnisse einer der jetzigen oder künftigen slawischen EU-Amtssprachen sehr hilfreich und nützlich. Im EuroCom-Ansatz sehe ich daher einen wertvollen Zugang zu einer diversifizierten Mehrsprachigkeit der Europäer der Zukunft, wobei ich darauf hinweisen möchte, dass man sich auch der Verwendung von kleineren Brückensprachen nicht verschließen sollte, bei denen die zu überbrückende “Sprachkluft” weniger breit ist.

Literatur Arntz, R. 2004. “Drittsprachen in der Übersetzerausbildung: Portugiesisch als “Kontrastsprache”.” In Und sie bewegt sich doch... Translationswissenschaft in Ost und West. Festschrift für Heidemarie Salevsky, I. Müller (ed.), 17–31. Frankfurt am Main: Lang. Besters-Dilger, J. 2002. “Spontane Interkomprehension in den slawischen Sprachen.” In Eurocom. Mehrsprachiges Europa durch Interkomprehension in Sprachfamilien. Tagungsband des Internationalen Fachkongresses im Europäischen Jahr der Sprachen 2001 Hagen, 9.-10. November 2001, G. Kischel (ed.), 385–393. Hagen: Fernuniv. Hagen. Habovštiak, A. 1993. Zo slovensko-slovanských lexikálnych vzťahov (Zu den slowakisch-slawischen lexikalischen Beziehungen). Bratislava: Vydavateľstvo Slovenskej akadémie vied. Pfandl, H. 1995. “Zum Bonus und Malus des Russischen.” In Sprachenpolitik in Mittel- und Osteuropa, R. Wodak and R. de Cillia (eds), 93–102. Wien: Passagen Verlag. Okuka, M. (ed.). 2002. Wieser Enzyklopädie des Europäischen Ostens, Band 10. Snell-Hornby, M. 2006. The Turns of Translation Studies. New paradigms or shifting viewpoints? Amsterdam/Philadelphia: John Benjamins.



Slowakisch: Brückensprache zur slawischen Welt?  Zybatow, L. N. (ed.). 2004. Translation in der globalen Welt und neue Wege in der Sprachund Übersetzerausbildung. Innsbrucker Ringvorlesungen zur Translationswissenschaft II. Frankfurt am Main: Peter Lang Verlag Servicecenter zur Europäischen Mehrsprachigkeit http://www.eurocomcenter.eu (accessed 17.08.2007) Institut für Europäische Interkomprehension http://www.eurocomprehension.info (accessed 17.08.2007) Webseite für die Slawische Sprachenfamilie http://www.eurocomslav.de (accessed 17.08.2007) Ján Figeľ, EU-Komissar für Bildung, Kultur und Mehrsprachigkeit http://ec.europa.eu/commission_barroso/figel/index_de.HTM (accessed 16.05.2006)

Translation Studies and mass media research Rachel Weissbrod

Bar Ilan University, Ramat Gan, Israel

Taking Israel as a case-study, this paper argues that Translation Studies can contribute to research dealing with multilingualism and multiculturalism as reflected in the mass media. Following ideological and demographic changes in Israeli society, the use of languages other than Hebrew is nowadays apparent in the public as well as in the private sphere. The change is clearly recognizable in Israeli mass media, all the more so because in the age of trans-national broadcasts, keeping multilingualism under control is no longer feasible. Socio-linguists and communication researchers have investigated the new situation. Distinguishing between Hebrew and non-Hebrew media, they have sought answers to questions such as: What is the impact of the exposure to foreign languages on the globalization and Americanization of Israel? What are the implications of using minority languages – mainly Arabic and Russian – for social integration, and how does it affect the construction of social identity? However, research has usually ignored translation which blurs the very distinction between media in Hebrew and other languages. The present paper attempts to point out what can be gained by taking translation into consideration in investigating the media in Israel and possibly in other countries. It deals mainly with Israeli television which provides translation into Hebrew, Arabic and Russian on a daily basis. Keywords: multilingualism, multiculturalism, media translation, social identity, social integration

1. Introduction This paper sets out to demonstrate the relevance of translation and the potential contribution of Translation Studies to mass media research. It focuses on research dealing with the multilingualism and multiculturalism of Israel as manifested in its mass media. Israel was conceived as a nation-state and, as Stewart Hall (1991) has aptly remarked, nation-states strive to obliterate the differences between ethnic, religious and other groups in the name of a unifying idea. However, since Israel consists of Jews and

 Rachel Weissbrod

Arabs, veteran Israelis and newly arrived immigrants, it has always been linguistically and culturally heterogeneous. Two of the languages used by Israelis, Hebrew and Arabic, enjoy the status of official languages. In fact, Hebrew is the dominant one, protected not just by law but also by the national ideology of the majority. Since the 19th century, Hebrew was transformed from a non-working language used mainly for religious purposes to a “living” one, a process that had been initiated by the Jewish Enlightenment Movement in Europe and continued by Zionism. Its revival is often referred to as “a miracle”. Arabic, which is spoken by the Arab minority (1.4 million; about 20% of the 7 million population of Israel), is marginal despite its official status and most Israeli Jews do not speak or read it. Other languages are used mainly by immigrants (Spolsky and Shohamy 1999). Until recently, the use of the immigrants’ languages was limited due to a strict melting-pot policy. They were used mainly in private communication (Even-Zohar 2005) and in some public domains controlled by the state authorities, such as Israeli radio (Caspi and Limor 1999). A gradual decline of the melting pot-policy and a mass immigration from the former Soviet Union in the 1990s, have brought about great changes. Nowadays, languages other than Hebrew are used at cultural events, in the educational system and in other public domains. This applies especially to Russian, the language of almost 800,000 immigrants from the former Soviet Union.1 The decline of the melting-pot policy is part of globalization processes, whose effect is also manifest in the penetration of English into Israeli culture. Though few Israelis use English to communicate with each other, it is encountered everywhere – in advertisements, street signs, names of companies, shops and restaurants, and in surfing the internet. English is an agent of globalization (Crystal 1997), but its penetration into Israeli culture is also one of the symptoms of its growing Americanization (Rebhun and Waxman 2000; Segev 2002). The changes described are clearly recognizable in the Israeli mass media; all the more so because in the age of trans-national broadcasts, keeping multilingualism under control is no longer feasible. A large number of Israeli newspapers and magazines are published in languages other than Hebrew, including Arabic, English, Yiddish, Russian, Polish, Hungarian, French, German and Amharic (Epstein and Kheimets 2006). Radio broadcasts in Arabic, English and the immigrants’ languages are provided by Kol Israel (“The Voice of Israel”), the radio branch of Israel Broadcasting Authority (IBA) as well as by regional radio stations which legally operate on Israeli territory since the 1990s (Caspi and Limor 1999). Israeli television, too, has local channels in Russian, Ukrainian, Georgian, Amharic and Tigrigna – the languages of the largest immigrant groups. In addition, Israelis can access foreign channels via satellite 1. The numbers of immigrants and Israeli Arabs (not including the inhabitants of territories occupied since 1967) are taken from Statistical Abstract of Israel 2007 (No. 58) published by the Israeli Central Bureau of Statistics. On line at: http://www1.cbs.gov.il/reader (accessed on November 27, 2007).



Translation Studies and mass media research 

and cables. In November 2007, the digital basic package of the local cable company included 45 channels in languages other than Hebrew. The main ones are English (17), Arabic (10) and Russian (5). Other channels are in German (3), French (3), Italian (2), Turkish (2), Hungarian (1), Spanish (1) and Hindi (1). The multilingualism of the mass media consumed in Israel has attracted the attention of socio-linguists and communication researchers who have sought answers to questions such as: – What role do the mass media play in processes of globalization and Americanization which Israel currently undergoes? – What are the relations between Hebrew and the minority languages on one hand, and between the minority languages themselves, on the other hand, in Israeli mass media? – What is the relationship between the consumption of Hebrew and non-Hebrew mass media and the construction of social identity? – Does the consumption of mass media in languages other than Hebrew lead to the segregation of the immigrants or, on the contrary, allow for a smoother and more gradual integration? The questions are varied, and so are the answers. However, the works of research I have examined are similar in that they distinguish between Hebrew and non-Hebrew mass media and do not take account of translation which blurs this very distinction. In this paper, I would like to suggest the possible gains of adding a Translation Studies perspective to research dealing with mass media in Israel. Special reference will be made to Israeli television which provides translation into Hebrew, Arabic and Russian on a daily basis.

2. English and globalization Epstein and Kheimets (2006) seek an answer to the question of whether the exposure of Israelis to mass media in foreign languages promotes the domination of English and the Americanization of Israel. They refer to the often heard claim that all over the globe, English has become the “hegemonic” and “neo-colonialist” language which not only creates linguistic and communicative inequality and discrimination between the speakers of English and those of other languages, but also indirectly controls many aspects of their lives. Examination of the mass media in Israel leads them to call this claim in question. Globalization, they say, seems to cause neither the total dominance of English nor the decline of other languages. On the contrary, the technological innovations associated with globalization contribute to the flourishing of both print and electronic media in numerous languages. The accessibility of TV channels in a variety of languages via cables and satellite testifies to this.

 Rachel Weissbrod

Epstein and Kheimets highlight the significance of the accessibility of TV channels in many languages, but they do not refer to the fact that most of them are only accessible to viewers who speak the relevant languages. The majority of Israelis depend on translation in watching TV productions in languages other than Hebrew. Translation into Hebrew is provided by the local channels as well as by some “international” ones which are offered to subscribers of the cable subscribers and satellite companies (such as BBC Prime, MGM, Hallmark and National Geographic). Due to economic considerations, the main mode of translation is subtitling, in which source and target languages are copresent, whereas dubbing is confined to children’s films and programs.2 Since the majority of imported films and programs on the local channels are English-speaking, and the “international” channels mentioned above broadcast exclusively in English, Israelis’ exposure to English is immense. This leads to a certain modification of Epstein and Kheimets’ claim regarding the relative position of English in Israeli TV. According to Danan (1991:€612), dubbing is “an assertion of the supremacy of the national language and its unchallenged political, economic and cultural power within the nation’s boundaries”. By implication, dubbing – which is the norm in many European countries today – can be viewed as a sort of resistance to the growing power of English. Israel, which is to a large extent a subtitling country, does not manifest such resistance.

3. Hebrew vs. the minority languages Epstein and Kheimets are interested in the relationship between English and other foreign languages accessible to mass media consumers in Israel. Other researchers are more concerned with the relations between Hebrew and the variety of languages used by Israelis. The successful transformation of Hebrew into a “living” language, on the one hand, and the awareness that it is only spoken in Israel, on the other hand, led Israelis, including academicians, to wonder if it is under threat. Several attempts have been made to protect it by legislation. One of them is the proposal of a Hebrew language law (Shavit 2002) which refers specifically to the media: “The Hebrew language will be given precedence in all Israeli media. The broadcasting of Israeli films, plays, poems and songs in Hebrew will be guaranteed” (my translation).3 Socio-linguists such as Spolsky and Shohamy (1999) do not share this anxiety. They believe that the current multilingualism of Israel does not “threaten” Hebrew: Hebrew at the moment is well enough entrenched in Israeli general language practice to be unthreatened, and to be able safely to relax some of the excessive 2. See Kaufmann (2002) for the philosophical and ideological implications of “co-presence” in translation. 3. Zohar Shavit is a prominent researcher in the field of culture studies who served as a cultural affairs advisor to the Minister of Science, Culture and Sports and as a member of the Board of Directors of the Second Television and Radio Authority.



Translation Studies and mass media research 

defensiveness that is endangering the other languages and that is sapping away the national language capacity (ibid.: 257).

As regards English, they think that “a serious danger signal would be to find minority members or new immigrants preferring to learn English rather than Hebrew”; but so far, “Israeli Arabs and Russian immigrants are both quite definite in seeing the instrumental and economic value of Hebrew over English” (ibid.: 257). TV translation could have provided them with more evidence. Though Hebrew is far from exclusive, it is unquestionably the major target language on Israeli TV. This is manifested as follows: a. Translation into Hebrew is the most common one. As mentioned, it is available regularly on the local channels. Among the non-local ones, Hallmark is the only one which currently provides Russian in addition to Hebrew translation. b. Due to a recent law made in consideration of the deaf and hearing-impaired, local films and programs are broadcast with subtitles in the language of origin. Since most local productions are in Hebrew, the subtitles, which can be considered an intra-lingual translation, are in Hebrew, too. c. Dubbing, which is only provided in children’s films and programs, is always in Hebrew. This ensures that the younger generation is being raised on Hebrew (as long as they watch local channels). Based on their conviction that Hebrew is under no threat, Splolsky and Shohamy (ibid.) express their opinion that a better maintenance of the weaker languages is in order. One area in which a better maintenance is obviously needed is TV translation. While translation into Hebrew is usually direct, translation into other languages is mediated through Hebrew with all the implications for quality. Based on viewers’ responses and a preliminary research of Arabic subtitling (Weissbrod 2008), the poor quality of the mediated translation is manifested in ungrammatical Arabic, literal translation of Hebrew idioms and puns, and wrong deciphering of polysemic Hebrew words which reflect its total subordination to the mediating one.

4. Arabic vs. Russian Both Arabic and Russian are marginal in comparison with Hebrew, but they do not occupy an equal position in the Israeli mass media. According to Adoni, Caspi and Cohen (2002, 2006) who have compared the local production of journals and newspapers in the two languages, Russian is in a much better position. Examination of their relative positions vis-à-vis each other in TV translation sheds more light on the issue. Due to regulation, on the one hand, and the high percentage of Russian speakers among the subscribers of the cable and satellite companies on the other hand, Russian

 Rachel Weissbrod

translation has become quite common in Israeli television today.4 It is provided either in the form of subtitles, or voice-over, or both. Voice-over as a mode of translation was practically unknown in Israel before the immigration from the former Soviet Union in the 1990s. It has probably been chosen because it is cheaper than dubbing and the immigrants had become accustomed to it in their countries of origin (see Grigaravičiūtė and Gottlieb 1999; Orero 2004 for the prevalence of voice-over in Eastern Europe). The recent move to digital broadcasting in Israeli TV has made it feasible to broadcast imported films and programs with Russian voice-over as an option, without forcing it on viewers who prefer to listen to the original – usually English – dialogue. Locally produced Hebrew programs, too, are often broadcast with Russian translation of either type or both. Channel 10 (which is financed by commercials, yet accessible only to the subscribers of the cable and satellite companies) is the most consistent in this respect: it provides Russian translation in all its broadcasts, including the locally produced Hebrew ones, from 17:00 o’clock daily. Translation into Arabic is less widespread despite its status as an official language. It is provided primarily by the low-rating public channels which are subject to the Israel Broadcasting Authority Law. The commercial channels as well as the cable and satellite companies are subject to regulation that requires a certain amount of broadcasting in – or translating into – Arabic. But they only provide the required minimum. Since they are motivated by commercial considerations, they are much more attentive to the needs of the Russian-speaking minority. From the point of view of the cable and satellite companies, Israeli Arabs are unattractive customers, because they form only a small percentage of the subscribers. According to a recent survey reported by Musawa (The Palestinian Center for the Independence of the Judiciary and the Legal Profession), 83% of the Arab households in Israel watch Arab channels produced in the neighboring countries by means of private satellite dishes (Pollack 2007; see also Adoni, Caspi and Cohen 2006:€155). The marginality of Arabic translation in comparison not only with Hebrew but also with Russian translation is probably due to commercial considerations, but it is nevertheless charged with ideological and political implications. The two minority groups, Israeli Arabs and immigrants from the former Soviet Union, are essentially different. The immigrants are intent on retaining their cultural distinction and tend to regard Israeli culture as inferior to their culture of origin (Gomel 2006). However, most of them see themselves, and are seen by the majority, as an integral part of the Jewish people and of Israel as its state. As opposed to that, Israeli Arabs conceive themselves as a national minority. They do not share with the immigrants their sentiments for Hebrew as the ancient language of the Jews and they attach great importance to Arabic as a symbol of their national distinction (Pinto 2007). Whether the marginality 4. According to a press report (Galily 2004), 94% of Russian-speaking households – compared to 75% of veteran Israeli households – have access to multi-channel television. Noah Atlan of Yes reported to me (in May 2006) that 25% of the company’s subscribers are Russian-speaking.



Translation Studies and mass media research 

of Arabic translation in Israeli TV results from Israeli Arabs’ preference for foreign stations or leads to it, it is liable to increase their alienation from the state.

5. Dualism and a hybrid identity Acknowledging the inequality between the two minority groups leads from questions of production and supply to questions regarding the consumers themselves: What use do they make of the multilingual media at their disposal? What is the relationship between their consumption of media and the way they construct their social identity? These questions have been addressed by Adoni, Caspi and Cohen (2002, 2006). In 1999, they conducted a telephone survey prepared on the basis of previously held focus groups. In the survey, Israeli Arabs and immigrants from the former Soviet Union were asked about their media consumption habits. Media included newspapers, radio and television. In order to find out the relationship between media consumption and social identity, the interviewees were also asked about their attitude towards preserving their minority identity on the one hand, and adopting an Israeli identity on the other hand. Four groups of media consumers have been defined: adapters, who are high consumers of media in the language of the majority and low consumers of media in the language of the minority; separatists, who are high consumers of media in their native language and low consumers of media in the language of the majority; dualists, who are high consumers of media in both the language of the majority and their native language; and “detached” people who are low consumers of media in general. The distribution of the Israeli Arabs according to their media consumption patterns demonstrated that more than half (51.9%) were dualists. Only one third of the interviewees (32.7%) were separatists. Each of the other groups, the adapters and the detached, formed less than 10% (Adoni, Caspi and Cohen 2006:€82). The prevalence of dualism among Israeli Arabs makes sense considering the fact that many of them are fluent in both Arabic and Hebrew; they use Arabic to communicate with each other but they study Hebrew at school and use it in their everyday life – at work, shopping, visiting health-care institutions and places of entertainment, etc. The distribution of the immigrants demonstrated that despite the short time since their arrival in Israel, the percentage of both adapters and dualists was rather high (18.4% and 28.4% respectively). Yet, nearly half of the Russian immigrants (46.6%) reported a separatist tendency. The remaining small group reported that they did not consume any media whatsoever (ibid.: 131). The smaller percentage of dualists in comparison with the Arab interviewees is in accord with the fact that the newly arrived immigrants speak East-European languages, mainly Russian, and have more difficulties consuming Hebrew media. As hypothesized by the researchers, the data indicated that the choice of language correlated with the construction of social identity. Separatism, namely high consumption of media only in the language of the minority (either Arabic or Russian) correlated

 Rachel Weissbrod

with a strong minority identity, while high consumption of media in Hebrew (by the adapters) correlated with a strong Israeli identity. Dualism corresponded to what Hall (1993) called “a hybrid identity”. According to Hall, minority group members in modern societies “are the products of cultures of hybridity” and therefore “are obliged to inhabit at least two identities, to speak at least two cultural languages, to negotiate and ‘translate’ between them” (ibid.: 362). However, some of the findings of the above mentioned research were unexpected. Among Israeli Arabs, the dualists, and not the adapters, exhibited the strongest Israeli identity (Adoni, Caspi and Cohen 2006:€ 86). In other words, in contrast to what might have been expected, in the case of the Arabs, minority members who did not give up their own language and culture were more willing to accept the identity of the majority. Among the immigrants, separatists showed a weak Israeli identity, and the strongest Israeli identity was demonstrated by the adapters; but surprisingly, the Israeli identity of the dualists was very similar to that of the adapters. The fact that in the Arab group, the dualists, and not the adapters, exhibited the strongest Israeli identity, and the similarity between adapters and dualists in the immigrants’ group as regards their social identity, suggest that the construction of an Israeli identity does not require the exclusive use of Hebrew media. In fact, one can consume media in both Hebrew and one of the minority languages and yet possess a strong Israeli identity. Adoni, Caspi and Cohen (2002:€431; 2006:€87) regard the hybrid identity of dualist media consumers as fitting equipment for living in the Israeli multicultural society. Their view has been consolidated by Elias (2005). Elias investigated the absorption of immigrants from the former Soviet Union and found that “living” in two cultures rather than one (the Israeli) did not lead to segregation. On the contrary, it allowed for a better social integration. Consuming Israeli media in Russian, and even Russian media produced outside Israel, enabled them to gain up-to-date information about Israel, its political and cultural agendas, its myths and symbols, and to eventually find their place in “the national narrative”. Researchers in other countries, too, such as Kymlicka in Canada (Kymlicka 1995) and Gillespie in Britain (Gillespie 1995), found that retaining one’s connections with the culture of origin contributed to a successful integration into the new one. As mentioned, dualism is defined by Adoni, Caspi and Cohen as the high consumption of media in more than one language. However, it is possible to apply this term to the consumption of Hebrew media with the help of what might be called “intra-cultural” translation. Translation makes it possible for a member of the minority to access Hebrew media while reading or listening to translation in his or her native language. This applies especially to TV translation. When translation takes the form of subtitles one can actually listen to Hebrew and read a text in another language. Taking translation into consideration can thus lead to a broader view of the category of “dualism”. In light of the vital importance of dualism for developing a hybrid identity – which, as we are told, is so much needed for living in a multicultural society – it is worth asking if



Translation Studies and mass media research 

consuming translation, too, plays a role in constructing such an identity. However, answering this question requires further research of the issue.

6. Concluding remarks As suggested, the multilingualism of mass media in Israel manifests itself not only in the accessibility of foreign products (newspapers, TV channels etc.), and in the local production in languages other than Hebrew, but also in translation involving the various languages spoken by Israelis. Taking translation into consideration can contribute to mass media research – in Israel and possibly in other countries – whether one is interested in the impact of multilingualism on immigrants’ absorption, the relations between minority languages, or the role of English as an agent of globalization and Americanization. However, while insisting on the necessity of a Translation Studies perspective, one should take into account the possibility that Translation Studies, too, can benefit from closer relations with mass media research conducted in the framework of socio-linguistics and communication studies. After all, raising the question of what might be the role of translation in the construction of social identity has been triggered, in this paper, by research on communication. A better collaboration between researchers in the neighboring fields of socio-linguistics, communication and Translation Studies seems to be beneficial to all parties.

References Adoni, H., Caspi, D. and Cohen, A.A. 2002. “The consumer’s choice: Language, media consumption and hybrid identities of minorities.” Communications: European Journal of Communication Research 27: 411–436. — 2006. Media, Minorities and Hybrid Identities: The Arab and Russian Communities in Israel. Cresskill, NJ: Hampton Press. Caspi, D. and Limor, Y. 1999. The In/Outsiders: Mass Media in Israel. Cresskill, NJ: Hampton Press. Crystal, D. 1997. English as a Global Language. Cambridge: Cambridge University Press. Danan, M. 1991. “Dubbing as an expression of nationalism.” Meta 36(4): 606–614. Elias, N. 2005. Media Uses as Integration Strategy: The Case of the Immigrants from the FSU in Israel. Tel Aviv: Chaim Herzog Institute for Communication, Society and Politics. [In Hebrew.] Epstein, A.D. and Kheimets, N.G. 2006. “Between globalization and localization: The linguistic diversity of the Israeli mass-media.” Paper presented at the 5th Conference of the Israeli Association for the Study of Language and Society. Ra’anana: The Open University of Israel (June 4). Even-Zohar, I. 2005. “Who is afraid of the Hebrew culture?” Papers in Culture Research. On line at: http://www.tau.ac.il/~itamarez/ (accessed on November 5, 2007). Galily, L. 2004. “The empire of television in Russian.” Ha’aretz (September 26). [In Hebrew.] Gillespie, M. 1995. Television, Ethnicity and Cultural Change. London and New York: Routledge.

 Rachel Weissbrod Gomel, E. 2006. The Pilgrim Soul: Being a Russian in Israel. Tel Aviv: Kinneret, Zmora-Bitan. [In Hebrew.] Grigaravičiūtė, I. and Gottlieb, H. 1999. “Danish voices, Lithuanian voice-over: The mechanics of non-synchronous translation.” Perspectives 7(1): 41–80. Hall, S. 1991. “The local and the global: Globalization and ethnicity.” In Culture, Globalization and the World-System: Contemporary Conditions for the Representation of Identity, A. King (ed.), 19–39. London: Macmillan in association with the Department of Art and Art History, State University of New York at Binghamton. — 1993. “Culture, community, nation.” Cultural Studies 7(3): 349–363. Kaufmann, F. 2002. “La coprésence de l’original et de sa traduction.” Dans Identité, altérité, équivalence? La traduction comme relation (Hommage à Marianne Lederer), I. Fortunato (éd.), 323–338. Paris: Minard. Kymlicka, W. 1995. Multicultural Citizenship.€Oxford: Clarendon Press. Orero, P. 2004. “The pretended easiness of voice-over: Translation of TV interviews.” JosTrans 2: 76–96. Pinto, M. 2007. “On the intrinsic value of Arabic in Israel – Challenging Kymlicka on language rights.” Canadian Journal of Law and Jurisprudence XX(1): 143–172. Pollack, Y. 2007. “83% of the households in the Arab sector do not watch Israeli channels.” TheMarker (November 26). [In Hebrew.] Rebhun, U. and Waxman, C. 2000. “The ‘Americanization’ of Israel: A demographic, cultural and political evaluation.” Israel Studies 5(1): 65–91. Segev, T. 2002. Elvis in Jerusalem: Post-Zionism and the Americanization of Israel. New York: Metropolitan Books. Shavit, Z. 2002. “Principles of a Hebrew language law.” Ha’aretz (April 16). [In Hebrew.] Spolsky, B. and Shohamy, E. 1999. The Languages of Israel: Policy, Ideology and Practice. Clevedon: Multilingual Matters. Weissbrod, R. 2008. “Implications of Israeli multilingualism and multiculturalism for translation research.” In Beyond Descriptive Translation Studies: Investigations in Homage to Gideon Toury, A. Pym, D. Simeoni and M. Shlesinger (eds), 51–65. Amsterdam/Philadelphia: John Benjamins.

Register shifts in translations of popular fiction from English into Slovene Marija Zlatnar Moe

University of Ljubljana, Slovenia

Major shifts in the register occur when works of popular fiction are translated into Slovene. This paper analyzes parts of seven novels of various subgenres, and finds a general shift towards an unmarked formal register and a neutral standard variety of the language. Formal or archaic texts (e.g.€The Lord of the Rings) become less formal, colloquial-style texts (e.g.€Bridget Jones’s Diary) become more so. Concomitantly, distinctions as to class and status, or past and present are blurred. Neutrality of style makes the texts appear less interesting or, at worst, “badly written” to the reader. Keywords: translation, popular fiction, register shifts

1. Introduction Many contemporary global bestsellers, such as Bridget Jones’ Diary, Harry Potter, and The Da Vinci Code, have been translated into Slovene over the past years. However, they have rarely achieved the same immense popularity or even cult status in the Slovene culture at large as they have elsewhere.1 While some of the reasons are undoubtedly connected to the characteristics of the target culture and the tastes of Slovene readers, these are not the only reasons. I will show how major shifts in the register – especially in the level of formality – occur when works of popular fiction are translated into Slovene. There is a shift towards an unmarked, formal standard language, which neutralises the style of the books. In order to determine the norms operating in translation of popular fiction into Slovene, I have studied seven books of different genres, focusing on how those norms influenced the shifts in register on the way. 1. Though in some cases, they do gain popularity within their intended reader community – the translation of the Harry Potter series (with the exception of the sixth book), for example, is popular among young readers.

 Marija Zlatnar Moe

2. Popular fiction 2.1

Reading popular fiction

Popular fiction is a huge field filled with genres and subgenres. It is not to be confused with the field of literature, not even literature that gains popularity. The two fields operate in different ways, they are discussed in different terms, and the readers hold completely different expectations towards them (Gelder 2004:€ 14–20). While literature values originality and inspiration, popular fiction is praised for its adherence to a particular genre and the hard work the writers put into it. Literary novels are praised for their beauty of expression, the original way in which they address the fundamental questions of our existence, while a good example of popular fiction is one with a strong plot, believable characters, fast-paced action and a clear ending (or, alternatively, a clear hook for a sequel). The field of popular fiction is characterised by conformity to genres, and there is a great abundance of them, from romance (with subgenres such as historical romance – or even medieval or Victorian romance), to horror, thriller, fantasy, science fiction, “chick lit” and so on. Literature, on the other hand, is more concerned with individual works, limiting the notion of genre to the classic trinity of prose, poetry and drama. The reader knows exactly what she will find in for example a good horror story, and a good horror story provides it. If it fails to do so, the reader may reject it. The crime has to be resolved, evil vanquished, the heroine married to the man she wants, the nice girl must get the job/place/guy she has worked for. If this does not happen, the reader is not looking at an original popular novel, but at a bad one. Another characteristic of popular fiction is that there is a lot of it. Many writers (usually called writers, not authors or novelists) are extremely prolific – Stephen King has produced over 50 novels, Agatha Christie 90, Terry Pratchett around 40 only in the Discworld series, the late Barbara Cartland produced over 500 romance novels, and so on. Exceptions include J. R. R. Tolkien with only two novels (collections of different completed or uncompleted texts were published posthumously), or Helen Fielding with three to date. They often write series featuring the same main character (Lestat, Sherlock Holmes, Harry Potter, James Bond), or sometimes a rotating set of characters. The authors also sometimes interact with readers in a way which is not found in the field of literature. Popular writers often travel and present and sign their books, and many of them communicate with their fans via web pages, blogs, message boards etc. While the readers move freely from popular fiction to literature and back again, they approach the two fields in different ways. Typically, popular fiction is for pleasure and to escape from the real world while literature is for personal growth and enrichment. We expect to have fun reading a popular novel. We do not make this demand of a literary work. Sometimes it happens. But just as often, reading literature is demanding,



Register shifts in translations of popular fiction from English into Slovene 

difficult and depressing. This is how a (professional) Slovene reader expressed those different expectations: The ideal scene, then, looks like this: a comfortable sofa, warm socks on my feet, a cup of aromatic tea, coffee or hot chocolate, and in my lap – a book. What kind of book? Unlike the “typical” summer book, which ought to be easy to carry as well as to read, so that it isn’t too heavy to be carried to the beach in a bag, and so that it doesn’t burden the already warm head too heavily; an autumn book can be totally real and totally serious. Big. Huge. A novel, that masterly entangles the reader in the threads of an epic fresco. A story, developing slowly, winding into many different directions. Characters and worlds to be slowly discovered while we read. Autumn reading is delicious. It takes time. (Košir 2006:€16, my translation)

However, readers have different demands and expectations not only towards the two fields, but also towards their authors. Popular fiction can achieve enormous popularity, seldom achieved by literature (an exception that springs to mind are Jane Austen’s novels which are therefore classed in an intermediate category of “popular literature”). The characters and the books acquire a large following of dedicated fans that know absolutely everything about the particular novel/series/character/author, the novels get made into (often hugely successful) films and similar. “Fandom” (a word denoting every aspect of being a fan of some segment of popular culture) is characteristic of popular culture and consequently of popular fiction, and a very unusual phenomenon in the literary field (Jane Austen, again, has become an exception). In many cases, fandom is as much fun as the book that started it, with many events, competitions, speculations, questions to the writer, trashing the films based on the books, and so on. So popular fiction very often transcends the solitary pleasure of being alone with the hot tea, warm socks and a book, and acquires a social dimension that is not typical for literature, not even popular literature. Connected to fandom is the industry providing tie-in products such as Harry Potter pyjamas or Sherlock Holmes teacups. As it partakes in these social and commercial dimensions, The Lord of the Rings is treated here as a work of popular fiction because it behaves like one, whether or not its literary merits are those of serious literature.

2.2

Translating popular fiction

Just as the field of literature differs from the field of popular fiction, so does translating within those fields. For one, more popular fiction gets translated than is the case with literary works. Translators are thus able to make money by translating popular fiction. Sometimes, with the most popular texts (such as the Harry Potter series or The Lord of the Rings), some of the book’s fame also rubs off on the translator, who is given the opportunity to speak about his/her work in the popular press and television programmes. On the downside, deadlines can be very short, because publishers want to follow the trend, catch the release of the movie, or just do not wish to lose too many readers

 Marija Zlatnar Moe

to the original.€The knowledgeable fans of the book can be helpful, but on the other hand, they are often extremely protective of the object of their affection, and fiercely opposed to changes. It has also so far been quite impossible to gain official professional recognition, of the kind expressed through awards for outstanding translations, from the Slovene literary translator community by translating popular fiction. All the winners of the Slovene award for the best literary translation have translated texts belonging to the canon of world literature. This is a fate popular fiction translators share with the writers of popular fiction, who also very seldom get recognition from the literary award community. When translators do talk about their popular fiction translation, their attitude to the texts often differs considerably from their attitude towards the literary texts they work with. From their public statements, it seems that they do not deem popular fiction to be on a par with literature, and that they are a bit embarrassed to be associated with it. But what they say publicly is not the only indicator of the different approach towards translating popular literature. As Lefevere pointed out when he was talking about three different translations of Brecht into English, “Writers and their work are translated differently when they are considered ‘classics’, when their work is recognised as ‘cultural capital’, and when they are not” (Bassnett and Lefevere 1998:€109). This is also true of Slovene literary translators and popular fiction. The texts tend to become more stylistically neutral, or more target-oriented. One of the strategies to achieve this is changing the register.

3. Register in translation “Register matters,” wrote C.E. Landers on literary translation (Landers 2001:€61). As Lefevere (1992:€58) points out, “Except in grammar books or primers on style, language is never used in a vacuum: it is always used in a certain situation”. This is also true for popular fiction. Authors of such texts use different registers for characterisation, to describe relations between characters, to bring out the temporal and social setting of the story, or to mark emotional intensity or the lack of it. All this helps readers to get immersed in the story, something that is of great value to the readers of popular fiction. A brief look at readers’ reviews on the internet will show that a book is considered good if it provides an escape from reality. On the way from English into Slovene, however, the register tends to change. Possible reasons include translators’ inexperience, the disdainful attitude toward translation of popular fiction, short deadlines, or translators’ deeply rooted convictions about what good language is. Whatever the reason, the outcome is a register that is close to the neutral normative variant of Slovene, shorn of many idiosyncrasies of the English text, irregularities, hidden meanings and comments, implied relations between the characters etc. There is a shift in the formality level as well as in field specific varieties of language, and this shift takes place both when the register is perceived as too low for a book in



Register shifts in translations of popular fiction from English into Slovene 

Slovene and when it is perceived as too high. In either case, the register is lowered slightly below the neutral.€It is as if translators feel the need to do something, but are not quite able to figure out what. Thus, when in the source text kings and wise old wizards speak their words of wisdom with appropriate grandeur, in the translation the same kings and wizards talk in a lower colloquial way, often further marked with rural features. These shifts may not be visible to the typical target reader who does not access a popular text in its source form (unlike translation studies scholars and die-hard fans). But (s)he may read a book and wonder what the fuss was about, since the book turns out to be rather boring and awkwardly written. Sometimes, of course, popular fiction is boring and awkward, but the works of popular fiction surveyed here have clearly been helped by Slovene translators to become slightly more boring and awkward than the English source version.

4. Research material I have analyzed seven books by some of the most popular writers in the field, translated from English into Slovene. Several different genres have been included in order to ensure that the translation strategies really were at work in the whole field of popular fiction, not only in a particular genre or even in the work of one translator or in the translation of one book. The books included in the study are: AB Nick Hornby: About a Boy (translated by Mojca Krevel); first published in 1998, Slovene version: 2002; genre: lad lit2 TR Maeve Binchy: Tara Road (translated by Aleksandra Rekar); first published in 1998, Slovene version: 2005; genre: romance HP J. K. Rowling: Harry Potter and the Order of the Phoenix (translated by Jakob Kenda); first published in 2003, Slovene version: 2003; genre: fantasy for children CoM Terry Pratchett: The Colour of Magic (translated by Maja Novak); first published in 1983, Slovene version: 2004; genre: satirical fantasy HF Helen Fielding: Bridget Jones’s Diary (translated by Maja Novak); first published in 1996, Slovene version: 2001; genre: chick lit DVC Dan Brown: The Da Vinci Code (translated by Nataša Müller); first published in 2003, Slovene version 2006; genre: thriller ROTK R. R. Tolkien: The Lord of the Rings – The Return of the King (translated by Branko Gradišnik); first published in 1958, Slovene version: 2003; genre: fantasy. As is common with popular fiction, several of the books studied come from a series about the same topic, and in such cases I included one book in the series (this is the 2. Based on the categorisation at Amazon.com.

 Marija Zlatnar Moe

case with BJ, TR, and ROTK). In the case of Pratchett’s Discworld series, I chose the only one that was available at the time of analysis (CoM). In the case of the Harry Potter series, I excluded the sixth book from the draw, because of the extraordinary circumstances of its translation.3 All the books studied and/or their authors belong to the canon of their respective genres – or even shaped the genres they represent (for example ROTK and BJ). All the authors have had more or less successful films made based on their books. They have acquired a faithful following of fans willing to buy whatever they write. Most of them communicate with their fans through web pages (the obvious exception being J. R. R. Tolkien). There is a lot of fan fiction written based on their books (with a possible exception of Maeve Binchy). Last but not least, all of them have gained a certain degree of popularity and economic security from their books.

4.1

The translators

The Slovene translators of those books are a diverse group.€On the basis of bibliographies of their published work (COBISS data) and biographical information of the kind provided by publishers, the group consists of two experienced translators who also are published fiction writers in Slovene, and of four more or less experienced professional or semi-professional translators who also write and/or translate non-fiction. By the summer of 2007, the least experienced had translated five novels, the most experienced, 38 novels. In some cases they are experienced translators now, but were at an early stage of their career at the time of the books in question; for instance, one of the translations studied was only the fifth in the career of a translator who had by 2007 published 22 translated novels as well as several of her own. She is represented in this paper with two books, in very different genres, one at the beginning of her translating career, and the other when she had become an experienced translator.

4.2

Method

I concentrated on those parts of the books which are most diverse in terms of register – including as many different levels of formality and fields as possible within each individual book. These are the parts where there is much interaction between characters, most notably dialogues; where the characters differ in social position, role, age, or prestige; or where the style is extreme (e.g.€dialogue between teenagers). In sum, for 3. The publisher decided to replace the original translator with another one who then translated the book in record time. The publisher’s decision caused an interesting discussion in the media about translation and translators, and it also infuriated the fans of the series who complained to the British publisher. The direct result of the whole affair is that the final book was again translated by the original translator, who also retranslated the sixth book. Both books were published by another publishing house.



Register shifts in translations of popular fiction from English into Slovene 

each book, I have analyzed either an actual chapter if the chapter as a whole met these criteria, or suitable excerpts from different chapters totaling about a chapter’s length. The kinds of shift I focused on were: – increased formality – decreased formality – change of field – stylistic changes that affect the contents of the text – words and phrases which deviated markedly from target-language norms These are the strategies that directly influence the register level of a text. A text containing segments which deviate from the norms in the relevant language, for example, is generally considered badly written; the same goes for a text that becomes less coherent because of shifts of meaning. The occurrence of a shift was determined on the basis of typical usage of a certain word, phrase or expression in the source and target languages. In cases of archaisms, neologisms and the like, it was determined with the aid of dictionary definitions in both languages and/or frequency and typical occurrences of an item in Slovene corpora. When such aids were not available, as in cases of the author’s or translator’s personal neologisms, the register of an expression was determined on the basis of an analysis of morphemes, word-formation etc.

5. Findings 5.1

Increased formality

The largest number (416 shifts out of 1,287 that were found in all the categories together) of shifts are those that increase formality – this goes for the whole study as well as for each individual novel. These changes are most visible in dialogues, except in BJ, where the overall formality in the whole text is higher than in the source text. Other interesting examples are TR, which is even slightly more formal than the rather conservative source text; and AB, where the chapters written from the point of view of the child are generally slightly more formal than the source text, while the chapters narrated by the adult are more formal only within the dialogues. Apart from these three cases of an overall higher formality level, formality is increased in: – informal conversations, for example conversations among hobbits, teenagers, friends in a pub etc.; – speeches of a low-status character (servants, children – for example Marcus’s chapters in AB);

 Marija Zlatnar Moe

– dialogues between members of different social classes4; – teenage or children’s language, which is almost universally more formal in the Slovene versions; – informal writing, such as the whole of BJ, or the informal letters in HP. A higher level of formality is achieved in various ways. One way is through lexical choices: more formal synonyms, sometimes unusual words or phrases, or archaisms. Translators sometimes choose to leave out or embellish very informal or vulgar expressions. Another way to increase formality is a more formal syntax and word/clause order: on the one hand translators form sentences that are longer and more complex than the source texts, and on the other they shorten long informal sentences, especially in the case of polysyndeton; or they shift the word order within each individual clause (a clause can be more or less formal depending on where the participle or the verb is). Translators also sometimes choose to make incomplete sentences from the source text complete in the target text; this is especially visible in BJ, where the diary form enables the author to be quite elliptical.€On the level of morphology, a higher formality level can be achieved by choosing more formal suffixes and endings: joče instead of joka (se), suče instead of suka, etc. A direct consequence of this upwards shift is a neutralisation of the style, which can in many cases result in less interesting texts. Because the rules about the plot, characters, endings etc. are so strong, most of the fun of popular fiction is in how it is written. If the style is lost, the reader is left with a story that is almost a formula and characters that are little more than types. And there are instances when the reader, instead of enjoying his/her portion of escapism, becomes increasingly irritated by the book (s)he is reading. This can easily happen when the text differs very much from expected usage – a typical example would be bookish-sounding children and teenagers.

5.2

Decreased formality

Decreased formality is the third most common shift (215 out of 1287 registered shifts), after increased formality and shifts in meaning (though it is often also a side effect of meaning shifts). The majority of cases of decreased formality are found in ROTK. They occur when the source text is very solemn, or when the characters discuss deeper issues such as emotions, philosophy or ideology. In ROTK, the starkest difference in the level of formality is found in the speech of the king, who sounds much more kingly and epic in the source text than he does in the target text; another instance, interestingly, is when one hobbit tries to explain to another hobbit how he feels about all the noble and legendary beings he has been around. Similar cases can be found in DVC 4. In ROTK, formality increases when characters of higher status talk to lower status characters, but not when they talk among themselves. This is the case also in the CoM, in the dialogue between the patrician and the failed wizard, where the patrician’s informality in the source text not only signals a dialogue across social classes, but it is also a part of the characterization.



Register shifts in translations of popular fiction from English into Slovene 

(the quiz which enables the heroes to enter the eccentric professor’s house); and in CoM, where the whole cosmological introduction to the series is rather less formal than its English equivalent. Less often, there are cases where the formality suddenly falls – deeply – and contrasts with the immediate context, where the paragraph or the sentence, clause or phrase belong to the prevailing level of formality in the text, but a single word belongs to an entirely different level of formality. These examples are jarringly obvious to the reader because they are not in agreement with the rest of the sentence. Interestingly, it is usually the disturbing informal element that is closer to the source text than its more formal vicinity. Third, proper names are usually less formal in translation, especially in fantasy and children’s fiction. Sometimes names of people or places are formed similarly to colloquial Slovene words borrowed from German (which have been unacceptable in normative language for at least a century). The level of formality is further lowered if those names are written phonetically (Rajterska, Ajzengart). Slovene readers have been taught in school to avoid such words as signs of an informal manner if not simply bad style.5 Nevertheless, the whole of the Rohan culture in LOTR was named according to this strategy, as was the main villain in HP. Names generally contribute a great deal to the lowering of the overall register in many of the books. In HP, some names that are deliberately funny are attributed to persons whose role in the story is not supposed to be funny6 and which do not have especially funny names in the source text.7 A clear naming strategy is sometimes absent – as shown in the CoM where some names are translated, others are not, names that follow the same rule are translated differently (Twoflower becomes Flowerflower, but Ninereeds remains Ninereeds). Another point in case is the use of possessive forms of women’s family names, which is clearly considered politically incorrect, informal and a sign of a low style. Terminology specific to different universes can also cause downward shifts in formality level. In HP, for example, all that is scary and sinister for some reason ends up with a Slovene name that is just not scary enough. For instance dementors become morakvarji in Slovene, a word with no clear meaning and associations ranging from something humid to a bad dream. In CoM the terminology is made very complicated and long without any clear strategy – some basic geographical terms of Discworld are given in a descriptive way and do not function as terms at all. In LOTR rangers, who are basically an army unit, are translated by a pejorative neologism for “wanderers.”

5. Relevant sources including style manuals and school textbooks are cited in Zlatnar Moe (2005). 6. Severus Snape is called Robaws Raws in Slovene, which we could roughly retranslate as “Brute Brawl”; He turns out to be the great tragic hero of the saga. Neville Longbottom (Velerit, retranslated “Greatarse”) also turns out to be a hero. 7.

For more on the names in the HP translation, see Žižek (2003).

 Marija Zlatnar Moe

Sometimes interjections and metaphorical language are drawn from a field that is alien to the setting of the book. In BJ, for example, the translator repeatedly uses language drawn from old Slovene folk tales, proverbs and idioms in the diary of a thirtysomething Londoner. In archaising texts such as ROTK, formality is decreased where modernisms are used, and on the morphological level, where in very formal situations the less formal suffixes or endings are used. The level of formality is also often lowered as a side effect of other types of shifts I have studied, such as shifts of field and meaning, and by individual stylistic decisions such as the omission of the more formal type of polysyndeton.

5.3

Shifts of field

Shifts of field are less numerous than formality shifts (31 out of 1287 registered shifts), but they move in the same direction: towards the neutralisation of style. Thus biblical, legal, teenagers’, children’s, journalistic, and ritual English are all translated into the same neutral formal Slovene. One interesting example is from HP, where teenage language changes into children’s language. In other works studied too, the teenage variant seems to cause the most problems to the translators. They generally decide to neutralise it, but sometimes throw in individual colourful expressions that stand jarringly out from the neutrality of the rest.

5.4

Other shifts that influence the register level

5.4.1 Shifts of meaning Worryingly enough, meaning shifts are the second largest group of shifts in my study (293 out of 1287 registered shifts), and they occur more often with translators who are supposed to be more experienced. Direct causes for those shifts, such as are evident from the texts, are mostly interference between English and Slovene, cases of word-for word translation, or overcautious attempts to avoid interference. For example, in HP, Harry is urgently advised not to leave his uncle’s house (the building itself) while in Slovene this changes to “home”, which is often the more usual way of translating the English term house. But this does not work, because it is only the house that is protected; the garden in front of it, for example, is dangerous. Sometimes shifts in meaning are caused by clear reading mistakes (in TR the translator at one point obviously read “two houses” as “two hours”). In some cases they appear to result from misunderstanding of either the source text or the universe in which the story is happening: for instance, in ROTK the word prestol (throne) is used for the seat of the Steward of Gondor, who most definitely does not use the throne, as the book explains at length. In the cases I studied, shifts of meaning often influence the formality level, as the Slovene translator chooses a more formal close synonym. Exceptions to this rule include two cases of shifts in meaning that led to changes in



Register shifts in translations of popular fiction from English into Slovene 

characterization, but that did not influence the register of the text (both in DVC), as well as a few shifts that caused a character’s speech to be less coherent, though not necessarily less formal. 5.4.2 Stylistic shifts Stylistic shifts fall into three categories. First, there is a group of shifts influenced by the norms of “good writing” in Slovene. These are most typical of the less experienced translators in my study. The most common among those shifts is compression – translators frequently decide to say the same thing with less words than the original author. One typical example is found in TR, where one of the characters talks about “the man who is married to my sister”, thereby distancing himself as much as possible from a very unpleasant character, who, in the Slovene translation, becomes “my brother-inlaw”. Another very common change is omission of repeated items, according to the Slovene norm, which is quite different from the English norm. There is one contrary example (in the work of a more experienced translator) where the style is much less economic and one- or two-word expressions are replaced by whole sentences or even paragraphs for no apparent reason (in CoM). Second, there are changes that prettify the nastier parts of the source text. In HP, dangerous things appear less dangerous; in BJ, many remarks are less acerbic than in the English version, Sophie Neveau in DVC is softer than her American equivalent, etc. Thus the style is neutralised, and the whole text is a little bit more formal.€Similarly, vulgarisms tend to be embellished or left out, but then, suddenly there is a burst of vulgarity, leaving the reader wondering what possessed the character suddenly to say such a thing. Also, vulgarisms fall victim to the Slovene reluctance to repeat: the same four-letter vulgarism is translated differently each time in the same scene (the quarrel between Will and Marcus’s mother in AB), which reduces the intensity of the situation and contributes to a higher formality level. Third, there are text segments not in line with target-language norms. This category is rather small, and is more apparent in the case of more experienced translators. In my study, more than half of the examples come from one and the same book, and they are also characteristic of some other of this experienced translator’s many works. Most cases seem to be the result of interference and word-for-word translations, the rest are awkward collocations and neologisms. As a rule, such occurrences in the novels studied decrease the formality level, either because they are not in line with the norms of good writing in the target language,8 or because they cause the text to be accidentally funny where it is not supposed to be.

8. I refer to norms of good writing as determined in style manuals and school textbooks, since the dialogues typically use the normative version of written Slovene, and do not conform to the norms of spoken discourse, which is why the non-prescriptive solutions are jarring.

 Marija Zlatnar Moe

6. Conclusion My survey of seven translations shows that shifts in register occur frequently in translating popular fiction from English into Slovene. As a rule, they shift the style from the extremes towards neutrality, which changes it considerably. The most frequent type of change is increased formality; it is followed by lexical changes (some contributing to increased formality and some to the opposite), decreased formality, and changes of field-specific language into a more general standard variant. This often makes the texts less interesting to read, sometimes also influencing the contents of the novels, making them duller than the original.€On the basis of what translators publicly say about their work, one reason for these shifts could be their negative attitude towards popular fiction. Others are evident from the texts – for instance opting for standard target style norms instead of the authors’ individual stylistic choices, and attempting to “correct” the style of low-status characters. Some translators move away from either source- or target-language standard norms and adopt an idiosyncratic style – possibly, again, because they feel that they can do in popular fiction what is not expected in “real literature”. This probably explains why the reception of world best-sellers by professional readers as well as by the general public is often less enthusiastic in Slovenia than in many other markets.

References Bassnett, S. and Lefevere, A. 1998. Constructing Cultures: Essays on Literary Translation. [Topics in Translation 11]. Clevendon: Multilingual Matters. Gelder, K. 2004. Popular Fiction: The Logic and Practices of a Literary Field. London and New York: Routledge. Košir, T. 2006: “Branje po jesensko ali debele knjige mojega življenja.” Premiera 132: 16. Landers C.E. 2001. Literary Translation: A Practical Guide [Topics in Translation 22]. Clevendon: Multilingual Matters. Lefevere, A. 1992. Translating Literature: Practice and Theory in a Comparative Literature Context. New York: The Modern Language Association of America. Lefevere, A. 1998. “Acculturating Bertolt Brecht.” In Constructing Cultures: Essays on Literary Translation, S. Bassnett, and A. Lefevere (eds), 109–122 [Topics in Translation 11]. Clevendon: Multilingual Matters. Zlatnar Moe, M. 2005. “(Ne)ponavljanje blizu skupaj stoječih prvin v književnih prevodih.” Vestnik 39, no. 1–2: 187–204. Ljubljana: Društvo za tuje jezike in književnosti. Žižek, U. 2003. Prevajalske strategije prevajanja lastnih imen v slovenskem prevodu romana Harry Potter – kamen modrosti pisateljice J. K. Rowling. Diploma thesis, University of Ljubljana.

Getting the ACCENT right in Translation Studies Ian A. Williams

University of Cantabria, Spain

This paper describes a methodological approach to Translation Studies, Application of Corpus-based Contrastive Evaluation for Natural Translation (ACCENT), which is particularly suitable for non-literary texts. The approach, which employs corpus techniques to create linguistic profiles based on collocation, colligation, semantic preference and semantic prosody, is illustrated in a study of first-person verb use in the Results sections of English and Spanish biomedical research articles. In the texts containing this feature, the frequency was significantly higher in Spanish than in English (92 versus 30; P = 0.002). In English, the main associated rhetorical function was to express a non-standard procedure or a personal methodological choice, whereas the main function in Spanish was to state a result, mainly through a perception verb. Comparison of English source texts with their Spanish target texts revealed direct transfer of the source language profile rather than adaptation to the target language profile. Strategic transfer options based on the linguistic profiles proved capable of correcting the differences observed between the corpora. It is concluded that Translation Studies using corpus and contrastive techniques not only yield valuable empirical data but also provide insights into discourse patterns that may go unnoticed in routine translation practice. Keywords: corpus studies, contrastive studies, discourse patterns, scientific writing, rhetorical features

1. Introduction In the last 20 years, developments in computer technology have contributed to a spectacular increase in corpus and contrastive studies. These methodologies have a great deal to offer Translation Studies both through empirical descriptive studies and as a means of making and testing hypotheses on language in context. Enhanced computer potential has also provided trainee and practising translators with valuable tools through

 Ian A. Williams

translation memories and the ability to align source (SL) and target language (TL) texts in parallel corpora. It is thus possible that theory and practice could follow separate paths, whereby theorists could choose to carry out contrastive studies on comparable corpora to feed their speculations on cultural divergence, whereas practitioners would prefer parallel corpus data, possibly leading to the repetition or perpetuation of translation behaviour not in keeping with that found in comparable naturally occurring data. Just as it is through accents that we often identify non-native speakers, it is through that quality “It doesn’t ring true” that we recognise translated texts, a quality that has been associated with such phenomena as the third language (Duff 1981) or code (Frawley 1984), translationese (Gellerstam 1986), and interference and negative transfer (Toury 1995). Here ACCENT is used as mnemonic to represent a methodological approach to Translation Studies that is especially suitable for non-literary texts, such as biomedical research articles (RAs). In this context, the aim is to produce as natural a TL text as possible through a “covert translation” (House 1997), one that could pass as a native language (NL) text. ACCENT refers to the Application of Corpus-based Contrastive Evaluation for Natural Translation. Application because, for non-literary texts, Translation Studies needs to go beyond description and provide directives in the guise of laws (Toury 1995) or guidelines on translation options in specific contexts (Williams 2004; 2005; 2007; 2008). The ideal corpus design includes both comparable and parallel components as in the English-Norwegian Parallel Corpus (Johansson 1998:€8). This allows contrasts from different viewpoints: interlinguistic analysis confronts similar text types in two languages; intralinguistic analysis contrasts TL with NL texts; and comparison of SL and TL texts provides insights into actual translation practice. Evaluation is both quantitative through statistical analysis and qualitative through corpus techniques such as extended units of meaning involving collocation, colligation, semantic preference and semantic prosody (Sinclair 1991; Tognini-Bonelli 2001), reflecting lexical, grammatical, semantic and discourse-functional patterns incorporated into a linguistic profile. The quantitative data and profiles serve to draw the attention of translators, revisers and translators’ editors to problem areas regarding certain lexical items and syntactic structures and thus contribute to the production of a more natural translated text. This paper illustrates the approach with a study of first-person verb use in the Results sections of 192 biomedical RAs comprising a specialised corpus with parallel and comparable components.

2. Material and methods 2.1

The corpus

The study was performed on a computerised corpus composed of 192 RAs (approximately 500,000 words) with the typical Introduction, Methods, Results and Discussion (IMRAD) format, and divided into three subcorpora: 64 English SL texts



Getting the ACCENT right in Translation Studies 

(157,650 words); their 64 Spanish TL texts (185,000); and a comparable subcorpus of 64 Spanish NL texts (140,250) (for details, see Williams 2006). This study used only the Results sections, comprising 45,350; 52,930; and 33,930 words, respectively.

2.2

Rhetorical categories in the results section

Within the IMRAD framework, the communicative function of the Results section is fundamentally to present the evidence generated by research. In fact, 74% of all statements correspond to what is termed a Statement of Results (SOR) (Williams 1999). These SORs present numerical data, compare groups, express the changes observed due to intervention, or describe statistical associations, effects, and trends. However, one of every four statements has a different communicative function. About 8% are Procedures explaining how or why the data were generated, and almost always preceding a related SOR. Another 5% are metatextual Pointers, which refer to other parts of the article, often a table, graph or illustration with the data to be described or commented on. The remaining 13% are classed as Comments – statements more typical of the Discussion – and subdivided into Citations (3%), which compare new results with published findings, Explanations (5%), which account for unexpected outcomes or inconsistencies between results, and Implications (5%), which indicate the immediate significance of the data.

2.3

Methods and analyses

All first-person verbs in the Results section were located with the concordancing program of WordSmith Tools (Scott 1998). Quantitative analyses compared the subcorpora for number of texts with and without first-person verbs and, when this feature was present, for frequency of occurrence and in relation to rhetorical category. Qualitative contextual analyses created linguistic profiles consisting of collocation (lexical verb and nouns in the direct object (DO) position); colligation (tense and associated syntactic features such as adjuncts and subclauses); semantic preference (verb types: procedural, perception, possessive, desiderative); and semantic prosody (rhetorical function and associated nuances). Categorical variables were assessed with the χ² test, with Yates’ correction for 2 × 2 tables. First-person frequencies were analysed with a binomial distribution test. A value of ≤ 0.05 was considered statistically significant.

3. Results 3.1

Quantitative analysis

The analysis of texts with and without first-person forms showed a clear bimodal distribution in the English and Spanish subcorpora (Table 1), with more texts containing

 Ian A. Williams

Table 1.╇ Comparison of texts with and without first-person forms in the English and Spanish subcorpora Subcorpus

Texts with no 1st Person

Texts with 1st Person

χ² Value

P Value

47 36

17 28

3.427

0.064

English Spanish

Table 2.╇ Distribution of first-person forms according to rhetorical function in the English and Spanish subcorpora Subcorpus English Spanish

Procedures

SORs

Comment/ Metatext

χ² Value

P Value

17 16

â•⁄ 6 64

â•⁄ 7 12

24.128

< 0.001

no first-person verbs in the Results section. The higher number of Spanish texts including this feature (28 versus 17) did not quite reach statistical significance. However, in the texts containing the first-person feature, the frequency was significantly higher in the Spanish than in the English subcorpus (92 versus 30; P = 0.002 by the binomial distribution test). Thus, in these Spanish texts a first-person verb appeared once every 7 sentences compared to once every 20 sentences in the English texts. For the analysis of first-person tokens according to rhetorical category (Table 2), the metatextual Pointer and the Comment categories were combined because of the small numbers. The highly significant difference between the subcorpora is due almost entirely to the SOR function. The quantitative results show that Spanish authors not only tend to include more first-person forms in their Results sections, but they use them for a different rhetorical purpose.

3.2

Qualitative analysis: Linguistic profiles

In the linguistic profile for first-person use in the English texts (Table 3), the most frequent lexical verb collocate was the perception verb find (3 tokens), with five verbs appearing twice. The only two noun collocates repeated as DO were cases and hypothesis. The outstanding associated grammatical feature was past tense (25 of 30 instances). Syntactically, no clear preference for one type of fronted adjunct was apparent, and of the 8 adverbial subclauses, 4 came before a first-person verb in the main clause and 4 subclauses contained the first-person form. The outstanding semantic preference was for a procedural verb (17 tokens). As seen in the quantitative study, the main related



Getting the ACCENT right in Translation Studies 

Table 3.╇ Linguistic profile of first-person use in the English texts Collocation Verb Find Ask Assess Collect Perform Study

Colligation

Semantic Preference

Past tense 25 Verb type 3 Procedural 2 2 Syntax 2 Fronted 2 adjunct â•⁄ 6 2 Adverb â•⁄ 8 Direct Object subclause Cases 2 – preceding (4) Hypothesis 2 – included in (4)

Semantic Prosody

17

Functions and Features To state a procedure 17 – not routine or conventional – linked to rationale (6) – based on personal choice – involving novelty – due to ad hoc decision To state a result â•⁄ 6 To make a personal comment â•⁄ 5 To refer to the current text â•⁄ 2

communicative function of first-person forms is to state a procedure. However, the contextual analysis allowed qualification of this function: these forms are used not with routine or conventional procedures, but with those justified by logical reasoning, as in Example (1), or based on personal choice or decision (2), or involving some kind of novel approach, as in Example (3), where the author is the first to investigate these cul-de-sacs.



(1) Similar analyses by type of treatment (Table VI) were limited by small numbers of person years for patients receiving multiple courses of treatment, and we therefore aggregated some tumour groups for this analysis. (2) After 1 year, the patient had an undercorrection of 1 D, and we decided to reoperate. (3) As described in the ‘Materials and Methods’ section, we estimated the area of cul-de-sacs from the linear measurements in the following manner.

One further possibility is that researchers take an ad hoc decision to perform a procedure based on some element not foreseen in the study design. In (4), the experiment was repeated with larger doses because of the poor results obtained with the small dose:

(4) Because the error in delivering 0.5 U was so large, we asked four nurses to dispense 1.0 U (10 µL) using the 0.3-mL syringe.

Thus, the main characteristic of first-person use in the English texts is the expression of non-standard methods and of personal choices and decisions. The authors assume responsibility for their actions, and so leave readers free to decide whether to accept or reject the validity of the results, in what Hyland (1998:€181) has identified as a readeroriented hedging strategy.

 Ian A. Williams

Table 4.╇ Linguistic profile of first-person use in the Spanish texts Collocation Verb Encontrar Observar Hallar Apreciar Ver Aplicar Comparar Obtener Realizar Registrar Tener Descubrir Destacar Detectar Querer Seguir Direct Object Diferencias Casos Caso Asociación Cambios histológicos Pacientes

30 â•⁄ 9 â•⁄ 8 â•⁄ 5 â•⁄ 5 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 3 â•⁄ 2 â•⁄ 2

Colligation

Semantic Preference

Tense Past Present Pres. perfect

Verb Type 50 Perception 23 Procedural 16 Possessive

Syntax Fronted adjunct – spatial (13) – topic (9) – temporal (6)

Other Aspects â•⁄ 31 Capacity â•⁄ 8 (poder)

Adverb subclause – preceding (10) – included in (9) Projected nominal clause

Semantic Prosody 55 16 â•⁄ 6

Functions and Features To state a result 64 To state a procedure 16 – linked to SOR (9) To refer to the â•⁄ 7 current text To make a personal â•⁄ 5 comment

19

11

16 â•⁄ 6 â•⁄ 3 â•⁄ 2 â•⁄ 2 â•⁄ 2

In the linguistic profile for the Spanish texts (Table 4), the outstanding verb collocate was encontrar ‘find’ (30 of 92 tokens), followed by another four perception verbs: observar ‘observe’, hallar ‘find’, apreciar ‘notice’, and ver ‘see’. No procedural verb had more than 3 tokens. The Spanish texts also provide more DO collocates, the main one being diferencias ‘differences’, usually in the statistical sense. Three nouns refer to patients or cases, asociación ‘association’ also appeared in statistical contexts, while cambios histológicos ‘histological changes’ was related to perception verbs. For colligation, the past tense also predominated (50 of 92 instances), but a good many verbs appeared in the present (23 tokens) and the present perfect (16). Syntactically, Spanish first-person forms were associated with a fronted adjunct in 31 contexts, with a preference for



Getting the ACCENT right in Translation Studies 

spatial adjuncts, often including caso/s ‘case/s’ or paciente/s ‘patient/s’, but SORs also had topic-introducing adjuncts, as shown in (5): (5) En cuanto a los tumores originados en el hueso, encontramos tres osteomas que produjeron exoftalmos y pérdida de visión. ‘With regard to tumours arising in bone, we found three osteomas that caused exophthalmos and loss of vision.’

First-person forms had related adverbial clauses in 19 contexts, 10 appearing before the main first-person verb while the other 9 subclauses included the first person. One syntactic feature that distinguished the Spanish texts was a projected clause introduced by a perception verb: (6) Si comparamos estos datos con el global de partos, observamos que entre ambos grupos no existen diferencias estadísticamente significativas (P>0,05). ‘If we compare these figures with those for the total number of deliveries, we can see that there are no statistically significant differences between the two groups (P>0.05).’

As with the individual verb collocates, the clear semantic preference for Spanish firstperson forms in this section is for perception verbs (55 tokens), followed at a distance by procedurals and possessives. Capacity (poder ‘can, be able to’) was an associated feature in 8 contexts. In contrast to the English texts, the main communicative function is the SOR. The contextual study revealed that there is often a blurring of the semantics of the firstperson form, sometimes including and sometimes excluding the readers. The inclusive forms, as in Example (6), invite the reader as a peer to participate in the logical presentation and interpretation of the data (Hyland 1998:€183). In addition, in the Spanish texts first-person Procedures were mostly linked to the SOR, as in Example (6), in a Procedure + SOR pattern. Finally, metatextual reference was also more prominent in these texts, often linked to the explicit drawing of the reader into the deductive process: (7) En relación con las complicaciones, podemos dividirlas en precoces y tardías. ‘In relation to complications, we can divide them into early and late.’

3.3

Translation behaviour

Having established the reference values and linguistic profiles for first-person use in the English and Spanish texts, we now look at what the translators did when transferring the English Results sections into Spanish. Of the 30 English SL instances, 25 were translated as Spanish first-person forms. Those rendered by other means were 3 Procedures, expressed in the reflexive Se passive; a metatextual statement designating the abbreviations identifying different pathological groups, also in the Se passive; and an Explanation, in which first-person reference was expressed by the possessive

 Ian A. Williams

adjective. In contrast, in 3 contexts a first-person form was added to the TL texts. One was a non-finite metatextual reference expressed in a full finite relative: ‘the two PDS specimens mentioned earlier’ los dos especímenes de SDP que mencionamos antes (‘that we mentioned earlier’). The second was in a Citation, also involving a non-finite to finite change: ‘as reported’ Como ya describimos (‘As we previously described’). The third context was a comment with a switch from an impersonal infinitive to a firstperson form that aptly invites the reader to share in the deductive process:

(8) Therefore, it is appropriate to ask how representative the study population is compared to the background or reference populations in the three study sites. Por tanto, podemos preguntarnos si la población del estudio es representativa de las poblaciones tomadas como referencia en los tres lugares de estudio. This gives a total of 28 tokens (14 Procedures, 6 SORs, 5 Metatext, 3 Comments), and a profile that conforms to the English SL rather than the Spanish TL profile.

4. Application 4.1

Strategies for Spanish-English translation based on the English profile

As seen in the quantitative analysis, the discrepancy between Spanish and English use was found in the SOR category, since the figures for the other categories were roughly the same. However, in formulating corrective strategies, we must also consider the linguistic profiles, especially the associated implications of first-person verbs in Procedures. The recommendations for each of the communicative categories can be formulated as follows: 1. Metatextual and Comment categories – first-person forms are candidates for direct transfer unless local factors indicate otherwise. These will entail such concepts as brevity and clarity, with a preference for shorter non-finite forms instead of full finites. (9) En lo que respecta al serotipo C, ya hemos dicho que el 100% fueron varones de seis años. ‘For serotype C, as mentioned above, 100% were 6-year-old males.’

2. Procedures – first-person forms are candidates for direct transfer provided the context includes non-standard techniques, rationale, personal choices or ad hoc decisions. Otherwise, they are better expressed by impersonal means such as the passive. 3. SORs – first-person forms are candidates for transfer by other means unless specifically indicated for first-person expression. Justification of first-person use would be to achieve better information flow, as can occur in the opening SOR in the Results section. In (10), the version (a) is far more satisfactory than version (b) with its awkward sentence-final passive verb, which is less informative than the long grammatical subject and thus contravenes both the information (end-focus) and end-weight principles (Biber et al. 1999: 896–8):



Getting the ACCENT right in Translation Studies 

(10) Durante el período de tiempo estudiado hemos encontrado 17 pacientes con rotura diafragmática, con edades comprendidas entre los 19 y 61 años (edad media 40,3 años). (a) ‘In the study period we found diaphragmatic rupture in 17 patients aged between 19 and 61 years (mean 40.3 years).’ (b) ‘In the study period 17 patients with diaphragmatic rupture aged between 19 and 61 years (mean 40.3 years) were found.’ The other means to express SORs impersonally include the following. 1. Spanish sequences with an adjunct containing paciente/s, caso/s or some similar noun plus a perception verb plus a DO expressing a characteristic can be transferred by the typical English S + V + DO sequence with patient/s or case/s as subject of have or show: (11) En 2 enfermos encontramos antecedentes de radioterapia cervical, 10 y 17 años antes de la intervención. ‘Two patients had a history of cervical radiotherapy performed 10 and 17 years before surgery.’ 2. Similarly, sequences with an adjunct or a non-finite subclause containing a technique or analytical method can be transferred by active forms of show, demonstrate or reveal with the technique expressed as an abstract rhetor: (12) Realizando una valoración similar respecto al síndrome de CID, apreciamos que la capacidad discriminativa superior se consiguió en los intervalos 0–12 y 0–24 horas (Youden: 0,93, SE: 0,04) utilizando el sistema de PRISM. ‘A similar assessment for DIC syndrome showed that the highest discriminatory capacity was obtained for the intervals 0–12 and 0–24 hours using the PRISM system (Youden: 0.93, SEM: 0.04).’ 3. Statistical SORs, especially those with diferencias, can be rendered by existential There + be or by a passive perception verb: (13) No hallamos diferencias entre niveles de los diferentes COC respecto a la edad gestacional y el tono general de los RN. ‘(There were) no differences (were found) in the levels of the various organochlorine contaminants for gestational age and general tone of the newborn.’ 4. Spanish possessives tener ‘have’ and obtener ‘obtain’ can be expressed through the possessive adjective our with stative be. (14) Según esto, obtenemos buenos resultados en 35 casos, es decir, el 81,39%. ‘According to these criteria, our results were good in 35 cases (81.39%).’ 5. Some instances where the perception verb takes a clausal object simply mark the status of the following proposition as an SOR, and can be considered redundant:

 Ian A. Williams

(15) En el caso de los enfermos parcheados con la batería específica y con positividades, observamos que el EACP supone el 70,5% de los casos estudiados. ‘For patients with positive results in the specific battery patch test, ∆ allergic occupational contact eczema accounted for 70.5% of the cases studied.’ When the strategic options were applied to the Spanish contexts (Table 5), the main alternatives were passives, active have and show (Williams 2008), and the existential structure. Eight Spanish first-person forms were considered redundant according to the English profile, and minor options were structures involving stative verbs, and other active verbs (e.g.€occur and include). Of the 92 Spanish tokens, 20 were deemed appropriate for direct transfer into English. However, a further three related verbs were added, as in (16), where the infinitive becomes a full finite: Table 5.╇ Application of Spanish-English translation options to Spanish first-person forms Transfer Procedure

No

No. by Rhetorical Category

First person

23 *

Procedure SOR Comment Metatext

10 â•⁄ 5 â•⁄ 4 â•⁄ 4

Passive

22

Procedure SOR Metatext

â•⁄ 6 15 â•⁄ 1

Active have/ show

21

SOR Metatext

20 â•⁄ 1

Existential

15

SOR

15

Redundant

â•⁄ 8

Procedure SOR Metatext

â•⁄ 1 â•⁄ 6 â•⁄ 1

Stative

â•⁄ 4

Procedure SOR Comment

â•⁄ 1 â•⁄ 2 â•⁄ 1

Other active verb

â•⁄ 2

SOR

â•⁄ 2

* Three first-person forms were added in the application: 2 Procedures and 1 Statement of Results (SOR)



Getting the ACCENT right in Translation Studies 

(16) Al comparar los tumores mayores y menores de 1 cm (tabla 2), encontramos diferencias significativas en la variedad anatomopatológica predominante, invasión extratiroidea, presencia de adenopatías, mortalidad y clasificación TNM. ‘When we compared tumours larger and smaller than 1 cm (Table 2), we found significant differences for predominant pathological type, extrathyroidal extension, lymph node involvement, mortality and TNM classification.’ With regard to rhetorical function, the distribution of the 23 first-person forms shown in Table€5 corresponds to the English profile (see Table€2), but is rather light on Procedures, since only 8 of the 16 Spanish procedural contexts included explicit reference to rationale, novelty, choice, or personal decision. However, as this study was limited to those Spanish sentences containing an overt first-person form, there may be other impersonal contexts with features that would make them suitable for first-person expression in English, as in (17), where the authors decide to perform a subgroup analysis based on the patients’ characteristics: (17) Se investigó, también, si existía esta correlación entre los enfermos en ritmo sinusal, sin que se encontrara una significación estadística. ‘We also investigated whether this correlation existed in patients in sinus rhythm, but it did not reach statistical significance.’ Here the reported question also makes the first-person the most appropriate option. If there were a few other similar cases, the above profile would come very close to that seen in the English texts.

4.2

Strategies for English-Spanish translation based on the Spanish profile

As seen in Section€ 3.3., most English first-person forms can be transferred directly into Spanish. However, the quantitative analysis and the linguistic profile show that this is not sufficient to bring the TL texts into line with the Spanish discourse style due to the lack of first-person SORs. In view of the bimodal distribution observed in the corpus (Table 1), we can predict that 44% of Spanish TL texts will require first-person forms, whereas only 27% of English SL texts will include them. In addition, the expected first-person frequency will be three times that found in the English texts, with practically all the required additions being SORs. Therefore, for the TL texts to conform to the Spanish profile and discourse style, it is first necessary to identify English SL texts suitable for amplification or inclusion of first person verbs. Candidate texts for amplification and inclusion of first-person use will be (1) texts already containing first-person forms in the Results section; (2) texts with the possessive our in the Results section; (3) texts with a relatively high frequency of first-person forms in the Discussion.

 Ian A. Williams

The transfer options required are the inverse procedures to those in the above section. 1. English SVO structures with patient/s as subject can be transformed into the Adjunct + perception verb + DO pattern with paciente/s, enfermo/s, caso/s etc. as headnoun in the adjunct. 2. Abstract rhetors representing techniques and investigational procedures as grammatical subjects of show, demonstrate and reveal can be represented in Spanish adjuncts or non-finite clauses together with a first-person perception verb (Williams 2008). 3. Statistical results expressed in passive or existential structures can be personalised, especially with the verbs encontrar and hallar ‘find’. 4. SORs involving stative and existential verbs can sometimes be expressed through Spanish possessives tener ‘have’ and obtener ‘obtain’. 5. In complex clauses with the Procedure + SOR pattern, a first-person reporting frame (observamos que ‘we observed that’) can introduce the SOR. The application of these options to an SL Results section with two first-person Procedures is shown in Table€6. The last two sentences have been omitted because they were not considered suitable for further first-person use. The published Spanish translation (not shown) only transferred the two first-person Procedures from the English SL text, and the remaining statements including all the SORs were rendered impersonally. The proposed version, like the published translation, opens with an impersonal statement establishing two groups of patients, followed by a standard metatextual Pointer. The first shift is seen in sentence 3 (S3), with the introduction of the inclusive first-person in a subclause como podemos apreciar ‘as we can see’, which invites the peer reader to actively participate in the interpretative process. The immediacy of this process is also reflected by a switch to the present time sequence in the SORs that follow, with original S4 appended to S3 in a subclause. In S5, the collaborative process continues as the reader’s attention is directed to the figure: Estos resultados los podemos ver reflejados también– ‘These results [them] we can see reflected also…’ S6 illustrates a passive-to-active transposition to express a statistical result. After another Pointer (S7) comes the first-person Procedure, as in the published translation. S9 contains a Procedure + SOR pattern in which the former is expressed by an infinitive and the latter is presented by the reporting frame observamos que ‘we observe that’. This reestablishes the collaborative interpretative viewpoint after the methodological flashback, in which the first-person refers only to the authors. S10 is another Procedure + SOR sequence in which the English finite passive has been made an infinitive related to the perception verb comprobamos que ‘we found that’. The sequence S11-S12 constitutes a third Procedure + SOR pattern, this time extending over two sentences because of the information density. The Procedure is as in the published translation, and the SOR reporting frame is now encontramos que ‘we found that’.



Getting the ACCENT right in Translation Studies 

Table 6.╇ Application of English-Spanish translation options to an English sample text 1. A total of 572 patients had had atypical symptoms (either alone or combined with typical symptoms), whereas 2555 patients had had typical symptoms alone.

Un total de 572 enfermos habían tenido síntomas atípicos (solos o asociados a síntomas típicos), mientras que 2.555 habían presentado síntomas típicos únicamente.

2. The number and type of outcome events in all patients are shown in Table 1.

En la tabla 1 se recogen el número y el tipo de episodios de valoración final en todos los casos.

3

Patients with atypical symptoms had a higher risk of a major cardiac event and a lower risk of stroke than did patients with typical attacks.

4

There was no significant difference in the overall risk of major vascular events between the groups.

Como podemos apreciar, los enfermos con síntomas atípicos tienen un riesgo de episodios cardíacos graves superior y un riesgo de ictus inferior al de los enfermos con episodios típicos, sin que existan diferencias significativas entre ambos grupos en el riesgo global de episodios vasculares importantes.

5

These findings are also illustrated by the Kaplan-Meier survival curves (figure).

Estos resultados los podemos ver reflejados también en las curvas de supervivencia de Kaplan-Meier (Fig. 1).

6

No difference in outcome was found either between patients with atypical symptoms only and those with a combination of typical and atypical symptoms, or between patients with a TIA and those with a minor ischaemic stroke.

No hemos encontrado diferencias en cuanto a la evolución entre los enfermos que tenían sólo síntomas atípicos y aquellos que presentaban una combinación de síntomas típicos y atípicos, ni tampoco entre los enfermos que presentaban AIT y los que habían sufrido ictus isquémicos menores.

7

Table II shows the baseline characteristics of the two groups.

En la tabla II se muestran las características basales de ambos grupos.

8

We adjusted for minor differences between patients with atypical and typical symptoms by a multivariate analysis.

Tuvimos en cuenta las pequeñas diferencias existentes entre los enfermos con síntomas atípicos y típicos mediante análisis multivariante.

9

After adjustment for single variables and for all relevant variables simultaneously, HR (atypical vs typical) for stroke remained at 0.6 (95% CI 0.4–0.9), and that for cardiac events slightly increased to 1.5 (1.1–2.0).

Tras realizar este ajuste para las variables analizadas de forma individualizada y para todas las variables relevantes consideradas simultáneamente, observamos que la PR (casos atípicos frente a casos típicos) del ictus seguía siendo de 0,6 (IC del 95%, 0,4–0,9), mientras que la PR de los episodios cardíacos aumentó ligeramente hasta 1,5 (1,1–2,0).

10 If age and diastolic blood pressure were entered as continuous instead of dichotomised variables, the difference in outcome between the two groups was again significant.

Al introducir la edad y la tensión arterial diastólica como variables continuas en vez de dicotómicas, comprobamos que la diferencia de la evolución entre ambos grupos era también significativa.

11 In a separate analysis we assessed which of the atypical symptoms in particular implied an increased risk of cardiac events.

En otro análisis, valoramos cuáles eran precisamente los síntomas atípicos que implicaban un aumento del riesgo de episodios cardíacos.

 Ian A. Williams

12 In all patients with atypical symptoms, the cardiac to cerebral events ratio was 1.5 compared with 0.6 in patients with typical attacks.

En el conjunto de enfermos con síntomas atípicos, encontramos que la proporción de episodios cardíacos respecto a cerebrales era de 1,5, frente a 0,6 en los enfermos que presentaban episodios típicos.

Thus the revised text now has 8 first-person forms: 4 in SORs, 2 in Procedures and 2 in the Comment/Metatextual categories. As mentioned above, there is a blurring of the exclusive-inclusive division as the writer draws the readers into the deductive process so that they feel they are participants not only in the interpretation of the results but also in the underlying methodological choices: had they been in that situation, they would have done exactly the same. In the proposed version, the first-person Procedures no longer stand alone as two individual choices, but have been incorporated into an overall coherent discourse strategy that will be familiar to Spanish readers of biomedical RAs.

5. Conclusions The study has revealed that Spanish authors using first-person forms in their Results sections have a different rhetorical purpose and employ a different discourse strategy from their counterparts publishing in English medical journals. The higher frequency provides cohesion, while the combination of inclusive and exclusive use serves to implicate the reader in the unfolding discourse and in all types of statement, but especially the SOR. In contrast, the small number of instances in the English texts suggests that the first person functions as a signal to the reader that something significant is being focussed on, namely that through exclusive use of the first person the writer is releasing the reader from any commitment to the non-standard methods or personal choices described. As an illustration of the methodological approach, the study has shown that the differences – excesses and deficits – identified by contrastive analysis and the rich linguistic data obtained with corpus techniques provide a solid empirical basis for the formulation of strategic options that can be directed at any deficiencies detected in actual translations evaluated by the same methods. The study also indicates the usefulness of combining parallel and comparable corpora in Translation Studies in order to detect discourse patterns that could go unnoticed in translation practice and with the use of a translation corpus alone.



Getting the ACCENT right in Translation Studies 

References Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. 1999. Longman Grammar of Spoken and Written English. London: Longman. Duff, A. 1981. The Third Language. Recurrent Problems of Translation into English. Oxford: Pergamon Institute of English. Frawley, W. 1984. “Prolegomenon to a theory of translation.” In Translation: Literary, Linguistic and Philosophical Perspectives, W. Frawley (ed.), 159–175. London/Toronto: Associated University Presses. Gellerstam, M. 1986. “Translationese in Swedish novels translated from English.” In Translation Studies in Scandinavia [Lund Studies in English 75], L. Wollin and H. Lindquist (eds), 88– 95. Lund: CWK Gleerup. House, J. 1997. Translation Quality Assessment: A Model Revisited. Tübingen: Narr. Hyland, K. 1998. Hedging in Scientific Research Articles. Amsterdam/Philadelphia: John Benjamins. Johansson, S. 1998. “On the role of corpora in cross.linguistic research.” In Corpora and Crosslinguistic Research: Theory, Method and Case Studies, S. Johansson and S. Oksefjell (eds), 3–24. Amsterdam/Atlanta, GA: Rodopi. Scott, M. 1998. WordSmith Tools. Oxford: Oxford University Press. Sinclair, J. 1991. Corpus, Concordance and Collocation. Oxford: Oxford University Press. Tognini-Bonelli, E. 2001. Corpus Linguistics at Work. Amsterdam/Philadelphia: John Benjamins. Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins. Williams, I.A. 1999. “Results sections of medical research articles: Analysis of rhetorical categories for pedagogical purposes.” English for Specific Purposes 18 (4): 347–366. Williams, I.A. 2004. “How to manage patients in English-Spanish translation: A target-oriented contrastive approach to Methods.” Target 16 (1): 69–103. Williams, I.A. 2005. “Thematic items referring to research and researchers in the Discussion section of Spanish biomedical articles and English-Spanish translations.” Babel 51 (2): 124–160. Williams, I.A. 2006. “Towards a target-oriented model for quantitative contrastive analysis in Translation Studies: An exploratory study of theme-rheme structure in Spanish-English biomedical research articles.” Languages in Contrast 6 (1): 1–45. Williams, I.A. 2007. “A corpus-based study of the verb observar in in English-Spanish translations of biomedical research articles.” Target 19 (1): 85–103. Williams, I.A. 2008. “Semantico-syntactic environments of the verbs show and demonstrate and Spanish mostrar and demostrar in a bilingual corpus of medical research articles.” International Journal of Corpus Linguistics 13 (1): 38–74.

Die Kirche im Dorf oder die Regierung im Wald lassen Zum Übersetzungsproblem der Namen von Ämtern und Ähnlichem für Nachrichtenzwecke im Medium Radio Dieter Hermann Schmitz

Universität Tampere, Finnland

The most convincing way to answer the question why Translation Studies matters is to point to the practical benefits TS-results or reprocessed field experience offer in handling practical translation tasks. TS provides translators on the job with help for decision-making and problem-solving. A permanent difficulty when translating news for radio broadcasts lies in the names of organisations such as political parties, associations and institutions of all kinds. In my article I will discuss this issue, specifically for the language pair Finnish-German. For example, what can be done with ‘Suomen Ääni- ja kuvatallennetuottajat’, the name of a Finnish association for record industry products that could – at least in principle – be translated word by word. In theory, translators have a wide range of possibilities: for instance, they can transfer SL words directly into the target text, or invent new expressions. However, in practice, they are usually not free to choose. Instead, they have to look for precedents or check for existing translations in use. A procedure depicting the process of finding possible solutions is presented in the form of a decision tree. Keywords: Applied TS, proper names, news translation, didactics of translation

Vorbemerkung Im ersten Teil meiner Überschrift spiele ich auf zwei Wendungen oder geflügelte Worte an: Die erste der beiden, „die Kirche im Dorf lassen“, gilt als lexikalisiert und wird im Duden für Redewendungen und sprichwörtliche Redensarten umschrieben, im Sinne von „etwas im vernünftigen Rahmen belassen, nicht übertreiben“. Sie beruht

 Dieter Hermann Schmitz

wahrscheinlich auf der Vorstellung, dass „der angemessene Platz der Kirche in der Mitte des Dorfes ist.“ (Duden 1992:€384 f.) Der zweite Teil spielt auf eine populäre Fehlübersetzung an, die vor Jahren durch finnische Prospekte in deutscher Übersetzung geisterte und zumindest in kleinen Fachkreisen in Finnland zur stehenden Wendung geworden ist. Dort war Metsähallitus, das finnische Amt für Staatswälder, durch eine plumpe Glied-für-Glied-Übersetzung übertragen worden. Der Name des Amtes ist auf seiner Oberflächenstruktur ein Kompositum zweier Appellativa und segmentierbar in metsä, das mit Wald, Waldung, Forst, Gehölz usw. übersetzt werden könnte, sowie hallitus, das je nach Kontext Regierung, Kabinett oder Vorstand, Aufsichtsrat, Direktion aber auch Verwaltung, Amt u.Ä. bedeuten kann. In besagtem Prospekt war dann von Waldregierung die Rede. Mit dieser unfreiwillig komischen Lösung bin ich beim Kern eines Problems, auf das man zwangsläufig stößt, wenn die Namen von Ämtern und Ähnlichem etwa für Nachrichtenzwecke übersetzt werden. Dabei handelt es sich im Folgenden um Beobachtungen und Erfahrungen aus der Praxis, aus denen heraus sich ein Vorgehen bei der Lösung derartiger Probleme entwickelt hat, das einer Art Entscheidungsbaum ähnelt und hier zur Diskussion gestellt werden soll. Auch wenn es entstanden ist für die engen Zwecke der Übersetzung von Radionachrichten im Sprachenpaar Finnisch-Deutsch, ist es im Prinzip auch für ähnliche Zwecke in anderen Sprachenpaarkombinationen anwendbar.

1. Problemstellung Das Interesse an dieser Problematik leitet sich zunächst aus persönlicher Betroffenheit ab. Im Alltag der universitären Lehre begegnen wir dem Übersetzungsproblem derartiger Namen im Rahmen eines Kurses mit quasi-authentischem Charakter. (Zur Kategorisierung von Übersetzungsaufträgen, insbesondere für die universitäre Lehre vgl. auch Schopp 2002). Für den uni-eigenen Sender, Radio Moreeni, werden in der Translationswissenschaft regelmäßig Nachrichten vom Finnischen ins Deutsche übersetzt und anschließend durch Studenten auf Deutsch eingesprochen und ausgestrahlt. Quasi-authentisch nenne ich diese Übersetzungen einerseits deshalb, weil Auftrag oder Adressatenkreis von Seiten des Radiosenders nicht definiert wurden, mithin Kursleiter und Studenten bei der Gestaltung der Sendung und der Auswahl ihrer Ausgangstexte relativ freie Hand genießen, andererseits deshalb, weil diese Arbeit nicht bezahlt wird, jedenfalls nicht mit barer Münze, bestenfalls mit Studienpunkten für die Studenten und der Möglichkeit, ein wenig Erfahrung in der journalistischen Radioarbeit zu sammeln. Immerhin erreichen die (auditiven) Texte ein potenzielles Publikum von rd. 300.000 Hörern im Großraum Tampere; die Übersetzungen erlangen dadurch sicherlich einen anderen Status als Übungen für den Ordner. Gesendet werden die Nachrichten auf Deutsch einmal wöchentlich. Als Quelle dienen Tickermeldungen der finnischen Nachrichtenagentur STT (Suomen Tietotoimisto). Die Texte von STT werden als freie



Die Kirche im Dorf oder die Regierung im Wald lassen 

Grundlage verwendet, über deren Auswahl, Zusammenstellung, Kürzung und redaktionelle Bearbeitung von Studenten und Kursleiter entschieden wird – ohne jedoch den propositionalen Gehalt der Nachrichten anzutasten oder in Frage zu stellen. Was einen echten Übersetzungsbedarf anbelangt, ist ein solcher nur bei einer Konzentration auf Meldungen aus Finnland gegeben, da STT-Nachrichten aus aller Welt in der Regel ihrerseits Übersetzungen darstellen, anderweitig nachzulesen und somit kaum übersetzungsrelevant sind. Es zählt in den westlichen Medien sozusagen zum guten Ton – eher noch: zu einer ethischen Pflicht – Informationen in einen Zusammenhang zu stellen, Aussagen anderer als solche kenntlich zu machen und Quellen zu nennen. In der klassischen Lasswell-Formel gehört neben den Angaben zum Was, Wann und Wo auch an zweiter Stelle die Information des Woher oder Von Wem, bevor an späterer Stelle auf Wieso und Weshalb eingegangen wird (vgl. Gawlas 1998:€236.). Als Gegenstand einer Nachricht, als zitierte Experten und nicht zuletzt als Quelle tauchen überaus häufig Einrichtungen unterschiedlichster Art auf: Ein Statistikamt berichtet von neuen Entwicklungen, ein Wirtschaftsforschungsinstitut verweist auf die letzten Trends, eine Behörde gibt eine Mitteilung heraus, Verbände nehmen Stellung zu tagesaktuellen Problemen usw. Das heißt, bei fast allen Nachrichten begegnet einem das Problem der Namen; Namen von Entitäten, die ich unter ‚Organisation’ kategorisieren möchte. Sie auszulassen und unübersetzt zu lassen, würde gegen westliche Normen des Radio-Journalismus und der Hörgewohnheiten verstoßen (von La Roche et al.€2004:€212).

2. Namen von Organisationen: Abgrenzung und Überblick Wie bereits deutlich geworden ist, betrachte ich die Bezeichnungen für Ämter und Ähnliches als Namen, hier zu verstehen als Namen im engeren Sinne, als Eigennamen (nomen proprium). In Anlehnung an die DIN 2342, Begriffe der Terminologielehre, definiere ich „Name“ als sprachliche Bezeichnung eines Individualbegriffs, der vom einzelnen Gegenstand ausgehend gebildet wird. (Baxmann-Krafft 1999). Allen Namen ist prinzipiell die Funktion der Identifikation gemeinsam. Für die Namen von Organisationen – und hierzu zähle ich neben Ämtern und Behörden auch Ministerien, Verbände, Parteien, Vereine, Bündnisse, Forschungseinrichtungen, Institute aller Art, politische Zusammenschlüsse, Interessengemeinschaften, Einrichtungen aus Kultur und Wirtschaft etc. – bin ich geneigt, eine eigene Gruppe von Namen anzunehmen, die ich im Folgenden kurz von verwandten Namengruppen absetzen möchte. Eine Besonderheit ist, dass Namen von Organisationen sich häufig aus Appellativa zusammensetzen, in ihrer Gesamtheit und ihrem Stellenwert aber Eigennamencharakter annehmen. Sie erscheinen dann in relativ festgefügten Formen (Europäischer Rat ≠ Europarat) und Schreibweisen (Gewerkschaft „ver.di“ mit dem Binnenpunkt).

 Dieter Hermann Schmitz

2.1

Organisationsnamen vs. Anthroponyme

Namen von Organisationen unterschieden sich natürlich in erster Linie von anderen Namensgruppen durch ihr andersgeartetes Denotat. Von den Anthroponymen (Personennamen) im Besonderen unterscheiden sie sich aber auch insofern, als Letztere in der heutigen (jedenfalls westlichen) Übersetzungstradition den stärksten Hang zur sprachübergreifenden Invarianz aufweisen; zumindest bei real existierenden Personen, im Gegensatz zu Charakteren aus fiktiven Welten, wie sie uns in der Belletristik begegnen. Schlimmstenfalls treffen wir (wiederum aus westlicher Perspektive) bei den Personennamen auf das Problem der Transliteration, etwa von Namen aus dem Russischen, Chinesischen oder Arabischen, die in anderen Schriftsystemen als dem lateinischen wiedergegeben werden. Wenn von einem Hang zur Invarianz die Rede war, so sollte klar sein, dass damit kein absolutes Primat verbunden ist und es sicherlich Ausnahmen gibt, ganz zu schweigen von gewissen lautlichen Angleichungen. Ich gehe aber davon aus, dass „Angela Merkel“ auch in anderen Sprach- und Kulturräumen „Angela Merkel“ genannt wird und nicht (in der Rückübersetzung) „Engel Grenzhüter“. (Der Familienname „Merkel“ leitet sich von „Markward(t)“ ab. Vgl. Bahlow 1988:€336 u. 328.)

2.2

Organisationsnamen vs. Toponyme

Die oben genannte Tendenz zur Invarianz lässt sich in jüngster Zeit latent auch bei den Toponymen (Ortsnamen) feststellen. Es gilt heute im Deutschen als politisch korrekt, statt von Peking eher von Beijing zu sprechen. Im Zusammenhang mit Toponymen begegnet einem das besondere Problem der Exonyme, also der Namengebung von außen. Gute Beispiele hierfür sind die Namen ehemaliger Kolonialländer (z.B. Rhodesien – Simbabwe), ähnlich auch Birma [finnisch: Burma], das sich selbst Myanmar nennt und zunehmend auch in westlichen Nachrichten so genannt wird. Gewisse Ortsnamen lassen erkennen, dass sie direkte Lehnübersetzungen sind (z.B. finnisch Itämeri – Ostsee, obwohl für Finnland besagtes Meer im Westen und Süden liegt).

2.3

Organisationsnamen vs. Produktnamen

Produktnamen, insbesondere die neueren Datums (Maoam, Smart, Tesa...), unterscheiden sich u.a. von den Namen für Organisationen durch den Anspruch ihrer oft weltweiten Singularität und Unverwechselbarkeit, durch ihre Kreativität, häufig gepaart mit Kürze und Prägnanz, und durch mögliche rechtliche Bestimmungen wie die eingeschränkte Verwendung durch Dritte. Ob man auch noch eine weitere Kategorie für Firmennamen (Shell, Plus, Ikea, BenQ...) aufmachen müsste, will ich nicht weiter problematisieren; eher denke ich, dass Firmennamen (insbesondere die älteren Schlags wie Rheinmetall, Westdeutsche Farbengesellschaft...) unter das Hyperonym „Organisation“ zusammenfassbar sind,



Die Kirche im Dorf oder die Regierung im Wald lassen 

und zunehmend eine starke Überlappung mit Produktnamen aufweisen (vgl. Coca Cola als Produkt- und Firmenname).

2.4

Organisationsnamen und Realien

Aus translationswissenschaftlicher Perspektive ist natürlich auch die Frage der Realien oder Kultureme von Interesse, die sich vielfach mit den hier aufgezählten Gruppen von Namen überschneiden. Die Realien ragen aber weit über die Gruppe der hier vorgestellten Namengruppen hinaus. Kujamäki zählt zu selben auch Bezeichnungen aus den Bereichen Berufsleben, Verwaltung, Maßeinheiten, Traditionen, Flora, Fauna, Landschaft, Nahrung, Kleidung, bauliche Einrichtungen und Freizeit (Kujamäki 1998:€26f.) In diesem gerafften Überblick, bei dem es mir nicht um scharfe Definitionen, sondern eher um eine lose Abgrenzung ging, die zwangsläufig vielfache Überschneidungen aufweist, stehen die Namen von Organisationen den Appellativa am nächsten: Sie setzen sich, wie schon weiter oben angemerkt, in der Regel aus selbigen zusammen, tragen im landläufigen Sinne eine Bedeutung und sind damit prinzipiell auch übersetzbar. Lexikalische Lücken als Problem ergeben sich normalerweise nicht, Glied-fürGlied-Übersetzungen sind daher bildbar.

3. Fallbeispiel Als Beispiel möchte ich eine Kulturnachricht vom Januar 2006 heranziehen, die am 25. des Monats als STT-Nachricht 170 ins Netz gestellt wurde. In dieser ging es um statistische Angaben zum Verkauf von Tonträgern in Finnland. Sie enthielt außer Verkaufszahlen und genauen Angaben zu den Top 20 der Verkaufslisten in- und ausländischer Interpreten auch Angaben zur Rock-Gruppe Nightwish, ihren früheren Erfolgen sowie zur Zusammensetzung der Band. Insgesamt hatte der Text eine Länge von rd. 1400 Zeichen. Es lässt sich natürlich über die Bedeutsamkeit und Übersetzungsrelevanz dieser Nachricht streiten; sie war keinesfalls der Aufmacher, sondern gehörte zu den „weichen Nachrichten“ im Bereich Kultur und Unterhaltung gegen Ende der Sendung. Die Wichtigkeit der Nachricht ist aber hier zweitrangig. Interessant ist die Nennung der Quelle im letzten Satz des Textes. Dort stand sinngemäß: „Die Angaben beruhen auf den Statistiken der Suomen Ääni- ja kuvatallennetuottajat.“ Selbst Studierende im zweiten oder dritten Studienjahr, die an theoretischen Einführungen und ersten Übungen bereits einiges hinter sich haben, neigen erfahrungsgemäß stark dazu – nach einem flüchtigen und normalerweise erfolglosen Blick ins zweisprachige Wörterbuch oder seinem elektronischen Pendant – einen solchen Namen in seine semantischen Bestandteile zu segmentieren und sich an einer Gliedfür-Glied-Übersetzung zu versuchen.

 Dieter Hermann Schmitz

Problematisch erscheint die zitierte Bezeichnung schon allein deshalb, weil sich, in der Terminologie Kollers, im lexikalischen Bereich eine Diversifikation ergibt (bei Koller einer von fünf Typen potentieller Äquivalenz). Die folgende Übersicht soll das verdeutlichen: Suomen

Ääni-

ja

Kuva

– tallenne –

tuottajat

Finnlands Finnische(r) ...

Stimme Geräusch Schall Ton Klang Laut ...

und sowie ....

Bild Bildnis Gemälde Foto Fotographie Porträt ... Eindruck Impression Vorstellung ...

Aufzeichnung Aufnahme Speichersatz ...

Hersteller Produzent Erzeuger ... Urheber Verursacher ... Fabrikant Anbauer Producer Realisator ...

von Finnland ... Finnland [nachgestellt]

Als mögliche Ergebnisse wären also denkbar: Finnische Ton- und Bildaufzeichnungshersteller oder Finnlands Schall- und Fotographieaufnahmeproduzenten. Es ließen sich eine Reihe weiterer Wortketten bilden, die z. T. unfreiwillig komisch wirken oder annähernd unverständlich sein könnten. Durch den Kontext der Nachricht über den Erfolg einer Rock-Gruppe und die Verkaufszahlen von Alben können wir eine Organisation dieses Namens durch unser Weltwissen wohl irgendwie zuordnen. In ihrem Zusammenhang ergibt eine wie auch immer geartete Glied-für-Glied-Übersetzung – wegen „sprechender“ Namen – auch einen gewissen Sinn. Aber es stellt sich die Frage, ob solche Übersetzungen auch gut genug sind und ob sie akzeptabel und legitim sind?

4. Grundsätzliche Lösungsmöglichkeiten In der Fachliteratur, in Handbüchern und praktischen Ratgebern, werden für Fälle der Übersetzung von Organisationsnamen eine Reihe unterschiedlicher Lösungsmöglichkeiten angegeben – abhängig von der Funktion der Übersetzung, dem Auftrag, der Kommunikationssituation, dem Sprachenpaar, den ZT-Rezipienten sowie einer Reihe weiterer Faktoren. Am dringlichsten erscheint dieses Problem bei gesellschaftlich relevanten und auch international bedeutsamen Organisationen, wie z.B. Ministerien und Verwaltungseinrichtungen, und für Texte und Kommunikationssituationen, die über das bloße Informieren oder Unterhalten weit hinausreichen und von Bedeutung für



Die Kirche im Dorf oder die Regierung im Wald lassen 

die Rechtsprechung sind, z.B. Vertragsabschlüsse, wirtschaftliche Transaktionen. Die Palette denkbarer Lösungen umfasst folgende Möglichkeiten: 1. die Verwendung eines funktionalen Äquivalents und dessen mögliche Erweiterung mithilfe lexikalischer Mittel (Beispiel: DE: Amtsgericht – FI: käräjäoikeus); 2. eine Paraphrasierung, erklärende Umschreibung sowie der Rückgriff auf Editionstechniken (Beispiel vom Frühjahr 2007, nach den finnischen Parlamentswahlen: Seinerzeit war die Rede von der Einrichung eines kuntaministeriö (wörtlich: Gemeindeministerium). In einer Pressemitteilung der Finnischen Botschaft in Berlin wurde diese umschrieben mit Ministerium, zuständig für Kommunalwesen und Verwaltungsangelegenheiten (http://www.finnland.de/de/ [16.4.07]); 3. Generalisierungen (Beispiel: die finnische Partei Kansallinen Kokoomus, meist: Nationale Sammlungspartei, wird gelegentlich übersetzt mit die Konservativen); 4. Bildung eines formalen Äquivalents (Beispiel: denkbar für das US-amerikanische Federal Bureau of Investigation FBI DE: Bundesuntersuchungsamt); 5. Fremdwortentlehnung (etwa wenn in deutschen Texten das finnische Parlament eduskunta genannt wird); 6. die Neuschaffung eines (abstrakten oder neutralen) Ausdrucks, der keine falschen Erwartungen und Vorstellungen erweckt. (vgl. für viele: Kinnunen 2006:€44). Hinzu kommen verschiedene Mischformen sowie als weitere Alternative die Auslassung, die aber nur sehr bedingt verwendbar ist. Die hier genannten Möglichkeiten zeigen zwar grundsätzliche Lösungswege auf, erwecken aber zum Teil den etwas trügerischen Eindruck, als würde jeder Übersetzer gleichsam wieder bei Null anfangen. Sie sind perspektiv angelegt und mögen glauben machen, als stünde die Entscheidungsgewalt dem jeweiligen Übersetzer gleichsam jedes Mal von Neuem zu. Natürlich sollte Studierenden eine kritisch-reflektierende Herangehensweise und der Mut zu eigenen Entscheidungen vermittelt werden, doch für die Zwecke unserer Radionachrichten, zumal bei einmaliger auditiver Aufnahme, empfiehlt sich viel eher die Rückversicherung bei Autoritäten, die Suche nach dem „translatorischen Präzedenzfall“ und dessen Evaluierung sowie eine Einschätzung, was an Übersetzungen möglicherweise schon bekannt und etabliert ist. Vereinfacht ausgedrückt: Statt abzuwägen, wie übersetzt werden könnte, sollte die Problemlösung mit der Recherche danach beginnen, wie bereits – in Bezug auf Zweck und Genre – übersetzt worden ist. Wir gehen bei unserer Arbeitsweise von einem nachgeordneten Übersetzen aus, mit einem retrospektiven (deskriptiven) Ansatz. Für diese Zwecke möchte ich in Abschnitt 6 eine Art Entscheidungsweg vorstellen, der eine solche Vorgehensweise stufenweise veranschaulicht. Vorab aber noch einige Anmerkungen zu rechtlichen Aspekten, Normen und Gepflogenheiten.

 Dieter Hermann Schmitz

5. Namengebung Die weiter oben erwähnte festgefügte Form und Schreibung von Organisationsnamen leitet sich nicht zuletzt vom freien Recht auf Namengebung ab, das nur durch gewisse rechtliche Bestimmungen eingeschränkt ist (für Deutschland etwa durch das Namensrecht im BGB oder die Markenverordnung, die auch geschäftliche Bezeichnungen und geographische Herkunftsangaben schützt). Damit einher geht m.E. auch das Recht auf die Übersetzung eines Namens. Personennamen werden uns nach gewissen rechtlichen Regelungen und Traditionen von anderen verliehen und dürfen nur unter bestimmten Voraussetzungen geändert werden. Die meisten Makrotoponyme, etwa die Namen von Ländern, entwickeln und festigen sich oft über historisch längere Zeiträume, statt punktuell festgelegt zu werden. Eine Organisation hingegen verleiht sich ihren Namen häufig selbst und kann ihn unter Umständen relativ beliebig ändern. (Was nicht ausschließt, dass es um die Namen von Fächern und Instituten an Universitäten unter Umständen gehörigen Zank gibt.) Häufig wird die Namengebung durch einen deklarativen Akt zeremoniell begangen und bemüht dazu konkrete Äußerlichkeiten (Gebäude, Insignien, Vereinsund Firmenlogos, Symbole wie Fahnen, Briefköpfe usw.). Dieser Akt der Namengebung ist m.E. auch verbunden mit einem gewissen Recht auf die Form der Übersetzung eines Namens. Ein Übersetzer hat somit oftmals nicht freie Hand, wie die oben aufgelistete Liste an Übersetzungsmöglichkeiten suggerieren könnte. In der übersetzerischen Praxis hat sich folgende Arbeitsweise als nützlich erwiesen.

6. Entscheidungsweg für die Praxis der Radioübersetzungen Die erste Frage im Zeitalter der Vernetzung lautet sinnvollerweise: Gibt es auf den Internetseiten einer Organisation bereits eine Übersetzung des Namens in der betreffenden Zielsprache? Anders formuliert: Hat bereits eine Namengebung in fremden Sprachen („Selbstübersetzung“) stattgefunden?

6.1

Selbstübersetzung?

Lautet die Antwort „Ja“, ist das Problem quasi schon gelöst. Ein Beispiel: Wenn es etwa darum geht, den Namen eines Staatsbetriebs zu übersetzen, der dem finnischen Verkehrsministerium untersteht, namentlich ilmailulaitos (wörtlich: ilmailu ≈ Luftfahrt / Aeronautik / Flugwesen + laitos ≈ Institut / Anstalt...), so sollte der erste Blick auf die Net-Seiten des Ministeriums gehen. Dort findet sich keine Bildung mit „-institut“, „-anstalt“, sondern der Name Finnische Luftfahrtsadministration.



Die Kirche im Dorf oder die Regierung im Wald lassen 

6.2

Nachschlagewerke?

Findet sich keine Übersetzung in Texten, deren Textsender mit dem Bezeichneten identisch ist, muss sich als Folgefrage anschließen: Steht ein verlässliches Nachschlagewerk zur Verfügung? Dazu gehören im Falle Finnlands Veröffentlichungen der finnischen Staatskanzlei, Valtioneuvoston kanslia, oder anderer Ministerien und ihrer Sprachendienste, an denen Stäbe von Fachleuten mitwirken. Dazu zählt auch die Online-Terminologiebank der Staatskanzlei, VALTER, die einen anderen Stellenwert besitzt als zweisprachige Wörterbücher oder beliebige Paralleltexte aus dem Netz. Ein Beispiel: So erfährt man im Hochschulglossar Finnlands, Korkeakoulusanasto, dass die Humanistinen tiedekunta der Universität Tampere (humanistinen ≈ humanistisch + tiedekunta ≈ Fachbereich, Fakultät...) auf Deutsch offiziell nicht etwa „Humanistischer Fachbereich“ heißt, sondern den (übersetzten) Namen „Philosophische Fakultät“ trägt. Ähnliche Fakultäten anderer Universitäten tragen den formal selben finnischen Namen, haben aber andere Übersetzungen. Auch bei der Anerkennung von (normativen) Autoritäten, wie etwa dem genannten Sprachendienst der finnischen Staatskanzlei, verbietet sich natürlich keinesfalls eine kritische Distanz, insbesondere bei fehlender Aktualität. Das genannte Werk, Yliopistosanasto, stammt von 1999 und bedarf sicherlich einer Überarbeitung, zumal seither Dutzende Institute neu gegründet und umstrukturiert worden sind und sich ihre Namen entsprechend geändert haben. Daneben ließe sich eruieren, ob es zulässige Synonyme gibt.

6.3

Schwesterorganisationen?

Findet sich der Name einer Organisation nicht in einem Nachschlagewerk, dessen Rang und Verlässlichkeit wir anerkennen, schließt sich erneut eine Folgefrage an. Diese könnte lauten: Gibt es im anderen Sprach- und Kulturraum direkte Schwesterorganisationen? Das kann z.B. dann der Fall sein, wenn Organisationen verschiedener Länder Mitglieder einer Dachorganisation sind. Äußerlich kommt dies dadurch zum Ausdruck, dass auf Internetseiten wechselseitig aufeinander verwiesen wird oder Links bestehen. Ein Beispiel: die Organisation Eurooppalainen Suomi (eurooppalainen ≈ Europäer(in) / europäisch + Suomi ≈ Finnland) ist ein eingetragener Verein, der politisch unabhängig ist und zur Förderung des europäischen Gedankens beitragen möchte. Eurooppalainen Suomi ist kein Organ oder Ableger der Europäischen Union, wenn der Verein selbiger auch nahe steht und die EU und den Europa-Gedanken ideell voranbringen möchte. Sein deutsches Gegenstück trägt den Namen Europäische Bewegung Deutschland, in Österreich entsprechend Europäische Bewegung Österreich. Die finnische Organisation müsste also in Anlehnung daran Europäische Bewegung Finnland genannt werden. Solche Schwesterorganisationen sind aber eher selten anzutreffen.

 Dieter Hermann Schmitz

6.4

Parallelübersetzungen?

Falls also auch die letzte Frage verneint wird, schließt sich die folgende Frage an: Verwendet die Organisation Übersetzungen ihres Namens in andere Sprachen, mit deren Hilfe man eine Analogiebildung ins Deutsche vornehmen könnte? In Finnland informieren viele Organisationen im Internet nicht nur auf Finnisch, der Sprache der Mehrheit, sondern auch auf Schwedisch, der zweiten Landessprache, sowie oft auch auf Englisch als der weltweiten lingua franca. Ein Beispiel: Die finnische Partei Perussuomalaiset, 1995 gegründet und bekannt für ihre markigen Sprüche, war im Frühjahr 2007, während der Parlamentswahlen, mehrfach auch Gegenstand der ausländischen Presse (perus ≈ Grund / Basis / Fundament... + suomalaiset ≈ Finnen). Auf den parteieigenen Internetseiten bezeichnet sich die Partei selbst als schwedisch sannfinländarna (sann ≈ wahrhaftig / wirklich / richtig / echt... + finländarna ≈ Finnen) und englisch True finns. Das Schwedische und das Englische als dem Deutschen verwandte Sprachen können ggf. bei der Übersetzung FI-DE helfen. Hier aber stellt sich die Frage, ob diese Parallelübersetzungen ausreichend sind. Was im Vergleich auch ein wenig stutzen macht, ist dass das finnische „perus-“ als Wortbildungselement im Sinne von „Grund-, Elementar-, Erz-“ semantisch nicht unbedingt an die Bedeutung „wahr, wahrhaft, wahrheitsgetreu; ehrlich, richtig, tatsächlich...“ denken lässt, im gleichen Maße wie das adjektivische „sann-“ des schwedischen Namens oder das englische „true“.

6.5

Translatorischer Präzedenzfall?

Falls wir die obige Frage verneinen müssen, weil ein gewisses Rest-Unbehagen bleibt oder überhaupt keine Parallelübersetzungen vorliegen, stellt sich die Anschlussfrage: Gibt es anderweitig den „translatorischen Präzedenzfall“? Wo ist der Name einer Organisation bereits einmal übersetzt worden, wann, von wem, zu welchem Zweck und für welches Publikum? Welche Autorität genießt der Textsender? Die Aufgabe der Studenten besteht hier also nicht – um es noch einmal zu betonen – darin, einen Namen (erstmalig) zu übersetzen, sondern bereits vorliegende Übersetzungen zu recherchieren, was mithilfe von Suchmaschinen im Internet relativ leicht möglich ist, allerdings ein wenig Erfahrung mit Suchfunktionen, die Kenntnis relevanter Internetseiten und Online-Dienste sowie eine Portion Urteilsvermögen und Erfahrung erfordert, welche Lösung man möglicherweise selbst übernimmt. Ein Beispiel: Für den Parteinamen Perussomalaiset ließen sich im Frühjahr ’07 folgende Belege von Übersetzungen ins Deutsche finden: (a) Der „Standard“, eine überregionale Zeitung aus Österreich, die sich vor allem im Raum Wien verkauft, verwendete in ihrem Internetangebot „Echte Finnen“. (b) Das „Dokumentationsarchiv“ einer Gruppe Mainzer antifaschistischer Studenten sprach von der Partei „Wahre Finnen“.



Die Kirche im Dorf oder die Regierung im Wald lassen 

(c) Auf einzelnen Seiten der freien Enzyklopädie Wikipedia war von der Partei der „Urfinnen“ die Rede. (d) Die Internetseite der Finnischen Botschaft in Berlin flüchtete sich vor der Wahl im März 2007 in einen Anglizismus und sprach von der „True Finns Party“ [sic!]. (e) Tilastokeskus, das Statistische Zentralamt des Landes, benannte die Partei „Basisfinnen“. Dies tat im Übrigen auch das Auswärtige Amt in Berlin. (Quellen: siehe Literaturverzeichnis) Bei der Auswahl und Bewertung spielt es eine Rolle, welche Autorität und Verlässlichkeit sich der Stelle zuordnen lässt, die eine bestimmte Übersetzung verwendet. Wie etabliert, wie bekannt, wie alt, wie häufig erscheint eine bestimmte Übersetzungsvariante? Ist der Name sprechend genug, um die Zielgruppe ausreichend zu informieren? Hier sollte klar sein, dass das Auswärtige Amt der Bundesrepublik Deutschland ein völlig anderes Gewicht hat und Ansehen genießt als eine Gruppe engagierter studentischer Antifaschisten aus Mainz oder ein Artikel in Wikipedia.

6.6

Konsultation des Denotats

Sollten alle bisher hier dargestellten Lösungswege zu keinem befriedigenden Ergebnis geführt haben, bleibt als letztes Mittel, sowie generell zur Rückversicherung, die direkte Nachfrage bei der betreffenden Organisation. (Da Nachrichtenübersetzungen in der Regel unter Zeitdruck entstehen, empfiehlt sich die unmittelbare Kontaktaufnahme per Telefonanruf.) Wenn man so will, ist dies die Konsultation des Denotats, die eben auch nur bei Organisationen möglich ist: Bei Personennamen ist sie unnötig, bei geographischen Namen unmöglich, bei Produktnamen hinfällig. Der hier vorgestellte Entscheidungsweg wurde als Abfolge von Leitfragen verlaufend dargestellt, was der Übersichtlichkeit und Einfachheit halber geschah. Eine strikte Reihenfolge oder strenge Hierarchie gibt es jedoch nur im Modell, die Praxis erweist sich oft als eine Kombination bzw. ein gleichzeitiges Nebeneinander-Ablaufen einzelner Rechercheschritte.

7. Schlussbemerkung Um auf das weiter oben angesprochene Beispiel zurückzukommen: Eine Glied-fürGlied-Übersetzung von Suomen Ääni- ja kuvatallennetuottajat erbringt zwar Ergebnisse, aber sie entsprechen nicht dem, was die betreffende Organisation selbst verwendet sehen möchte. Recherche und Nachfrage ergaben, dass die entsprechende Organisation auf Deutsch als Finnischer Verband der Phonographischen Industrie bezeichnet werden möchte, ein Ergebnis, das ein Übersetzen im engen Sinne schwerlich erbracht hätte. Und legitim ist das, was die betreffende Organisation verwendet wissen will: Das Recht

 Dieter Hermann Schmitz

auf den Namen sollte bei Organisationen auch das Recht auf die Übersetzung desselben umfassen.

Literatur Bahlow, H. 1988. Deutsches Namenlexikon. Familien- und Vornamen nach Ursprung und Sinn erklärt. Hamburg: Gondrom. Baxmann-Krafft, E.M. 1999. Normen für Übersetzer und technische Autoren. Deutsches Institut für Normung e.V. (Hg.). Berlin. Finnische Staatskanzlei/ Terminologiedienst. 2005. Korkeakoulusanasto [Hochschulglossar]. Helsinki: Yliopistopaino. Gawlas, Ch. 1998. „Texte von Presseagenturen.“ In Handbuch Translation, M. Snell-Hornby, H.G. Hönig, P. Kußmaul und P.A. Schmitt (Hg.), 236–237, Tübingen: Stauffenburg. Duden-Redaktion. 1992. Redewendungen und sprichwörtliche Redensarten. Duden Band 11. Mannheim: Dudenverlag. Kujamäki, P. 1998. Deutsche Stimmen der Sieben Brüder. Ideologie, Poetik und Funktion literarischer Übersetzung. Frankfurt a.M.: Peter Lang. Kinnunen, T. 2006. Käännös oikeudenkäyntiaineistonan [Translation as a source of law – a study of legal concepts and their translation]. Tampere: Tampere University Press. La Roche, W. v. und Buchholz, A. (Hg.). 2004. Radio-Journalismus. Ein Handbuch für Ausbildung und Praxis im Hörfunk. 8. Auflage. Schopp, J. 2002. „Übersetzungsübung und Übersetzungsauftrag.“ In Erikoiskielet ja käännöstoria [Fachsprachen und Übersetzungstheorie], 344–358. Vaasa: VAKKI:n julkaisut [Publikationen der VAKKI-Studiengruppe] No. 29.

Online-Quellen Belege für Übersetzungen des Parteinamens „Perussuomalaiset“: – Der Standard, Österreich: http://derstandard.at/ – Antifaschistisches Dokumentationsarchiv, Mainz: http://dokmz.akdh.ch/index.php – Wikipedia, die freie Enzyklopädie, Stichwort „Kurikka“: http://de.wikipedia.org/wiki/Kurikka – Finnische Botschaft in Berlin: www.finnland.de – Statistik Finnland, deutschsprachige Seiten: www.stat.fi – Auswärtiges Amt Berlin (Länderinfos): www.auswaertiges-amt.de/diplo/de/Laender/Finnland.html [alle abgerufen 2/2007]

Assessment and training

Magnifying glasses modifying maps A role for translation theory in introductory courses Heloísa Pezza Cintrão

University of São Paulo, Brazil

There has been some debate about the usefulness of translation theory in initial levels of translator training. Here we consider translation theory as a form of declarative knowledge abstracted from translation practice and empirical studies. We report on a study in which the data suggest that translation theory can have a positive impact on the performance of undergraduate participants in an introductory translation course. Keywords: development of translation competence, declarative knowledge, translation learning, empirical Translation Studies.

1. Introduction In 2000, Orozco concluded from her doctoral thesis that the data for her study on the development of translation competence had not shown any necessary correlation between knowledge about translation, analyzed in students’ responses to a questionnaire, and these same subjects’ efficiency in a translation task that they had carried out (Orozco 2000:€240–241). Perhaps this has been an important reason why, up to (and including) its 2003 version, the translation competence model proposed by the PACTE group, which was also developed at the Autonomous University of Barcelona, emphasizes that translation competence (herein referred to as TC) is predominantly procedural knowledge, though considering it as a type of expert knowledge which, as such, is characterized by the joining together of declarative and procedural knowledge: Translation competence is the underlying system of knowledge needed to translate. It includes declarative and procedural knowledge, but the procedural knowledge is predominant. It consists of the ability to carry out the transfer process from the comprehension of the source text to the re-expression of the target text, taking into account the purpose of the translation and the characteristics of the target text readers. It is made up of five sub-competencies (bilingual,

 Heloísa Pezza Cintrão

extra-linguistic, knowledge about translation, instrumental and strategic) and it activates a series of psycho-physiological mechanisms. (PACTE 2003:€58)

However, instead of assuming that declarative and procedural knowledge come together in TC as clearly separate components, from the point of view of translation training it would perhaps be more worthwhile ask how different types of declarative knowledge can interact with procedural knowledge in the TC development process. This may allow greater clarity concerning the possible impacts of declarative knowledge on the procedural knowledge in the translator’s performance and in the TC development process.1 In this article, an analysis of data from an empirical study on the development of TC will be presented. The study involved the design of an introductory translation course of twenty-eight hours. Longitudinal data were collected to observe the effects of this course on the participants’ translation performance. The data analysis to be presented here focuses on functional problems in translation tasks carried out by the subjects of the experiment and two control groups. In the course we opted for combining translation tasks performed by the students and case analyses with the explicit presentation of a set of concepts and principles selected from translation theory. This was despite controversies over the efficacy of the explicit presentation of theoretical concepts and proposals in introductory courses. The results of the data analysis suggest that the procedures adopted in the course had rapid and positive effects on the course participants’ ability to detect functional translation problems and on the quality of solutions they found for these problems. As a rationale for choosing this course design, we will mainly consider Chesterman’s (2000) and Shreve’s (1997) proposals on TC development. Building on these authors, we argue that a series of important factors for detecting and solving problems and for making translation decisions will not necessarily be grasped spontaneously from practical experience if these factors are not focused on so as to be consciously perceived by a subject by means of the explicit presentation of translation concepts and meta-reflection. These forms of declarative knowledge can work in the development of TC as a sort of ‘magnifying glass,’ allowing the learner to see what could, without the tools provided by conceptual and metacognitive work, go unnoticed indefinitely. In turn, the perception of crucial factors for detecting and solving problems and for making decisions will favor a ‘modification of maps’, that is, the transformation of simplistic mental schemes about the nature of translation into more complex ones, providing a more flexible ability to deal with different types of problems and to find more appropriate solutions to the functional requirements of a translation task. We trust this paper will make the following contributions: (1) provide evidence of one kind of impact of knowledge about translation sub-competence on 1. I tried to make this point in a conversation with the head researcher for the PACTE group, Prof. Amparo Hurtado, during her course at the Federal University of Minas Gerais, in Brazil, August 2004.



Magnifying glasses modifying maps 

the development of the strategic sub-competence of TC; (2) provide some empirical evidence of an aspect in which the development of TC has a considerable degree of independence from bilingual sub-competence; (3) present a possible method for measuring functional components in the analysis of translation products, devised for empirical studies on TC and its development.

2. Knowledge about translation in introductory courses The role of Translation Studies in introductory courses by means of explicit presentation of theory has created diverging stances between two specialists in translator training, Hurtado and Chesterman. In 1996, Hurtado presented a proposal for an introductory course on written translation into the mother tongue. She defined introductory courses as a “teaching space which allows students to grasp the fundamental principles that guide translation practice and to undertake appropriate working methods for dealing with the many specialized fields in translation later on” (p.€31, our translation). With regard to the course bibliography, she defended the idea that it should be made up of documentation sources of a general nature, such as grammar books, bilingual and monolingual dictionaries, encyclopedic dictionaries and so on. Publications on translation theory should be restricted to the course “Translation Theory,” to be taught later on (at her university, after the first year). “Bibliography on ‘theory’ finds its place in […] the course on ‘Translation Theory’” (Hurtado 1996:€37 note 9, our translation). Hurtado’s 1996 viewpoint about the role of translation theory in introductory courses seems to be related to what she states in a later book: Translatology (Traductología) consists of knowledge about something, that is, a type of knowledge called ‘declarative knowledge’ (Anderson 1983), characterized by being easily verbalized, acquired at once (through exposition), and essentially consciously controlled. Translating (Traducción), by contrast, is a know-how, that is, a type of ‘procedural knowledge,’ which is difficult to verbalize, is gradually acquired by means of practice and is essentially automatic. From this point of view, Translatology (Traductología) and translating are intrinsically different in nature (Hurtado 2001:€25). Chesterman (2000) takes a different stance concerning the explicit presentation of theoretical elements at introductory levels. Like Hurtado (2001) and PACTE (2003), he classifies TC as an ‘expert knowledge’ that is defined, following the Dreyfus & Dreyfus model, as learned knowledge whose development is not spontaneous and which goes through five stages. Unlike Hurtado (1996), however, Chesterman states that during the first (or novice) stage, the learner should not only be offered concrete instruments such as dictionaries, but also what he calls “conceptual tools,” that is, strategies and “key concepts” or “the most basic concepts” (ibid.: 80). Thus, Chesterman seems to attribute an important role to declarative knowledge at the novice stage, in which concepts “relevant to the skill in question” (ibid.: 77) should be explicitly

 Heloísa Pezza Cintrão

presented to the learner: “First, we need to teach certain things quite overtly”; “At the beginning, certain rules and concepts have to be learned consciously” (ibid.: 80); “At the novice stage in the growth of expertise [...] we need an explicit presentation of key concepts” (ibid.: 83). Whether done in an implicit or explicit, inductive or deductive manner, Hurtado and Chesterman agree that there should be a focus on fundamental principles in an introductory course. And both point out that there is not complete agreement in the field of Translation Studies with respect to what these principles might be. Hurtado seems to assign an important role to the Skopos rule set forth by Reiß & Vermeer (1984). Chesterman suggests that basic concepts traditionally central to theory be selected, especially those related to the beliefs that laypeople hold about translation, such as concepts like “source” and “target,” of different types of equivalence, of (un) translatability. Moreover, translation strategies, defined as “potentially conscious, goaloriented procedures for solving problems” (Chesterman 2000:€82), should be taught. Chesterman proposes a basic set of strategies, built on Vinay & Darbelnet (1958), Catford (1965), and Van Leuven-Zwart (1989, 1990). As for empirical evidence for or against these viewpoints, we can go back to the experimental study done by Orozco (2000) for her doctoral thesis. Orozco built three instruments for measuring TC. One of them, the “translation problems measuring instrument” (TPI) should measure, through the analysis of a translation done by the students, their skill in detecting different types of translation problems and in applying strategies to solve them. In the 2003 TC model by the PACTE group (of which Orozco was a member), the detection of problems and the application of strategies are presented as characteristic tasks of what is called “strategic sub-competence,” classified by PACTE as a type of predominantly procedural knowledge: Strategic sub-competence. Procedural knowledge to guarantee the efficiency of the translation process and solve the problems encountered. This is an essential subcompetence that affects all the others and causes inter-relations amongst them because it controls the translation process. Its functions are: (1) to plan the process and carry out the translation project (choice of the most adequate method); (2) to evaluate the process and the partial results obtained in relation to the final purpose; (3) to activate the different sub-competencies and compensate for deficiencies in them; (4) to identify translation problems and apply procedures to solve them. (PACTE 2003:€59)

Another instrument developed by Orozco was the ‘general concept of translation measurement instrument’ (GCTI), which should assess conceptions of translation and of TC by using student answers to a questionnaire. Considering PACTE’s TC model, GCTI is clearly linked to the “knowledge about translation sub-competence”, classified as declarative knowledge that can be implicit or explicit: Knowledge about translation sub-competence. Predominantly declarative knowledge, both implicit and explicit, about what translation is and aspects of the



Magnifying glasses modifying maps 

profession. It includes: (1) knowledge about how translation functions: types of translation units, processes required, methods and procedures used (strategies and techniques), and types of problems; (2) knowledge related to professional translation practice: knowledge of the work market (different types of briefs, clients and audiences, etc). (PACTE 2003:€59)

Orozco applied these instruments twice, within an eight-month interval, to 97 firstyear undergraduate students in Translation courses at three Spanish universities, and she made tests to measure the correlation between TPI and GCTI, which seems to be equivalent to observe the correlation between the ‘strategic sub-competence’ and the ‘knowledge about translation sub-competence’. Her findings were that there was no significant correlation between TPI and GCTI in either of the two measurements. From this finding we can suppose that the impact of translation theory in introductory courses should be very limited or even non-existent.

3. The impact of declarative knowledge at the ‘novice’ stage The experimental study presented in this article is connected with this debate. For the study, an introductory course in general written translation into the mother tongue in the Spanish-Portuguese language pair was devised. It connected systematically practice activities with explicit presentation of theoretical elements by means of ‘prototypical tasks and cases’. This should make especially clear the relevance of the concepts to be discussed, in an approach that we believe to be in line with Chesterman’s proposals regarding the explicit offer of conceptual or theoretical tools. Basically, this introductory class was aimed at deconstructing simplistic conceptions of translation as mere transference between codes at word level, favoring, in contrast, the construction of a more complex understanding of the translator’s task, using the integration of functional, discursive and cognitive concepts and proposals. The functional considerations included the concept of functional units (Nord 1998) as well as the Skopos rule (Reiß & Vermeer 1984) and the consequent need to consider the audience when making decisions. Included among the discursive concepts were considerations of dialect and register (Hatim & Mason 1990), cohesion and coherence, genres and text types, and relationship between language, culture, and worldview (or values). Among the cognitive aspects, we worked on issues such as awareness of our own inferential processes and problem-solving processes (Alves et al.€2000). Thus, the course was designed to be a learning space in which the participants should grasp the “basic principles which guide translation, and begin to develop appropriate translation working methods” (Hurtado 1996:€ 31). As for the main “basic principles which guide translation,” the principles explicitly presented and discussed were the Skopos rule, the essential link between language and culture (and cognition, Snell-Hornby 1988), and the concept of translation as communicative act of intercultural mediation. These were considered to

 Heloísa Pezza Cintrão

be fundamental principles for decision-making in translation and, therefore, guiding principles for the operation of strategic sub-competence, during an adequately executed translation process. The functional and (con)textual aspects of the translation were emphasized. Materials and translation tasks capable of demonstrating, in practice, the relevance and applicability of these principles and concepts were used (and were called, for this reason, ‘prototype materials’) (see Cintrão 2006). The course was given to eight undergraduate students who were, at that time, finishing a course in basic-level Spanish as a foreign language. The native language of all the participants was Brazilian Portuguese and they were all taking classes of Spanish as a foreign language at the same university. A prior interview and a profile questionnaire guaranteed that none of them had received classes of Spanish beyond a basic level before entering the university. In those aspects, they shared the same profile as a further seven undergraduates who comprised one of the control groups for the experiment. Four translations were done individually by the eight participants at different stages of the course outside the classroom. These translations composed a corpus for studying the effects of the training. The seven students in the control group did the same translations as the participants at the same moments. The translated source texts were all children’s stories by the Argentinian writer María Elena Walsh. The translation tasks giving the longitudinal corpus were as follows: Code

Source texts (children’s stories by Walsh)

T1 T2 T3 T4

“Historia de una Princesa” “El patio” “La Luna y la Vaca” “Historia de una Princesa”

Characters (with spaces)

Direction (Sp > Pt)

5028 3262 4170 5028

L2>L1 L2>L1 L2>L1 L2>L1

Moment before the course in the middle of the course after the course a few days after T3

A second control group was composed of six subjects with a university degree in Spanish (Language and Literary Studies), all of them with experience as Spanish teachers. These bilinguals only did the first translation task. The student control group was formed to isolate possible improvements in translation performance ascribable mainly to the progress in foreign-language learning and to classes on literature and linguistics taken at the university during the period of the experiment. The students in the two groups were instructed to restrict themselves to the undergraduate Spanish classes during the period of the experiment, that is, to not take any extra classes of Spanish. The students in the control group did not have any kind of training in translation. All the students were also asked never to talk with other volunteers about the individual translations done for the corpus collection. The translations done for the corpus were never commented on or discussed in the translation course. The control group of bilinguals should provide a further parameter of comparison for the performance in translation. They had different degrees of professional



Magnifying glasses modifying maps 

experience as translators but none had translation as their main activity. They did not have training in translation. This design of groups isolated the bilingual sub-competence in the form of expertise in language. We think it allows us to suggest that sensitivity to functional parameters in translation has a significant degree of independence from the bilingual sub-competence, can be improved faster than the competence in the foreign language, and receives significant impact from explicitation of fundamental concepts of translation. The translation collected in the middle of the course (T2) should allow us to assess the effects of the awareness of translation processes stimulated in the second part of the course. For the study presented here there was no interest in observing this component of the course in isolation; this second translation was thus not analyzed. For the corpus collection, the subjects came one by one to individual sessions and did the translation without time pressure. The translations were written in the same office, in the same computer, using the program Translog 2000 (Jakobsen 1999). The subjects were asked to translate for a target audience of Brazilian children. The same set of documentation sources on paper were available for all the subjects in each task and they also had free access to the Internet. Once the translation was finished, an interview was recorded where the subjects explained the reasons for their pauses and corrections, as registered by Translog. No value judgments were made about the translation solutions and the subjects knew that the researcher should not reply to any question of this nature. Each translation session allowed us to collect a text file containing the target text, a log file with real-time process data provided by Translog, a recorded interview discussing the reasons for pauses, corrections and decisions. Below we present a product analysis of passages selected from the target texts for T1, T3 and T4. In this analysis, two procedures for measuring the development of TC were applied, both devised with the intention of observing functional and (con)textual aspects of strategic sub-competence. We believe the results suggest that, at the novice stage, the explicit presentation of certain concepts and theoretical principles “relevant to the skill,” to quote Chesterman (2000:€77), contributes to a change in conceptions of translation, that is, a change in the students’ mental schemes about translation. This change was manifested in a significant improvement in the functional aspects of the strategic sub-competence, specifically in its nuclear tasks of detecting and resolving translation problems. The focus on functional and contextual aspects for establishing the indicators for our data analysis was due to the fact that several studies, such as those done by Toury (1986), Jääskeläinen (1989), and Halliday (2001), many of them empirical, coincide in pointing to the hierarchical prevalence of equivalence at functional and contextual levels as one of the most important operative principles in translation. This principle is deemed relevant for achieving quality in the end product, and is one of the most notable factors that differentiate the translational capabilities of experts and from those of novice translators. This is why we considered the ability to apply a functional and contextual hierarchy when making decisions to be an indicator of progress in the

 Heloísa Pezza Cintrão

development of TC, especially in the case of translation problems caused by a conflict between referential equivalence and functional or (con)textual appropriateness. The element observed when examining the data was, therefore, the ability to detect and solve functional and (con)textual problems, mainly from the perspective of the changes in this ability among course participants after four months of training. In order to operationalize this indicator, the translations for eighteen passages selected in the stories “Historia de una Princesa” (T1 and T4) and “La Luna y la Vaca” (T3) were considered. The selected passages presented problems whose solution depended on making decisions considering functional and (con)textual issues. In nine of these passages, in view of the translation brief, rhymes were supposed to receive a high equivalence value, superior even to the accuracy of the meaning of the rhyming words, as in the example “Los dos llegaron al templo en monopatín y luego dieron una fiesta en el jardín, una fiesta que duró diez días y un enorme chupetín”. Three of the segments contained cultural elements that, considering the audience and the purpose of the translation, required cultural adaptation as in, for instance, “... bailaron la rancherita y el pericón”. Six passages exhibited peculiar manners of speech, characterizing certain characters in the stories, as in “nopo puepedopo” (instead of ”no puedo”)2. Table 3 (appendix) shows an analysis of the translations of these eighteen passages, referred by numbers on an Excel spreadsheet. The translation of each passage by each subject was given a score in order to allow for quantification of the results, thus arriving at a percentage by group and task. In this table we register absence (zero) or presence (one) of indications of a positive response to functional appropriateness requirements in each passage, considering positive response indicators to be (a) any attempt to recreate the rhymes, (b) any application of a cultural filter (House 1981) in the case of cultural elements, (c) any attempt to recreate a distinctive manner of speech for the Princess and the Cow characters, in the case of playful children’s languages. The result was thus a binary evaluation. Table€1 presents the scores in percentages for each subject in the binary evaluation. In another Excel spreadsheet, a measurement of the quality of the solutions were also made by scoring each passage on a 0 to 3 scale instead of using binary values. For this qualitative assessment, zero kept on corresponding to the lack of any attempt to deal with the functional appropriateness requirement of the selected passage, according to the same criteria as for the binary evaluation (e.g.€no attempt to recreate rhymes in a rhymed passage). The score of one was given to an attempt to deal with the functional problem but with questionable results. The score of two was given to acceptable but not totally appropriate solutions. The maximum score of three corresponded to a fully satisfactory solution. Table€2 presents the qualitative scores in percentage for each subject.

2. A complete relation of the eighteen selected passages can be seen in Cintrão 2006:€450.



Magnifying glasses modifying maps 

Table 1.╇ Binary evaluation: scores per subject Bilingual control group

T1 (%)

P1

P2

P3

P4

P5

P6

80

40

70

100

80

50

S1

S4

S6

S7

S8

S11

S12

S13

â•⁄ 50 100 â•⁄ 70

40 88 90

30 63 90

10 63 10

40 38 70

70 63 80

50 50 70

10 13 60

S3

S5

S9

S10

S14

S15

S16

50 50 80

10 38 â•⁄ 0

60 75 90

80 25 80

10 â•⁄ 13 60

â•⁄ 0 13 20

50 25 70

Main student group

T1 T3 T4

Control student group

T1 T3 T4

Table 2.╇ Qualitative evaluation: scores per subject Bilingual control group

T1

P1

P2

P3

P4

P5

P6

47

13

43

90

63

37

S1

S4

S6

S7

S8

S11

S12

S13

40 75 47

17 67 53

20 42 67

â•⁄ 3 42 â•⁄ 7

20 33 40

40 38 47

23 38 47

â•⁄ 7 13 37

S3

S5

S9

S10

S14

S15

S16

37 42 50

â•⁄ 7 38 â•⁄ 0

30 58 43

40 13 60

â•⁄ 7 â•⁄ 4 30

â•⁄ 3 13 â•⁄ 7

17 21 43

Main student group

T1 T3 T4

Control student group

T1 T3 T4

 Heloísa Pezza Cintrão

The groups’ average scores (Graphs 1 and 2) for the tasks show the changes in the performance of the students from the first translation to the two translations done between three and four months later, one of them (T4) being a re-translation of the same text used for the first one.

80 70 60 50 40 30 20 10 0

70

Functional problems detection (%) (Binary evaluation) 59 37

PC

38

57

32

SC SM

SC SM

T1

T3

T1

68

SC SM T4

Graph 1.╇

80 70 60 50 49 40 30 20 10 0 PC T1

Solutions assessment (%) (Qualitative evaluation) 44 20

21

27

SC SM

SC SM

T1

T3

33

43

SC SM T4

Graph 2.╇

For the first translation of the story “Historia de una Princesa” (T1), the bilingual group of graduates attained an average score of 70% in detecting the higher hierarchical importance of the functional dimensions for this translation brief, while both groups of students achieved 37 and 38%. The bilingual control group (P) only performed task



Magnifying glasses modifying maps 

T1, and their score of 70â•›% on this first translation will be compared with the two final tasks (T3 and T4) performed by the students groups (Control Student=SC and Main Student=SM), for reference. After the course was over, the participants (SM) were able to detect 68% of the functional problems in the re-translation and 59% when translating a new story, showing a considerably stable performance in these two different cases. The group that did not attend the course (SC) achieved 57% in the re-translation of the source text they already knew (T4) and 32% in the translation of the new story (T3). It is fairly probable that their performance when re-translating “Historia de una Princesa” improved due to the analytical attitude they had to adopt toward this text during the interview carried out for T1. On the other hand, Jääskeläinen (1986) pointed out that when subjects have to deal with a text and a task which raise problems related to function and audience, their awareness of contextual issues may spontaneously increase. However, unlike the students who had participated in the translation course, the control group did not seem able to repeat the improved performance for T4 when dealing with a new text, and their score for functional problems detection is worse in T3 than in the first task T1 (See Graph 2). In the qualitative evaluation, the two student groups improved their performance after some months. This improvement is significantly higher for the main group, which, once more, almost reached the same average score as the proficient bilinguals had attained in T1. This makes us think that it is necessary to consider carefully what kinds of analytical attitudes could be used in translator training, what ways practical activities can be organized, and what combinations of activities and reflections are insightful for learners. That is, when discussing translation training, it seems necessary to consider different types of declarative and procedural knowledge, and different interactions between them. Is some kind of declarative knowledge perhaps involved in practical activities that entail an analytical attitude and a conscious effort to explicate parameters for decisions? Were the students in the control group working on some kind of declarative knowledge in the interviews while explaining their choices and trying to verbalize their criteria? For this study, however, the specific declarative knowledge we intended to observe comprised the explicit presentation of concepts and principles taken from translation theory. From this point of view, the findings of our analysis seem to be different from (but perhaps not incompatible with) those observed by Orozco. Our results seem to suggest that declarative knowledge, acquired through explicit presentation of knowledge about translation, has a considerable impact on the strategic sub-competence of novices. This impact is manifested in our data in a short-term improvement in the detection and solution of functional problems in the subjects’ translations into their mother tongue. Over approximately four months of an introductory course, this functional aspect of strategic sub-competence improved in those beginner foreignlanguage learners to the point that they almost matched the results attained by the group of proficient bilinguals. This seems to us to be unthinkable for predominantly

 Heloísa Pezza Cintrão

procedural knowledge, a type of knowledge that is supposedly acquired more slowly and gradually, as in the case of the communicative proficiency in a foreign language.

4. Trying to explain the experiment results To look for possible causes of these results, let us focus on the performance of one student from the control group (S10) in the translation of passage 1.4 of “Historia de una Princesa”, which contains rhymes. In the first task (T1), student S10 had already decided to give precedence to the rhyme. Instead of the semantic translations ‘monopatín’ (Argentinean Spanish, meaning ‘scooter’) → ‘patinete’ (Brazilian Portuguese), ‘chupetín’ (AS, meaning ‘lollipop’) → ‘pirulito’ (BP), she translated ‘monopatín’ as ‘patim’ (‘roller skate’, Spanish ‘patines’), and ‘chupetín’ as ‘quindim’ (a typical Brazilian sweet), in order to rhyme with ‘jardim’ (‘garden’, ‘jardín’ in the source text). In the interview after this first translation (T1) she stated that she was very satisfied with this solution. However, in the re-translation (T4), four months later, though her performance at the product level remained the same for this passage, she was no longer happy with that solution. In the interview for the re-translation (T4), the student said (our translation): – Well, this time... I wasn’t as happy as I was the first time […] With the last passage, you know [...], because... Well, I really wondered: can I do this? I was also a little frustrated. – If you could change it to ‘quindim?’ ‘Quindim,’ ‘jardim.’ In the first translation you were happy with that. – Yeah, because... It was something like… ‘Wow! I made it rhyme,’ you know. [...] But now, I am so... confused [laughs]. [...] Lots of questions, and... no answer. – What, then? The question of whether you could change it… – Yeah.

Thus, the confidence of this student who had not taken the pilot course ‘worsened’ as time went by. A few days before her re-translation of “Historia de una Princesa”, this insecurity had negatively affected her performance in the translation of passage 4.2 of “La Luna y la Vaca”. S10 had translated the passage “bailaron la rancherita y el pericón” (“they danced the rancherita and the pericón”) as “dançaram a rancherita e o pericón,” maintaining the names of the Argentinean dances. In the interview which immediately followed, I asked: “Are you happy with your solution for the dances?” And the student responded: “No. Not at all. But, well, it’s hard to say. Could I change them?” The data from the verbal protocols of this student give us some clues for understanding the way in which knowledge about translation can change the mental schemes that guide decision-making in the translation process. In view of the data presented, our proposal is that explicit presentation of concepts and principles in the initial stages



Magnifying glasses modifying maps 

of learning can work as a ‘magnifying glass,’ lending visibility to the fundamental mechanisms of translation, such as, for instance, the different levels of equivalence or correspondence that can be given precedence in different tasks, in relation to the textual type, audience and intended effects. The mere presence of an important parameter for decision-making in a translation task does not mean that an individual will always detect it and/or recognize its importance spontaneously, since, as pointed out by Shreve (1997), expectations and objectives play a critical role in the perceptive capacity of a subject or organism. A certain type of mental schema concerning translation leads many novices to have difficulty in producing “culturally appropriate stylistic and text-typological markers in the target text” (Shreve 1997:€135). The development of the knowledge about translation sub-competence by means of the explicit presentation of theoretical concepts and principles “relevant to the skill” (in the words of Chesterman) seems to be capable of qualitatively modifying, quite quickly, the ability to detect problems and to apply strategies for resolving problems, “pushing” (to borrow an image from Toury 1986) the segmentation of translation units towards more macro levels and “pushing” the establishment of equivalence towards more complex linguistic levels than those of lexical meanings, which is an important factor in the development of TC. Is it possible to arrive at this same qualitative change in mental schemata and at subsequent improvements in functional performance without the explicit presentation of concepts and principles? In many cases, the answer is probably yes. However, it is worth mentioning that an experimental study by Sun & Sluzarz (2001) regarding the interaction between implicit and explicit learning (in general terms, not specifically in translation) pointed out that subjects to whom the rules of a task are explicitly presented improve their performance more rapidly when compared with subjects that need to deduce the rules using pure repetition of a practical task, or when compared to subjects who are requested to infer the underlying rules of a task using pure practice (a group of subjects who underwent this last situation in an experiment conducted by these authors showed worse performance as time went on, at least in the short term, in the case of complex tasks).

Acknowledgements FAPESP – Foundation for Aid to Research of the State of São Paulo financed my participation in the 5th EST Congress. José Luiz Vila Real Gonçalves referred me to the text by Sun & Sluzarz. Sarah Johnson translated the first version of this article into English and Marielle Godbout helped us with a proofreading. The suggestions by the editors and the specially detailed comments and suggestions made by Anthony Pym resulted in significant differences in this final version as compared to the text originally accepted for publication. Christopher Scott-Tennent carried out a final proofreading.

 Heloísa Pezza Cintrão

References Alves, F., Magalhães, C., Pagano, A. 2000. Traduzir com autonomia. São Paulo: Contexto. Cintrão, H.P. 2006. Colocar lupas, transcriar mapas. Iniciando o desenvolvimento da competência tradutória em níveis básicos de espanhol como língua estrangeira. PhD Thesis. São Paulo: FFLCH – Universidade de São Paulo. Available at Chesterman, A. 2000. “Teaching strategies for emancipatory translation.” In Developing Translation Competence, Ch. Schäffner and B. Adab (eds), 77–89. Amsterdam/ Philadelphia: John Benjamins. Halliday, M.A.K. 2001. “Towards a theory of good translation.” In Exploring Translation and Multilingual Texts: beyond content, E. Steiner; C. Yallop (eds), 13–18. Berlin/New York: Mouton de Gruyter. Hatim, B. and Mason, I. 1990. Discourse and the Translator. London: Longman. House, J. [1977] 1981. A Model for Translation Quality Assessment. Tübingen: Narr. Hurtado, A. 2001. Traducción y Traductología. Madrid: Cátedra. Hurtado, A. 1996. “La enseñanza de la traducción directa ‘general’. Objetivos de aprendizaje y metodología.” In La enseñanza de la traducción, A. Hurtado (ed.), 31–55. Castelló de la Plana: Universitat Jaume I. Jääskeläinen, R. 1989. “Translation assignment in professional vs. non-professional translation: a think-aloud protocol study.” In The Translation Process, C. Séguinot (ed.), 87–98. Toronto: H. G. Jakobsen, A.L. 1999. “Logging target text production with Translog”. In Probing the Process in Translation: Methods and Results. G. Hansen (ed.), 9–20. Copenhagen: Samfundslitteratur. Nord, C. 1998. “La unidad de traducción en el enfoque funcionalista.” Quaderns. Revista de Traducció 1: 65–77. Orozco, M. 2000. Instrumentos de medida de la adquisición de la competencia traductora: construcción y validación. PhD thesis. Barcelona: FTI – Universidad Autónoma de Barcelona. PACTE. 2003. “Building a translation competence model.” In Triangulating Translation, F. Alves (ed.), 43–63. Amsterdam/Philadelphia: John Benjamins. Reiß, K. and Vermeer, H. J. [1984] 1996. Fundamentos para una teoría funcional de la traducción. trad. S. García Reina; C. Martín de León; H. Witte. Madrid: Akal. Shreve, G.M. 1997. “Cognition and the evolution of translation competence.” In Cognitive Processes in Translation and Interpreting, J. H. Danks; G. M. Shreve; S. B. Fountains; M. K. McBeath (eds), 120–136. Thousand Oaks: Sage. Sun, R. and Sluzarz, P. 2001. “The interaction of explicit and implicit learning: An integrated model”. Available at Last access: Dec. 2007. Snell-Hornby, M. 1988. Translation Studies: an integrated approach. Amsterdam/ Philadelphia: John Benjamins. Toury, G. 1986. “Monitoring discourse transfer: A text-case for a developmental model of translation.” In Interlingual and Intercultural Communication. J. House and S. Blum-Kulka (eds), 79–94. Tübingen: Narr.

Appendix: Table€3.╇ Binary evaluation

Magnifying glasses modifying maps 

Effects of short intensive practice on interpreter trainees’ performance Magdalena Bartłomiejczyk University of Silesia, Poland

The performance of ten interpreting students participating in a one-week practice programme in real conference conditions was measured using two tools established in research on interpreting: propositional accuracy scores and error analysis. These tools were applied to the students’ interpretations recorded in an experimental setting before and after the practice programme. The results suggest a slight average improvement in the students’ interpreting skills, yet none in their delivery. The study has some important practical implications on how the practice programme will be run in the future. Keywords: interpreter training, practical experience, error analysis, propositional accuracy, completeness

1. Introduction This paper is meant to demonstrate the practical usefulness of interpreting research by providing an example of how research tools applied to examine the quality of interpreted texts can contribute to the evaluation and development of teaching methods in interpreter training. The method in question is a short intensive practice programme in real conference conditions undertaken by interpreter trainees approximately in the middle of their 3-year-long translation and interpretation course. The main goals of this research project were to: 1. evaluate in an objective way, as precisely as possible, the effects of an intensive practice programme organised by the Institute of English (University of Silesia) in cooperation with the United Nations Office in Vienna on the performance of the participating trainees; 2. examine which quality aspects of the trainees’ output improved most considerably and which (if any) deteriorated after the practice; 3. gauge which students are likely to profit the most from such a practice.

 Magdalena Bartłomiejczyk

The findings of this study will have direct practical implications for the organisation of similar training activities in the future. After returning from Vienna, the subjective feelings of both the author and the participants were that the practice had considerably enhanced the trainees’ performance. However, there was no hard data to support this. As sending trainees for such practices requires a lot of organisational effort and is connected with considerable costs for the Institute as well as the trainees themselves, it is hoped that the research tools applied should help demonstrate that the performance of the participants actually improved as a result of intensive practice in real conference conditions. Otherwise, the effectiveness and, consequently, the value of similar practices might be called into doubt. Secondly, if any quality aspects of the participants’ interpretations are shown to have failed to improve or even to have deteriorated after the practice, this might suggest modifications to certain elements of the practice (such as preparation before the conference, trainer and peer feedback during the conference, follow-up etc.). Last but not least, potential candidates for such a practice programme considerably outnumber available booth places, which necessitates applying selection procedures. It would therefore be useful to find some indications as to which students are likely to draw the most profit from their participation. For example, if it is shown that the performance of students who had been relatively good at interpreting to begin with improved more than that of poorer interpreters, this would provide arguments for an assessment-based selection procedure rather than drawing lots or accepting candidates on a “first come, first served” basis.

2. Quality assessment in simultaneous interpreting Assessment of the quality of interpreting largely escaped scholarly interest until the early 1980s; however, it belongs to the group of top issues addressed by current conference interpreting research (Gile 2003). This is understandable, as quality assessment is crucial for achieving adequate quality levels in experienced professionals as well as in fresh interpreting school graduates. Ensuring this quality, in turn, is obviously in the best interest of customers and interpreters alike. Beyond any doubt, quality assessment is a very complex task which can be looked at from a variety of angles and carried out using a wide range of methods (see e.g.€Kurz 2001, Pöchhacker 2001, Gile 2003, Kalina 2005). The format of this paper does not allow me to elaborate on this, so I will limit my discussion to two assessment tools which were employed in this project, as will be described later on. The first tool, known as “propositional analysis” or “completeness score”, was originally developed by Kintsch 1974 to investigate recall of text content, and its practical application was described in great detail by Bovair and Kieras 1985. In interpreting research it has been used in a relatively small number of studies, for example, by Dillinger 1994 and Tommola and Helevä 1998. It focuses exclusively on the semantic aspect of interpretations and involves creating a text base for the source text, where



Effects of short intensive practice on interpreter trainees’ performance 

each propositional element is given a score value. Afterwards, the interpretations of the source text are analysed against the background of this text base: credit is given for each element which is present in both the interpretation and the source text. For a broad and deep discussion of theoretical and practical problems connected with using this tool in the context of interpreting research, see Tommola 2003. Tommola and Helevä 1998 investigated the effects of source text complexity and interpreting direction on interpreter trainees’ performance by means of propositional accuracy scores. For their experiment, they constructed sets of texts in English and in Finnish with varied surface structure complexity. The scores were significantly higher for simple texts than for complex ones (52% and 42%, respectively), and the study revealed a slight trend in favour of interpreting into B in the case of complex texts. Dillinger 1994, on the other hand, used a more complex scoring system than just 0 for a missing element and 1 if it was present. He marked 0 for omission, 1 for a semantic change, 2 for a paraphrase and 3 for word-for-word translation. This raises some serious objections, as changing the meaning of a proposition is often a much graver error than omitting it altogether, and word-for-word translations tend to be stylistically worse than paraphrases. This system probably explains his surprising results suggesting that experienced professionals are not markedly better at interpreting than complete beginners (bilinguals): the difference between the two groups amounted to only 17%. The above mentioned case is just one illustration showing that this tool must be treated with caution as a means of assessing the quality of interpretations. We must remember, for example, that in typical propositional accuracy scores meaning errors are not penalised with negative points, and in the case of incomplete sentences credit is still given for elements which are present even if the sentence as a whole is incomprehensible. Moreover, SI aims to transmit gist information and interpreters often omit certain peripheral elements (such as intensifiers or relative clauses) intentionally to cope with time constraints or because they see them as redundant. Therefore, it might make sense, as suggested by Tommola and Helevä 1998, to give different scores to superordinate and subordinate propositions. The other tool of interest here is more popular and probably does not need a detailed introduction. It is error analysis, where attention can be devoted to various aspects of the interpreting output: content as well as lexical and grammatical correctness, register, and numerous features affecting presentation (such as voiced hesitations or false starts). Barik 1975 and Kopczyński 1980 are good examples of this tool’s practical application. Barik 1975 introduces a sophisticated classification of “translation departures”, which fall into three main categories: omissions, additions and substitutions. Each of these categories comprises several subcategories. This classification, however, focuses exclusively on the semantic aspect. Moreover, it does not recognize differences in standards applied to conference interpreting versus written translation. Consequently, it was met with widespread criticism. Kopczyński 1980, on the other hand, proposes a

 Magdalena Bartłomiejczyk

classification of errors which seems more appropriate for conference interpreting. It contains five categories of errors: errors of competence, errors of performance, omissions and additions, errors of appropriateness and, lastly, the most serious type: errors of translation. Both propositional accuracy and error analysis have their inherent weaknesses. It seems that more reliable results can be obtained by combining the two tools, as was done in the study described in Färber 2002 and Kurz and Färber 2003. This study focused on anticipation in interpreting from German into English. One of the goals was to see whether the subjects’ use of anticipation correlated with selected quality criteria: completeness scores, grammar scores and the number of errors of two types: communicative inaccuracies and flaws in delivery. Completeness scores were calculated making a distinction between ‘core elements’ (sentence subject, verb and complement) for which 5 points were awarded, and ‘additional elements’, for which 1 point was awarded. The average completeness score was over 68%. Communicative inaccuracies were defined as translation errors such as incorrect terminology, meaning distortions, misinterpreted cohesions etc. Each sentence containing these received from 1 to 3 penalty points depending on the extent to which the original meaning was distorted. Flaws in delivery included repetitions, corrections, false starts, and voiced and silent hesitations. They were penalised with 1 point each. Each sentence was also evaluated for its grammatical correctness and rated with 0 if it was correct, with 1 penalty point if it contained minor grammatical errors and with 2 if it contained serious ones. Quality assessment as carried out in the study described above seems very comprehensive and relatively reliable (although inevitably subjective, as all the grading has to be performed by human referees). Consequently, it was decided that this type of assessment would, to a large degree, constitute a basis for the assessment of the material recorded within the present study.

3. Practice conditions The translator and interpreter programme at the Institute of English is undergoing constant change. It aims at preparing our students in an optimal way for the challenges of the market. Simultaneous interpreting classes take place in a computerised lab with 6 realistic interpreting booths, and trainers try to make as much use as possible of source texts having high ecological validity (authentic conference speeches, parliamentary debates, press conferences etc.). Even so, the artificial conditions of a classroom such as spatial and temporal detachment from the actual event and the absence of an authentic audience are impossible to fully compensate for and do not foster maximum involvement from interpreters. Therefore, in 2006 we established contact with the United Nations Office at Vienna and sent our first group of students for a short interpreting practice in real conference conditions.



Effects of short intensive practice on interpreter trainees’ performance 

Ten interpreter trainees participated in the 49th Session of the Commission on Narcotic Drugs (CND), which took place in March 2006. The conference lasted one week (five working days) with two sessions of 3 hours’ duration each day. Under the supervision of the author, the trainees interpreted from English into Polish in available ‘silent’ spare booths (taking relay from the English booth whenever the input from the floor was in any other UN official language). Although it is difficult to calculate this exactly, the estimated total interpreting time for each trainee was about 150% of one term’s worth of interpreting during regular SI classes, with our students having one SI class (90 minutes) per week over six terms. The students taking part in the practice were not selected on the basis of any criteria other than the fact that they belonged to the same fourth year group and all volunteered for participation, which involved covering costs of a one-week stay in Vienna (the number of available places coincided with the number of students in this group). At the time of the practice, they were approximately in the middle of their SI training, having completed three terms. Preparations for the conference lasted about one month and were coordinated during the group’s regular SI classes as well as during several additional meetings organised for this purpose. Relevant documentation had been made available in advance on the website of the United Nations Office on Drugs and Crime (UNODC), which greatly facilitated the preparations considering the specialized topic of the conference. The students studied the documentation thoroughly and compiled an English-Polish glossary containing all terms which initially posed problems. In addition, the preparation involved creating and interpreting texts dealing with illicit drugs (each student prepared a presentation on the topic to be interpreted by others) and searching for and studying Polish parallel texts. Direct follow-up to the conference included a debriefing session with a discussion of problems encountered and completing the glossary with vocabulary noted during the conference, which referred to both illicit drugs and conference procedures as such.

4. Experiment Two parallel sets of source texts in English (to be interpreted into Polish) were prepared, each of them containing two texts, one on a general topic and one on the topic of the conference, i.e. illicit drugs. General texts were fragments of speeches given by Madeline K. Albright and Jose Manuel Barroso before participants at an international conference organised in Gdańsk in August 2005 to celebrate the 25th anniversary of the trade union “Solidarność”. Specialized texts were fragments of opening addresses by UNODC executive director Antonio Maria Costa presented at CND sessions in 2004 and 2005. Directly before going to Vienna, five participants of the practice were asked to interpret the texts from one set and the other five interpreted the texts from the second

 Magdalena Bartłomiejczyk

set. After returning to Poland, participants were asked to interpret the set which they had not interpreted before. This second experimental session was organised one month after the practice, as I was interested in long-term effects rather than in immediate effects. The interpretations were recorded and transcribed. The initial, optimistic hypothesis was that the interpreting outputs recorded after the practice would prove to be of markedly better quality than the outputs recorded before the practice, and that this superiority would manifest itself in interpretations of both general and specialist texts. I also assumed that the overall improvement in the subjects’ interpreting skills would, above all, be reflected in differences when interpreting general texts.

5. Analysis In the present study, the recorded material is analysed from two distinct perspectives: transfer-related and target-oriented. Firstly, the interpretations are evaluated in terms of propositional accuracy. Secondly, the same interpretations are evaluated independently of the original message in terms of errors of various types they contain. From each source text, a fragment of about 430 words, preceded by a warm-up period of about two minutes, was selected for analysis of corresponding interpretations. The source text fragments were propositionalised by the author in such a way that core elements were given 2 points unless they were split into several constituents, while all other elements were given 1 point. Similarly to Färber 2002, core elements were defined as elements which carry essential information and have to be rendered in order for the whole sentence to make any sense (in terms of syntax, these were mainly sentence subjects, verbs, objects or complements). If any of these elements consisted of several constituents (e.g.€ “the roots of dissent and democracy”, a sentence subject), each constituent (i.e. “the roots”, “of dissent” and “of democracy” in this case) was given 1 point. In addition, logical links (such as “because”, “but”, “although”), which are also crucial for text cohesion, were treated as core elements. The scores for each selected source text fragment as a whole turned out to be very similar: 313 and 322 points for the general texts, 344 and 315 points for the specialist texts. Afterwards, all the interpretations were compared with the text base as described earlier, and points were awarded for each proposition rendered. Here is an example with a sentence from one of the general texts and its interpretation by one of the subjects:

(1) Historians will also celebrate the creation of Solidarność because its influence did not stop at the Polish border. historians // will celebrate // because // did not stop – 2 points each also // the creation // of Solidarność // its // influence // at the border // Polish – 1 point each



Effects of short intensive practice on interpreter trainees’ performance 



(2) Historycy będą również [h] świętowali st- [h] utworzenie Solidarności, ponieważ ich wpływ [h] nie zatrzymał się na granicy Polski. historycy // będą świętowali // ponieważ // nie zatrzymał się – 2 points each również // utworzenie // Solidarności // wpływ // na granicy // Polski – 1 point each The maximum possible score for rendering this sentence is 15 points, and the interpretation scored 14 points. The score was high in spite of a few clear imperfections: the misinterpreted pronoun “its”, which, in this interpretation, refers to the historians, three voiced hesitations and a false start. This example shows that propositional analysis cannot be treated as the sole criterion for good quality. Consequently, as in Kurz and Färber 2003, the propositional accuracy score was combined with error analysis. Here, however, the interpretations were analysed from the perspective of a monolingual listener, i.e. independently of the originals (to balance the fact that the first type of analysis is totally dependent on the source texts). The errors fall into two general categories: communicative inaccuracies and flaws in delivery. Communicative inaccuracies comprise six types of errors: silences, omissions, incomplete sentences, illogical sentences, lexical errors and grammatical errors. Silences are pauses longer than 10 seconds. Omissions are pauses longer than 3 seconds which do not have their counterparts in the original and which are obvious to the listener, for example because there is no logical connection between the sentences coming before and after the omission. Incomplete sentences lack certain elements which are crucial for their comprehension, such as a subject, or are obviously unfinished (for example, they end with “and”). Illogical sentences do not make sense, clearly do not fit the context or contradict something which was said earlier in the interpretation (very often, they reflect meaning errors, which are not accounted for in this classification as not always detectable by a monolingual listener). Lexical errors include those resulting from source language interference, words inappropriate in a formal context and other awkward target language expressions. Grammatical errors include different grammatical genders for a noun and its modifying adjective, a subject in the singular followed by a verb in the plural etc. As Polish is a highly inflected language, the potential for these errors is considerable. Errors of each type are given a specified number of penalty points: 20 points for silences, 10 points for incomplete and illogical sentences, 5 points for omissions, 3 points for lexical and grammatical errors. This weighing of errors is to some extent arbitrary, but the aim was to reflect their seriousness as perceived by a potential listener. Flaws in delivery, penalised with 1 point each, comprise voiced hesitations, repetitions, corrections and pronunciation errors. They do not endanger comprehension of the interpretation, but certainly make a bad impression on the audience and do not inspire trust in the interpreter, especially if they occur repeatedly.

 Magdalena Bartłomiejczyk

6. Results Propositional accuracy scores for each subject and averages for all the subjects before and after the practice are presented in Table€1. The last two columns show the relative improvement (positive values) or deterioration (negative values) expressed in percentage points (i.e. for her interpretation of general texts, S1 scored 81.15% before and 70.81% after the practice, and the difference between these values, -10.34, is given in the relevant column). A slight average improvement was noted (+ 4.37 for both text types), but the results vary a lot for individual subjects. Moreover, there are no clear correlations between the individual subjects’ scores before and after the practice, or between their scores for general and specialist texts. Only two subjects (who scored relatively high before the practice) show deterioration in their interpretations of general texts. Specialist texts, on the other hand, were interpreted less accurately after the practice by half the subjects; however, four other subjects showed considerable improvement of about 20€percentage points. Table 2 shows the results of error analysis. Communicative inaccuracies (CI) and flaws in delivery (FD) are presented as separate categories. The numbers refer to penalty points given for all the errors detected in the given interpretation. Therefore, positive values in the columns presenting the results after the practice should be interpreted as reflecting deterioration in the subjects’ performance. Table 1.╇ Propositional accuracy scores BEFORE

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 TOTAL

AFTER

general

drugs

general

drugs

81.15% 68.69% 49.20% 61.98% 69.01% 53.10% 61.18% 68.32% 71.74% 79.19% 66.36%

73.55% 72.38% 48.25% 49.71% 64.53% 18.41% 55.55% 39.36% 41.90% 46.35% 51.00%

– 10.34 + 12.68 + 11.05 â•⁄ + 3.86 + 12.36 â•⁄ + 1.85 â•⁄ + 1.12 â•⁄ + 1.33 â•⁄ – 4.96 â•⁄ + 4.20 â•⁄ + 3.31

â•⁄ – 7.20 – 13.65 â•⁄ – 0.95 â•⁄ – 3.04 â•⁄ – 3.90 + 20.83 â•⁄ + 1.43 + 20.52 + 20.31 + 19.93 â•⁄ + 5.43



Effects of short intensive practice on interpreter trainees’ performance 

Table 2.╇ Error analysis: penalty points BEFORE general

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 TOTAL

AFTER drugs

general

drugs

CI

FD

CI

FD

CI

FD

CI

FD

â•⁄ 17 109 172 â•⁄ 88 â•⁄ 71 â•⁄ 60 108 â•⁄ 40 â•⁄ 82 â•⁄ 40 787

â•⁄ 10 â•⁄ 43 â•⁄ 17 â•⁄ 31 â•⁄ 19 â•⁄â•⁄ 6 â•⁄ 39 â•⁄ 50 â•⁄ 31 â•⁄ 41 287

â•⁄ 46 119 120 â•⁄ 88 113 109 â•⁄ 73 122 163 â•⁄ 43 996

â•⁄ 30 â•⁄ 47 â•⁄ 27 â•⁄ 61 â•⁄ 43 â•⁄ 10 â•⁄ 63 â•⁄ 75 â•⁄ 55 â•⁄ 63 474

+ 49 – 51 – 46 – 27 – 20 – 38 â•⁄ – 9 + 73 â•⁄ + 2 â•⁄ – 8 – 75

+ 26 â•⁄ – 9 â•⁄ + 4 â•⁄ + 8 â•⁄ + 3 â•⁄ + 6 â•⁄ + 6 â•⁄ – 2 + 38 + 14 + 94

+ 12 – 31 + 15 + 46 – 32 – 28 + 10 –18 – 61 –12 – 99

+ 13 â•⁄ + 7 + 17 – 22 – 20 â•⁄ + 1 â•⁄ – 3 â•⁄ – 2 + 43 + 14 + 48

Again, the results for individual subjects show much variability. No obvious correlations can be detected. However, on average the subjects’ performance improved in terms of communicativeness (174 less penalty points for both text types, which corresponds to a drop of 9.76%) and deteriorated in terms of presentation (142 more penalty points, an increase of 18.66%). The results are more clear-cut for general texts. For this text category, the performance of seven subjects improved in terms of communicativeness. Presentation deteriorated in eight subjects. In order to present in more detail the changes in various aspects of the students’ interpreting performance, Table€ 3 shows the results of error analysis for each error category. The numbers reflect the occurrence of each type of error in interpretations of both types of texts by all the subjects. Flaws in delivery of all categories proved more frequent after the practice. As for communicativeness, interpreting performance seems to have improved in terms of its perceptible completeness (which, understandably, correlates with improvement in propositional accuracy), but failed to improve in terms of logical argumentation as well as lexical and grammatical correctness. The deterioration detected might to some extent be balanced out by the fact that, on average, the errors were distributed over longer target texts (as the interpretations recorded after the practice proved more complete).

 Magdalena Bartłomiejczyk

Table 3.╇ Error analysis: types of errors Type of error

BEFORE

AFTER

CI

Silence Omission Incomplete sentence Illogical sentence Lexical error Grammatical error

â•⁄â•⁄ 2 â•⁄ 48 â•⁄ 61 â•⁄ 50 â•⁄ 63 â•⁄ 68

â•⁄â•⁄ 0 (– 2) â•⁄ 27 (– 21) â•⁄ 39 (– 22) â•⁄ 61 (+ 11) â•⁄ 73 (+ 10) â•⁄ 85 (+ 17)

FD

Audible hesitation Repetition Correction Incorrect word pronunciation

483 â•⁄ 85 183 â•⁄ 10

562 (+ 79) â•⁄ 99 (+ 14) 220 (+37) â•⁄ 12 (+ 2)

7. Conclusions The improvement in interpreting skills of the students participating in the practice as measured by the tools used in this study turned out smaller than expected. I attribute this, to some extent, also to the fact that the interpretations before the practice were done after intensive preparations for the conference, which themselves might have had considerable influence on the subjects’ performance. Such a degree of motivation and hard work would most probably not have been possible to achieve in the case of preparations for any simulated event taking place in the classroom environment. The finding that the students’ delivery suffered constitutes a crucial departure point for plans on how to modify the practice conditions in order to prevent this in the future. It seems that more trainer and peer feedback given directly in the booth during the conference should refer to this aspect rather than to terminological problems, for example (these are not negligible as such, but play a smaller role in the students’ general performance). Participants could also be encouraged (as advocated inter alia by Déjean le Féal 1990) to record at least some of their interpretations during the conference on their own portable recorders and to listen to them critically afterwards, paying special attention to possible room for improvement in presentation (however, see Bartłomiejczyk 2007 for a discussion of potential problems connected with self-assessment). Another useful idea on how to make the participants more conscious of the impression they make on their audience was suggested by our hosts in Vienna during the latest practice (two more have already followed the one described here). In the future, after the initial one or two days of the conference, the sound from the booths occupied by our students could be broadcast to the floor and Polish delegates informed about the possibility of listening to the interpretations. In this way, the students would be confronted with a real (if small) audience, who, as an added benefit, could also be



Effects of short intensive practice on interpreter trainees’ performance 

asked to express their opinions on the interpretations. Clearly, the success of this method depends on whether the delegates concerned are willing to listen to the students’ output and provide comments. Against my initial hopes, no visible pattern emerged with respect to the improvement potential of better versus poorer interpreters. This, however, is also an important finding supporting a recruitment method based on the students’ motivation rather than on their interpreting skills at the time of selection. As a consequence, the latest practice was simply announced on the information board for students of the relevant programme and those eligible students who applied first were accepted. The above conclusions demonstrate that the research project has already proved useful to the practice programme itself and it is hoped that future changes planned on the basis of the findings will contribute to making such practices as effective as possible. However, I must also emphasise that not all the potential benefits from the practice could be measured by means of the tools applied here. For example, the students reported benefits such as getting familiar with a wide variety of accents of non-native speakers of English, becoming acquainted with conference procedures, a chance to pick up some tricks of the trade from experienced professionals (especially when taking relay from the English booth), and a great boost to their self-confidence. Last but not least, it seems that the certificates of participation obtained from the UN Office impress potential employers on the Polish market, which has helped some of the students find attractive jobs as translators and/or interpreters. The group graduated a few months ago and was evaluated as outstanding at their final exams, and my informal follow-up done by e-mail has shown that they are satisfied with a successful beginning to their professional careers.

References Barik, H.C. 1975. “Simultaneous interpretation: Qualitative and linguistic data.” Language and Speech 18(3): 272–297. Bartłomiejczyk, M. 2007. “Interpreting quality as perceived by trainee interpreters: Self-evaluation.” The Interpreter and Translator Trainer 1(2): 247–267. Bovair, S. and Kieras, D. 1985. “A guide to propositional analysis for research on technical prose.” In Understanding Expository Text, B.K. Britton and J.B. Black (eds), 315–364. Hillsdale, N.J.: Erlbaum. Déjean le Féal, K. 1990. “Some thoughts on the evaluation of simultaneous interpretation.” In Interpreting – Yesterday, Today and Tomorrow, D. Bowen and M. Bowen (eds), 154–160. Binghamton: State University of New York. Dillinger, M. 1994. “Comprehension during interpreting: What do interpreters know that bilinguals don’t?” In Bridging the Gap: Empirical Research in Simultaneous Interpretation, S. Lambert and B. Moser-Mercer (eds), 155–189. Amsterdam and Philadelphia: John Benjamins. Färber, B. 2002. Anticipation in German-English Simultaneous Interpreting. Unpublished M.A. thesis, Karl-Franzens Universität Graz.

 Magdalena Bartłomiejczyk Gile, D. 2003. “Quality assessment in conference interpreting: Methodological issues.” In La evaluación de la calidad en interpretación: investigación, Á. Collados Aís, D. Sanchez and D. Gile (eds), 109–123. Granada: Comares. Kalina, S. 2005. “Quality assurance for interpreting processes.” Meta 50(2): 768–784. Kintsch, W. 1974. The Representation of Meaning in Memory. Hillsdale, NJ: Erlbaum. Kopczyński, A. 1980. Conference Interpreting. Some Linguistic and Communicative Problems. Poznań: Wydawnictwo Naukowe Uniwersytetu im. Adama Mickiewicza. Kurz, I. 2001. “Conference interpreting: Quality in the ears of the user.” Meta 46(2): 394–409. Kurz, I. and Färber, B. 2003. “Anticipation in German-English simultaneous interpreting.” Forum 1(2): 123–150. Pöchhacker, F. 2001. “Quality assessment in conference and community interpreting.” Meta 46(2): 410–425. Tommola, J. 2003. “Estimating the transfer of semantic information in interpreting.” In La evaluación de la calidad en interpretación: investigación, Á. Collados Aís, D. Sanchez and D. Gile (eds), 125–146. Granada: Comares. Tommola, J. and Helevä, M. 1998. “Language direction and source text complexity. Effects on trainee performance in simultaneous interpreting.” In Unity in Diversity? Current Trends in Translation Studies, L. Bowker, M. Cronin, D. Kenny and J. Pearson (eds), 177–186. Manchester: St. Jerome.

Corpora in translator training A program for an eLearning course Kerstin Kunz, Sara Castagnoli, Natalie Kübler

Universität des Saarlandes, Germany, SSLMIT – Università di Bologna, Italy, Université Paris Diderot, France

Within recent years, corpora have gained considerable importance in Translation Studies, and a number of studies have also appeared which show their value for translator training (e.g.€Zanettin et al.€2003). However, results from a recent survey reveal that current practising and trainee translators still have insufficient awareness of corpora and expertise in using them to help in their translation workflow. In addition, while corpus linguistics courses are offered at some universities, no materials for self-learning are available to our knowledge: such materials might not only complement traditional courses, but would also be of special interest for professional translators, who are often under serious time constraints. This paper presents a free eLearning course on “Corpora for Translators” which has been developed by the EU-funded MeLLANGE project in an attempt to fill this gap.€It deals with the use of corpora for different translation-related activities (e.g.€source text analysis, translation proper, revision, terminology extraction) and is tailored to the needs of professional and trainee translators alike. Keywords: corpora for translation, translator training, e-learning, MeLLANGE, corpus studies

1. Introduction: Corpora in Translation Studies, Translator Education and the Translation Profession During the last decades the corpus-based approach has come to be recognised as one of the major research paradigms in Translation Studies, allowing scholars to study translation in a novel and systematic way, and addressing a variety of issues pertaining to the theory, description, and practice of translation (Laviosa 1998). If most descriptive corpus-based studies have focused on the analysis of the supposedly peculiar nature of the language of translations (see e.g.€the vast body of research on “translation

 Kerstin Kunz, Sara Castagnoli, Natalie Kübler

universals” inspired by Baker’s (1993) and Toury’s (1995) works), research on the applied side of the discipline has dealt with the promising applications of corpora to translator education and translation practice, which are enhanced nowadays by the increasing availability and usability of corpus tools. While reviewing all suggested applications of corpora to translation-related tasks would go beyond the scope of the present paper, we believe it is worth mentioning here some of the most relevant aspects and findings which have informed the conception of the MeLLANGE course on “Corpora and Translation” presented in Sections 2 and 3. Corpora can be seen as reference tools which may complement standard dictionaries and grammars, offering information on phraseology, register and frequency of linguistic phenomena which are often omitted or poorly documented in more traditional tools (Aston 1999). Several classroom experiences (e.g.€Zanettin 2001), for instance, have shown that target language corpora can help translators produce more natural-sounding translations, inasmuch as they can suggest conventional expressions and attested ways of performing given functions in specific text types; this is no surprise for translators, who have long been used to collecting reference texts in the target language matching the ST subject and genre. It has also been suggested that monolingual corpora of the source language can be consulted to better understand ST nuances (e.g.€Zanettin 2002). As for bi/multilingual corpora, the value of specialised comparable corpora for retrieving LSP terms and conventions is almost unquestioned (see e.g.€Bowker & Pearson 2002), while parallel corpora (and Translation Memories, as their alter ego) prove invaluable for observing strategies deployed by past translators (see e.g.€Pearson 2003; Johansson 2003). Despite such achievements and growing enthusiasm in a number of academic settings, it seems that the majority of translation students and teachers still need to be convinced of the great potential of corpus work for translator education and translation-related activities (Bernardini & Castagnoli 2008); the same goes for translation professionals, who appear to be largely unaware of corpora. This was revealed by a survey conducted in the framework of the EU-funded project MeLLANGE which aimed at defining user needs for learning materials on translation technology.1 The survey was designed to collect information about the use translators make of the Web, their attitude towards eLearning, as well as their awareness and use of corpora. Following two rounds of submissions in 2005 and 2006, 1,015 questionnaires were completed by professional (74%) and trainee (26%) translators from the UK (the majority), France, Germany and Italy. A concise and simplified definition of the word “corpora” was provided at the beginning of the Corpus section of the questionnaire in order to clarify what was meant by the term and hint at some differences existing between using “reference materials”

1. More information about MeLLANGE as well as full survey results are available from http:// mellange.eila.univ-paris-diderot.fr.



Corpora in translator training 

Table 1.╇ Corpus section (closed questions) of the MeLLANGE questionnaire (MeLLANGE 2006) Question

Answer (%)

Do you collect domain specific texts?

56.1 No 43.9 Yes

How do you collect them? (multiple choice allowed)

69.6 In electronic form 30.4 On paper

How do you use them? (multiple choice allowed)

53.4 Search through with software 46.6 Read them

Do you use ‘corpora’ in your translation practice?

58.2 No 41.8 Yes

If yes, do you use…? (multiple choice allowed)

25.9 22.8 18.8 15.9 12.9 3.7

What do you use to search them? (multiple choice allowed)

65.1 Search facility in word processor 20.2 Concordancer 12.1 Other search tools (specify: Trados, Concordance in translation memory) 2.0 UNIX utilities

If you do not use corpora, why? (multiple choice allowed)

41.0 21.0 20.2 7.5 6.7

Corpora of the target language Corpora of the source language Parallel corpora Domain specific corpora Comparable corpora General language corpora

Never heard about them I don’t have time to build them I don’t know how to use a concordancer I can’t see any advantage over Google I can’t see any advantage over translation memories 3.6 Other (1 specified – not sure if it will work with Macintosh)

Would you be interested in a service which 84.0 Yes quickly provides domain- and language-specif- 16.0 No ic corpora tailored to your needs? Would you be interested in a tool for extracting 83.4 Yes terms from a domain-specific corpus? 16.6 No Would you be interested in learning more about the potential that corpora offer?

85.9 Yes 14.1 No

 Kerstin Kunz, Sara Castagnoli, Natalie Kübler

and using “corpora”.2 As can be observed in Table€1, which summarises the results of the Corpus section, 43.9% of the respondents reported collecting reference materials, 69.6% of these specified that they collect texts in electronic format and 53.4% declared that they read these collections of texts rather than searching through them. An only slightly smaller percentage of respondents (41.8%) unexpectedly reported using corpora in their translation practice; the majority of these, however, specified that they use search facilities in word processors (65.1%) to search through them rather than concordancers (20.2%). In line with our expectations, the percentage of respondents who do not use corpora (58.2%) is largely determined by lack of information: 41% reported never having heard about corpora, and 20.2% admit to not having the skills to use them. Another 14% of negative answers (such as not seeing any advantage over Google or translation memories) may arguably have been prompted by the respondents’ limited awareness or superficial use of corpora; this hypothesis seems to be supported by the fact that 85.9% of respondents are interested in learning more about their potential. The MeLLANGE survey has therefore brought to light the need for tailor-made learning materials addressed to translators, which show the role corpora can play in the translation workflow and highlight their added value with respect to other tools and resources translators are currently more familiar with. The course we set out to present in the following section was created in an attempt to make up for the scarcity of available learning materials on the use of corpora for translation, in order to introduce practising translators to corpus principles and techniques that have already proved fruitful – and are indeed largely adopted – by scholars in TS and linguistics.

2. The MeLLANGE eLearning course on “Corpora for Translation” The MeLLANGE eLearning course on “Corpora for Translation” is an online course aimed at training translators in the use of corpora in the translation workflow, developed by the MeLLANGE project together with other eLearning materials for translators in the larger framework of devising a European Masters in Translation Technology. The aim of the MeLLANGE MA is to adapt the vocational training of translators and other language professionals to the new competences and skills required in the translation market. Participants therefore obtain advanced vocational training in the 2. The definition provided was: “Corpora are collections of texts in electronic form, usually grouped according to topic or type - contract, business letter, etc. Corpora may contain original texts in one language, comparable originals in two languages (comparable corpora), or originals and their translations (parallel corpora). Translation Memories are a special kind of parallel corpora. Corpora may be large and general, or small and specialised. They are (usually) not read cover to cover, so to speak, but searched by software programs (usually called “concordancers” or “corpus analysers”). One can list all the words contained in the corpus and see their frequencies, search for a word or expression (in context) and find out its typical patterns”.



Corpora in translator training 

application of new technologies in translation (e.g.€ Translation Memories, Markup Languages, Machine Translation and Corpora), in translation-related activities (e.g.€Localisation and Terminology Management), in Specialised Translation and in Information and Project Management, while basic competences and skills in translation are taken as a prerequisite. The “Corpora for Translation” course, in particular, deals with the possible applications of these innovative electronic resources to the practice of translation, instructing participants in why, how and when corpora can enter their translation workflow (more detailed information and examples are provided in Section€2.1 below). As all MeLLANGE courses address a dual audience (trainee and professional translators), they have been designed so that they lend themselves to different learning scenarios. On the one hand, they can be attended virtually, i.e. completely online, in a fully autonomous learning environment. With this scenario, we primarily target professionals who thus find themselves enabled to work at their own place, pace and time. Since the different course components are self-standing and, as much as possible, independent from each other, participants can also decide to take only certain components or sections. On the other hand, we foresee that components of our courses might be integrated into face-to-face sessions at advanced university level, thus giving rise to a blended learning scenario in which the MeLLANGE materials represent a repository of learning contents, ideas, activities, tools and corpus resources that teachers can use as they are or adapt and complement with assignments, forum postings, one-to-one and class-wide feedback, as well as face-to-face monitoring sessions.

2.1

Contents, principles and objectives

Having established (see Section€ 1 above) that there is limited awareness of corpora among translators, before even providing the rudiments of corpus work (e.g.€producing and interpreting concordances, sorting results, building small corpora, etc.) the course needs to build motivation for and interest in the use of corpora in translation, as its success will depend on the learners’ perception that corpus-based work can be relevant to their concerns. As a consequence, the different units which form the course (summarised in Table€2) revolve as much as possible around plausible translation-related tasks. This is particularly true for the units in the first section, which are meant to illustrate how corpora can assist translators in the different steps of the translation process: technical aspects are left aside in favour of guided reflective activities on case studies that require learners to start noticing and asking questions about aspects of language and texts that corpora can bring to light (see e.g.€Figure 1). Learning how to use corpora productively is not primarily a practical question: it implies learning about corpora, coming to understand the criteria used for their construction and the inferences which may be drawn from them, thus becoming aware of their potential as well as their limits (Aston 2000:€8).

 Kerstin Kunz, Sara Castagnoli, Natalie Kübler

Figure 1.╇ Screenshot from a reflective exercise in the unit on ST understanding

After this first – and mainly theoretical – section, learners are gradually introduced to a number of more practical issues. In Section€2 they learn how to interrogate corpora by making simple and elaborate queries. Section€3 deals with how to build specialised corpora for specific translation tasks, focusing on both theoretical aspects (e.g.€how to choose adequate reference documentation) and practical issues (e.g.€how to optimise web searches, download and save files). On a more advanced level, Section€4 introduces corpus annotation, i.e. tools and techniques for enriching corpora with information of different kinds such as text metadata (e.g.€author, publication date), linguistic information (Part-of-Speech tagging and lemmatisation), and alignment (in the case of parallel texts). Finally, the course closes with a specific application of corpora to terminology: Section€5 presents the process of identifying and extracting terms from specialised corpora, both manually and automatically. A number of considerations have been taken into account while drafting the course. First, as mentioned earlier, in order to help learners see the relevance of corpus work for their own activity, most units start by introducing a translation problem, preferably taken from an authentic translation task; the exercise shown in Figure 2, for instance, which is aimed at suggesting how a TL corpus can assist revision, is based on a real translation produced by a student and collected in the MeLLANGE Learner Translator Corpus.3

3. As its name suggests, the Learner Translator Corpus is a corpus of student translations collected in the framework of the MeLLANGE project (see Castagnoli et al. 2006). More information about the LTC can be obtained from the project website (http://mellange.eila.univ-paris-diderot.fr).



Corpora in translator training 

Table 2.╇ The MeLLANGE Corpora for translation course Unit

Contents

1 Overview

Introduction: why use corpora? Corpus use to understand a source text Corpus use to explore a text type Corpus use to produce a translation Comparable corpora Parallel corpora

2 Consulting corpora

Basic Advanced

3 Building your own corpus

To learn more about a text topic To produce or revise a translation

4 Encoding corpora

Introduction Structural mark-up and meta-data Linguistic annotation (manual and automatic) Querying annotated corpora Alignment

5 Applications: Term extraction

Manual Automatic

Second, corpus evidence is often compared with information obtainable from more familiar resources such as dictionaries and the Web: learners must be given enough information so as to understand and critically appreciate what corpora can offer them, otherwise they will hardly be motivated to keep using corpora beyond the duration of the course. Third, various exercises are included both within and in-between units, so as to force learners to participate actively in the learning process and to keep their attention alive. Unit-internal exercises allow students to check their in-progress comprehension, thus constituting a form of regular self-assessment, whereas unit-external exercises are more summative in nature while still not being intended for evaluative purposes. Since the course can be taken 100% online, without any intervention by a tutor or facilitator, most exercises are closed (e.g.€multiple-choice questions, matching/ ordering exercises) and participants can therefore receive instant feedback. However, other “open-ended” exercises in the form of assignments (e.g.€building a corpus, solving specific translation problems, applying a corpus annotation tool) requiring assessment and feedback by a tutor have also been devised for application to blended learning scenarios.

 Kerstin Kunz, Sara Castagnoli, Natalie Kübler

Figure 2.╇ Screenshots from an exercise on how to use a TL corpus to revise a translation

Use is also made of some collaboration tools typically used in eLearning settings, such as wikis, chats (for real-time synchronous discussions) and forums (for asynchronous discussions), in order to allow interaction between participants (and tutors, in blended-learning settings) and make them share knowledge and skills. Finally, besides learning contents and activities, the course includes suggestions for further readings and work, as well as links to software, websites on corpora and linguistics, and other MeLLANGE courses. Such a mix of eLearning methods was made possible by the



Corpora in translator training 

choice of specific learning environment and authoring tools, which are described in Section€2.2 below.

2.2

Creation and implementation

In this section we will discuss important aspects and criteria for the creation and implementation of the “Corpora for Translation” course from a technical, organisational and didactic perspective. Since, as was mentioned earlier, the course is designed for both academic and professional users, teaching material already available at several partner Universities had to be completed by new material, updated and adapted to the requirements of both target audiences and of the eLearning environment. For this purpose, trainers in corpus linguistics and translation have collaborated with specialists in eLearning as well as representatives of the translation industry, in order to create contents and activities which are innovative, relevant, motivating and up-to-date.4 In order to favour the usability, reusability and interoperability of our learning materials, we decided to create and distribute them with free, open-source tools which allow for customisation, and to produce materials which comply with existing eLearning standards (most notably the SCORM standard). The eXe XHTML editor (http://www.exelearning.org/) was used to create units offline and save them as SCORM packages which can be imported into any Learning Management System. For the purposes of MeLLANGE, contents and activities were implemented in a Moodle platform: Moodle (http://moodle.org) is an open-source Virtual Learning Environment which offers a wide range of interactive and collaborative activities while remaining very user friendly. It can also integrate exercises developed with other authoring tools, such as the Hot Potatoes (http://www.halfbakedsoftware.com/) software which we used to produce unit-external exercises, tracking participants’ performance and generating different types of reports. The “Corpora for Translation” course’s global architecture follows the principles outlined in the e-content and technical guidelines (MeLLANGE 2007) which the MeLLANGE Consortium produced and shared during the process of creating and implementing the courses, and which have been consistently applied in all MeLLANGE eLearning Courses. The course begins with an introductory page which gives information about its contents, its overall architecture, its learning objectives and target audiences, as well as about the order in which the different parts of the course should better 4. The authors would like to thank Guy Aston and Silvia Bernardini (Università di Bologna), Toni Badia (Universitat Pompeu Fabra), Cécile Frérot (Université Paris Diderot), Rüdiger Fries (Universität des Saarlandes), Silvia Hansen-Schirra (Johannes Gutenberg Universität Mainz) and Serge Sharoff (University of Leeds) for contributing to the development of the “Corpora for Translation” course; Susan Armstrong (Université de Genève), Carme Colominas (Universitat Pompeu Fabra), Stella Neumann (Universität des Saarlandes) and all other members of the MeLLANGE Consortium for useful comments and suggestions.

 Kerstin Kunz, Sara Castagnoli, Natalie Kübler

be followed. It continues with five Sections dealing with different topics which are relevant for the use of corpora in translation. Each Section€is further divided into several shorter units containing the basic contents of the course (see Section€2.1). Each unit encompasses an introductory page providing clear posting of prerequisites, learning outcomes and the duration of the unit, as well as a conclusion page summarising the key elements to be retained. Splitting the course into subtopics and units of manageable size allows participants to choose between going through all units or picking the ones they find most relevant for their needs. In addition, as has been reported by users testing the course (see Section€3 below), consistency in the structure and layout of the course and the units, as well as a reasonable length of units, are among the factors which can facilitate the learning experience.

3. Testing the course Testing the course was part of the project’s aims, in order to collect users’ feedback to be taken into account for subsequent improvements. Course testing has been implemented in different ways. First, a testing workshop was organised in May 2006, in which translation teachers from different CEE countries were introduced to the MeLLANGE methodology and the testing procedure so as to be able to oversee subsequent testing phases in their own Universities. Assessors were provided with testing guidelines introducing them to basic technical and didactical principles, e.g.€ about Moodle access, available MeLLANGE courses and course design, and were then asked to evaluate the courses. Second, the course was tested, and used, with students at some partner universities, with either a blended or fully online approach. In Paris, for example, a group of Master’s students who had no experience with corpora were asked to go through the entire online course and write a small report on corpus use for translation. Teachers testing the course thus had the possibility to discuss the feedback with their students directly and in situ. Third, the course was also tested by a number of professional translators who were given the possibility to take it and discuss specific issues on a special forum. A live chat was also organised in order to allow testers to ask questions about the course to MeLLANGE tutors. Professional translators, mainly members of the ITI, showed great interest in this innovative testing experiment. All testers were asked to fill in an evaluation questionnaire, an online form which was also accessible from the course website. The questionnaire was divided into two parts. The first part was aimed at collecting personal information about the testers, such as their experience as translators and web users, as well as previous experience with eLearning and knowledge of the subject matter. The second part was designed to collect feedback on four main points: general course presentation (mainly layout), navigation (e.g.€links between different parts), content (e.g.€relevance, clarity, usefulness of quizzes, see Figure 3) and overall satisfaction (see MeLLANGE 2007). While



Corpora in translator training 

Content 10. The writing style was easy to understand 11. The text was grammatically correct and free of errors. 12. The instructions were clear and sufficient. 13. The objectives of the unit were clear to me. 14. The activities were well chosen to fulfil the objectives. 15. The content was relevant. 16. The sequence of activities made sense to me. 17. The difficulty level was appropriate for me. 18. I found that the unit was too long. 19. I found the ratio of reading content to interactive activities good. 20. Quizzes helped me assess my level of knowledge. 21. Summaries were helpful.

NA O NA O

1 1

O O

2 2

O O

3 3

O O

4 O 4 O

NA O NA O NA O

1 1 1

O O O

2 2 2

O O O

3 3 3

O O O

4 O 4 O 4 O

NA O NA O

1 1

O O

2 2

O O

3 3

O O

4 O 4 O

NA O NA O NA O

1 1 1

O O O

2 2 2

O O O

3 3 3

O O O

4 O 4 O 4 O

NA O

1

O

2

O

3

O

4 O

NA O

1

O

2

O

3

O

4 O

Figure 3.╇ Screenshot from the content section of the evaluation questionnaire (MeLLANGE 2007)

the questionnaire was mostly made of closed questions, some open questions were included so as to give testers the possibility to express personal comments. A special section was also included for teachers, with questions on the perceived reusability of the materials. Returned questionnaires were then analysed with a view to extract useful feedback on content and layout to produce an improved version of the course. The analysis revealed overall satisfaction of the testers with respect to all four points under examination. Notwithstanding some differences between students’ and professionals’ observations, the most common remarks were that reading sequences were too long and that more examples and exercises would have facilitated the learning process; concerning layout, a number of testers lamented that course pages were too dense and not too appealing, which was reported to affect concentration. These were therefore the first and most substantial modifications carried out on the course as a consequence of the testing phase. The feedback also informed the drafting of a series of recommendations on how to create eLearning materials taking into account both students’ and professional translators’ needs (MeLLANGE 2007), which are available on the MeLLANGE website.

 Kerstin Kunz, Sara Castagnoli, Natalie Kübler

4. Concluding remarks: Course availability and prospects The MeLLANGE group intends to create a MeLLANGE community of users which will be able to use the learning materials described in this paper according to their needs. A fully working version of the course is available online (http://parles.upf.edu/ moodlercrc/course/view.php?id=33, login as guest) so as to allow potential users to (1) take it online, and (2) see how it can be implemented in an eLearning platform. The different units are also made available (http://mellange.upf.es/wp5/copy_of_exe-elpfiles-repository) as both SCORM packages (which can be imported on any SCORM-compliant eLearning platform) and eXe files, so that it is possible for teachers to adapt them to individual situations. For instance the course, which is presently fully available only in English, is currently being translated into French, with exercises and examples varying accordingly. Access to materials and updated information will constantly be provided on the MeLLANGE project’s website (http://mellange.eila.univparis-diderot.fr). Finally, we hope that members of the MeLLANGE community will continue to build learning materials drawing on the MeLLANGE course (e.g.€modified/ new units and examples integrating new corpora, corpus tools and insights in corpus linguistics as well as different language versions) and to share them, so as to create an ever-growing repository of eLearning materials on the use of corpora in translation.

References Aston, G. 1999. “Corpus use and learning to translate.” Textus 12: 289–314. Aston, G. 2000. “Corpora and language teaching.” In Rethinking Language Pedagogy from a Corpus Perspective, L. Burnard and T. McEnery (eds), 7–17. Frankfurt am Main: Peter Lang. Baker, M. 1993. “Corpus Linguistics and Translation Studies – Implications and Applications.” In Text and Technology – In Honour of John Sinclair, M. Baker, G. Francis and E. TogniniBonelli (eds), 233–250. Amsterdam/Philadelphia: John Benjamins. Bernardini, S. and Castagnoli, S. 2008. “Corpora for translator education and translation practice.” In Topics in Language Resources for Translation and Localisation, E. Yuste (ed.), 39–55. Amsterdam/Philadelphia: John Benjamins. Bowker, L. and Pearson, J. 2002. Working with Specialized Language: A Practical Guide to Using Corpora. London: Routledge. Castagnoli, S., Ciobanu, D., Kunz, K., Volanschi, A. and Kübler, N. 2006. “Designing a Learner Translator Corpus for Training Purposes.” Presented at TaLC 2006. Johansson, S. 2003. “Reflections on Corpora and their Uses in Cross-linguistic Research.” In Corpora in Translator Education, F. Zanettin, S. Bernardini and D. Stewart (eds), 135–143. Manchester: St.Jerome. Laviosa, S. 1998. “The corpus-based approach: A new paradigm in Translation Studies.” Meta 43(4): 474–479. MeLLANGE. 2006. Corpora and E-learning Questionnaire. Results Summary. Internal document, 20.04.06. Available online at http://mellange.eila.jussieu.fr/Mellange-Results-1.pdf



Corpora in translator training  MeLLANGE. 2007. Best practices in e-learning content creation and development. Internal document, September 2007. Available online at http://mellange.eila.univ-paris-diderot.fr/Best_ practices.pdf Pearson, J. 2003. “Using Parallel Texts in the Translator Training Environment.” In Corpora in Translator Education, F. Zanettin, S. Bernardini and D. Stewart (eds), 15–24. Manchester: St.Jerome. Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins. Zanettin, F., Bernardini, S. and Stewart, D. (eds). 2003. Corpora in Translator Education. Manchester: St Jerome. Zanettin, F. 2002. “Corpora for translation practice”. In Language Resources for Translation Work and Research, LREC 2002 Workshop Proceedings, E. Yuste-Rodrigo (ed.), 10–14. Zanettin, F. 2001. “Swimming in Words: Corpora, Translation and Language Learning.” In Learning with Corpora, G. Aston (ed.), 177–197. Houston, TX: Athelstan.

Psychology

Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung im Rahmen der interdisziplinären Forschung Caroline Lehr

University of Geneva, Switzerland

The following paper addresses the subject “Why Translation Study matters” from an interdisciplinary point of view. It shall outline a psycholinguistic experiment conducted with translation students in the context of an interdisciplinary research project. The experiment, which originally intended to test the cognitive development of a second language in the brain, provided results that could probably also give insights into the cognitive expertise of the translator. They suggest that language expertise and domain-specific expertise are at some points closely intertwined in the translator and that in interdisciplinary research, Translation Studies can not only play an important role by providing theoretical background for bilingual psycholinguistic experiments but also connect psycholinguistic and cognitive expertise research. Keywords: psycholinguistics, cognitive expertise research

1. Einleitung Mit der Erkenntnis, dass sich wissenschaftlicher Fortschritt oft an den Grenzen der Disziplinen vollzieht, hat interdisziplinäre Arbeit in den vergangenen Jahrzehnten für die Forschung immer stärker an Bedeutung gewonnen (Ullrich und Wilke 2006:€1). Auch in der ÜbersetzungswissenÂ�schaft, die in mancher Hinsicht schon immer interdisziplinär ausgerichtet war, kann diese Entwicklung beobachtet werden. Insbesondere im Rahmen der prozess-orientierten Übersetzungswissenschaft ist eine interdisziplinäre Zusammenarbeit mit der Neuro- und Psycholinguistik – Disziplinen in denen die kognitiven Prozesse des Sprachverständnisses, der Sprachproduktion und des Spracherwerbs mit empirischen Methoden untersucht werden – unerlässlich (Schwarz 1992:€8; Lee-Jahnke 2005:€359). Eine solche Zusammenarbeit kann, wie im Folgenden beschrieben, auch eine Verbindung zur Expertiseforschung herstellen, eines der am schnellsten wachsenden Gebiete der kognitiven Psychologie, das experimentelle Daten

 Caroline Lehr

nutzt, um kognitive Merkmale von Experten zu untersuchen (Hoffmann 1992:€ 29). Wie dies geschehen kann, und was dies für die weitere Forschung bedeuten könnte, soll im Folgenden erläutert werden.

2. Experimenteller Hintergrund Im Rahmen eines interdisziplinären Forschungsprojekts zwischen der Übersetzer- und Dolmetscherschule der Universität Genf und der Neuropsychologischen Abteilung des Genfer Universitätskrankenhauses1 werden psycho- und neurolinguistische Experimente mit Übersetzungsstudenten durchgeführt, mit dem Ziel, Einblicke in ihre bilinguale Sprachverarbeitung zu erhalten (Lee-Jahnke 2005:€359ff; Abutalebi et al. 2008). Eine solche Untersuchung der kognitiven Sprachverarbeitung von Übersetzungsstudenten ist nicht nur von Interesse, um allgemeine Einblicke in die kognitive Sprachverarbeitung und die kognitive Entwicklung einer Zweitsprache zu gewinnen. Ein im Rahmen dieses Forschungsprojekts in den Versuchsräumen des Genfer Universitätskrankenhauses (Laboratory of Experimental Neuropsychology) durchgeführtes Experiment mit 20 fortgeschrittenen Übersetzungsstudenten (3. und 4. Studienjahr), das ursprünglich dazu konzipiert war, die kognitive Entwicklung der Zweitsprache der Studenten zu untersuchen, zeigte ebenfalls interessante Ergebnisse, die möglicherweise Hinweise auf im Rahmen des Übersetzungsprozesses und der Entwicklung der kognitiven Expertise des Übersetzers automatisierte Teilprozesse liefern. Folgender experimenteller Hintergrund lag dem Experiment zugrunde: In psycholinguistischen Experimenten besteht der methodologische Ansatz zunächst einmal darin, sich auf Stimuli zu konzentrieren, die kleiner sind als ein gesamter Text, wie Wörter oder Sätze, um weitere Information über die Faktoren zu erhalten, die die Verarbeitung dieser Einheiten beeinflussen (Englund-Dimitrova 2005:€19). In dem im Rahmen des Forschungsprojekts durchgeführten Experiment wurde die Verarbeitung von semantisch verwandten und semantisch nicht verwandten Wörtern in der Erstund der Zweitsprache der Probanden unter Verwendung des Semantischen Priming Effekts getestet. Die Versuchskonzeption des Semantisches Primings geht aus den Priming-Verfahren hervor, die in psycholinguistischen Experimenten zu unterschiedlichen Zwecken sehr häufig eingesetzt werden, sowohl zur Untersuchung von Prozessen der Sprachrezeption als auch der Sprachproduktion. Priming ermöglicht eine Registrierung von Kontexteffekten im Mikrobereich, wie beispielsweise Reaktions-Vorteile durch die wiederholte Präsentation desselben Wortes (Repetition-Priming) oder wie in unserem 1. Dieses Projekt besteht zwischen der deutschen Abteilung der Schule unter der Leitung von Prof. Dr. Lee-Jahnke in Zusammenarbeit mit Prof. Dr. Jean-Marie Annoni (Hôpitaux Universitaires de Genève, HUG), Dr. Asaid Khateb (HUG), Prof. Dr. Theodor Landis (HUG), Dr. Jubine Abutalebi (Universität Mailand) and Dr. Alain Pegna (HUG).



Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung 

Fall Reaktionsvorteile durch zwei aufeinander folgende semantisch verwandte Wörter (Rickheit und Strohner 1993:€111f). Beim heutigen Stand der Forschung werden im Zusammenhang des Semantischen Primings nur Wörter mit einer konkreten Bedeutung betrachtet. Unter den 800 Wortpaaren, die wir für das Experiment zusammenstellten, befanden sich beispielsweise folgende semantisch verwandte und semantisch nicht verwandte Wortpaare in den Sprachen Deutsch und Französisch: agneau – cochon; courge – immeuble; Teppich – Vorhang; Kutsche – Zebra; Pfirsich – poire; cassis – Dampfer. Um nachvollziehen zu können, wie die semantische Verwandtschaft dieser Wortpaare die Sprachverarbeitung beeinflusst, muss man sich zunächst in Erinnerung rufen, wie Information über den Sprachgebrauch im Gehirn repräsentiert wird.

2.1

Kognitive Repräsentation

Die kognitive Repräsentation von Wörtern wird meist mit dem Konzept des mentalen Lexikons beschrieben, das sich im semantischen Gedächtnis befindet. Hinsichtlich des Aufbaus dieses Lexikons im Gehirn geht man heute davon aus, dass, wie es Dietrich (2002:€27) formuliert, “das Bild eines mehrdimensionalen Netzes mit Informationen als Knoten und Verbindungswegen zwischen bestimmten Informationen, stärkeren zwischen denen, die zusammen eine lexikalische Einheit bilden und stärkeren zwischen solchen gleicher Kategorie, eine brauchbare Heuristik darstellt.” Der Wortinhalt wird in diesem Netz von Repräsentationen durch die Relation eines Wortes zu seinen Nachbarn im Vergleich zum Verknüpfungsmuster anderer Worte dargestellt (Kress 1992:€223). Dies impliziert, dass die Repräsentationen semantisch verwandter Wörter näher beieinander liegen; je weiter sie sich räumlich voneinander entfernen, desto weiter ist auch ihr Inhalt semantisch entfernt. Wird nun ein Wort im Gehirn verarbeitet, werden die entsprechenden Repräsentationen im Gehirn aktiviert – unter ihnen auch die Repräsentation der Bedeutung. Der Spreading Activation Theory (Collins und Loftus 1975:€40ff) zufolge, breitet sich eine Aktivierung vom Informationsknoten der Bedeutung eines Wortes auf die umliegenden Nachbarknoten und damit auf die Bedeutungsrepräsentationen der assoziierten Wörter aus. In Experimenten zum Semantischen Priming Effekt werden zwei semantisch verwandte Wörter unmittelbar nacheinander gezeigt und von den Probanden im Zusammenhang einer ihnen gestellten Aufgabe verarbeitet. Wenn das zweite Wort verarbeitet wird, hat das erste Wort bereits die umliegenden Bedeutungsrepräsentationen der assoziierten Wörter aktiviert. Das zweite Wort trifft dann auf eine Preaktivierung und kann schneller verarbeitet werden. Semantisches Priming ist für die Untersuchung der bilingualen Sprachverarbeitung von besonderem Interesse, da der Effekt nicht nur für zwei Wörter einer Sprache, sondern auch für zwei Wörter zweier verschiedener Sprachen beobachtet werden kann. Dies gilt als einer der Hinweise darauf, dass semantische Information verschiedener Sprachen sprachübergreifend repräsentiert wird (Francis 1999:€209; Francis 2005:€251f).

 Caroline Lehr

3. Das Experiment Vor diesem Hintergrund konzipierten wir ein Experiment mit 20 Übersetzungsstudenten, die alle eine deutschsprachige Schule besucht und in einem deutschsprachigen Umfeld aufgewachsen sind und somit eindeutig als deutsche Muttersprachler bezeichnet werden können. Da es sich bei den Probanden um Studenten der Übersetzung handelt, die zudem in einem französischsprachigen Umfeld studieren, kann davon ausgegangen werden, dass sie ebenfalls über sehr gute Sprachkenntnisse in der französischen Sprache verfügen. Sie können daher gemäβ der auf Grosjean (siehe hierzu Francis 1999:€194) zurückgehenden, soziolinguistisch zentrierten Definition, der zufolge “Zweisprachigkeit” die regelmäβige Verwendung zweier oder mehrerer Sprachen im alltäglichen Leben umfasst, bedenkenlos als “zweisprachig” bezeichnet werden. Zu dieser Definition, die für psycholinguistische Experimente sehr häufig herangezogen wird, ist jedoch hinzuzufügen, dass die Probanden unseres Experiments die französische Sprache in einem durchschnittlichen Alter von 12 Jahren erwarben und daher als “späte” Zweisprachige bezeichnet werden sollten. Zwei Merkmale der Probandengruppe sind für die folgende Versuchsbeschreibung und -auswertung also von Bedeutung: Es handelt sich bei den Probanden sowohl um Personen, die als späte Zweisprachige gelten können, als auch um Studenten der Übersetzung, die sich bereits am Ende ihrer Ausbildung befinden. Bei den Probanden wurde der bereits näher erläuterte Semantische Priming-Effekt unter Verwendung der lexikalischen Entscheidungsaufgabe untersucht, die für die Untersuchung der bilingualen Sprachverarbeitung ein zentrales Verfahren darstellt. Die lexikalische Entscheidungsaufgabe besteht darin, in einem sehr kurzen Zeitfenster zu entscheiden, ob es sich bei dem gezeigten Stimulus um ein existierendes Wort oder eine sinnlose, einem Wort der betreffenden Sprachen ähnelnde Buchstabenfolge, ein Non-Wort, handelt. Zu den Wortpaaren mit semantisch verwandten und semantisch nicht verwandten Wörtern, wurden daher zum Zweck der Aufgabenstellung Wortpaare gebildet, die Non-Wörter enthielten. Im Fall einer impliziten Aufgabenstellung (wie in unserem Experiment) dient die lexikalische Entscheidungsaufgabe nicht zur Untersuchung der Worterkennung, sondern bildet den Rahmen für Beobachtungen zu einem bestimmten Mechanismus oder Effekt, der durch die Aufgabenstellung allerdings nicht erkennbar wird (Rickheit et al. 2002:€37). In unserer Versuchsreihe erfüllte jeder Proband die lexikalische Entscheidungsaufgabe in einer Einzelsitzung von insgesamt ungefähr einer Stunde. Die Probanden befanden sich hierfür alleine in einem geschlossenen Raum. Sie sahen die 800 semantisch verwandten und semantisch nicht verwandten Wortpaare (und die Wortpaare, die Non-Wörter enthielten) unterteilt in vier Blöcke von je 200 Wortpaaren, mit einer Pause zwischen jedem Block. Die Stimulus-Onset-Asynchronie (SOA), die Zeit zwischen dem Erscheinen des PrimeWortes und dem des Target-Wortes (erstes und zweites Wort eines Wortpaares), betrug in diesem Experiment 850 ms. Die statistische Analyse wurde mit Hilfe der ANOVA-Analyse durchgeführt. Es wurden sowohl die durchschnittlichen Reaktionszeiten



Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung 

der einzelnen Probanden auf semantisch verwandte und semantisch nicht verwandte Wörter als auch die durchschnittliche Anzahl der richtigen Antworten auf die lexikalische Entscheidungsaufgabe analysiert. Die inter- und intra-lingualen Bedingungen (beide Wörter stammen aus der gleichen Sprache oder aus zwei verschiedenen) wurden getrennt betrachtet.

4. Semantisches Priming und sprachliche Expertise Zunächst einmal ist von Bedeutung, dass die Ergebnisse des Experiments einen gleich stark ausgeprägten Semantischen Priming-Effekt in beiden Sprachreihenfolgen (das erste Wort ist ein deutsches oder ein französisches Wort) zeigten. In den für dieses Experiment herangezogenen bereits durchgeführten Experimenten (Keatley et al. 1994; Francis 1999), die nicht mit Übersetzungsstudenten durchgeführt wurden, variierte der Effekt jedoch, wenn das erste Wort des semantisch verwandten Wortpaares aus der Zweitsprache der Probanden stammte. Er war dann weniger stark ausgeprägt oder überhaupt nicht vorhanden, was wahrscheinlich auf unterschiedliche Wege der Sprachverarbeitung zurückzuführen ist, je nachdem ob es sich beim ersten Wort um ein Wort der Erst- oder der Zweitsprache handelt. Diese unterschiedlichen Wege der Sprachverarbeitung ergeben sich wahrscheinlich durch die Unterteilung der Repräsentation des für den Sprachgebrauch notwendigen Wissens. Aufbauend auf dem “Revised Hierachical Model” von Kroll und Stewart (1994), beschrieben in Keatley et al. (1994:€77), nimmt man beim heutigen Stand der Forschung an, dass das mentale Lexikon in hierarchische Ebenen gegliedert ist, wozu die konzeptuelle und die lexikalische Ebene zählen. Auf der konzeptuellen Ebene wird das Wissen über den Wortinhalt repräsentiert, auf der lexikalischen das Wissen über die orthographischen und phonologischen Formen der Wörter (Dietrich 2002:€ 27). Diese beiden Informationsebenen sind in der Erstsprache miteinander verbunden, in der Zweitsprache jedoch nicht unbedingt. In letzterem Fall geht man vielmehr davon aus, dass sich die Verbindungen mit einer hohen Kompetenz und einer fortgeschrittenen kognitiven Entwicklung der Zweitsprache herausbilden. Verfügen Probanden noch nicht über eine starke kognitive Ausprägung der Zweitsprache und über diese direkten Verbindungen zwischen konzeptueller und lexikalischer Ebene, verarbeiten sie Wörter der Zweitsprache unter Einbezug der lexikalischen Ebene der Erstsprache. Dies hat wiederum Auswirkungen auf den Semantischen Priming Effekt, der von einer Verarbeitung auf konzeptueller Ebene abhängt. Im Gegensatz zu den konsultierten Versuchsergebnissen zeigen die Ergebnisse unseres Experiments in beiden Sprachreihenfolgen einen gleich stark ausgeprägten Priming Effekt und sie deuten demnach darauf hin, dass die Übersetzungsstudenten bereits über eine kognitiv stark entwickelte Zweitsprache verfügen, da sie die semantisch verwandten Wörter beider Sprachen auf konzeptueller Ebene verarbeiten können. Eine solche kognitiv stark entwickelte Zweitsprache spricht für die sprachliche

 Caroline Lehr

Kompetenz unserer Probanden, könnte jedoch wahrscheinlich auch bei anderen Zweisprachigen beobachtet werden.

5. Automatisierung von Teilprozessen im Rahmen der kognitiven Expertise des Übersetzers Das ursprüngliche Interesse an dem im Rahmen des Forschungsprojekts durchgeführten Experiment zum Semantischen Priming Effekt bestand also darin, Aussagen über die kognitive Entwicklung der Zweitsprache der Studenten machen zu können. Vor dem Hintergrund einer Arbeitshypothese, die zunächst die rein sprachliche Expertise der Probanden betraf, konnten wir unter Verwendung einer experimentellen Methode der Psycholinguistik jedoch möglicherweise ebenfalls Einblicke in die kognitive Expertise des Übersetzers erhalten. Zu den Ergebnissen des Experiments lassen sich verschiedene Vermutungen anstellen, denen jedoch einige Vorbehalte vorausgeschickt werden müssen: In der zuvor beschriebenen Versuchsreihe wurde die kognitive Verarbeitung von Wörtern auβerhalb des Kontextes untersucht. Hierin besteht ein wesentlicher Unterschied zur Sprachverarbeitung im alltäglichen Leben und auch zur Sprachverarbeitung im Rahmen der Übersetzung. Zudem handelte es sich bei den untersuchten Prozessen um reflexgesteuerte Prozesse, die zwar die Grundlage für reflexionsgesteuerte Prozesse und damit eine fundamentale Ebene bilden, allerdings keinen Einblick in komplexere Sprachverarbeitungsvorgänge liefern, die für den Übersetzungsprozess ausschlaggebend sind. Dennoch ist zu bemerken, dass in der statistischen Analyse Besonderheiten auftraten, die mit der Übersetzung in Verbindung stehen könnten. So fiel in der statistischen Analyse auf, dass zwei semantisch verwandte Wörter der Sprachreihenfolge Französisch-Deutsch (Zweit- und Erstsprache) von unseren Probanden schneller verarbeitet werden konnten als zwei semantisch verwandte deutsche Wörter. Dies ist sehr ungewöhnlich, da die Erstsprache normalerweise auch bei Probanden mit hoher Sprachkompetenz in der Zweitsprache die am schnellsten verarbeitete Sprache bleibt. Die Rolle der Erst- und Zweitsprache scheint daher im Fall unserer Probanden nicht mehr eindeutig zu sein. Es sieht vielmehr so aus, als hätte sich eine Sprachreihenfolge herausgebildet, die für unsere Probanden nun die Bedingung darstellt, unter der ein Zugriff auf das mentale Lexikon am schnellsten möglich ist. Hierfür sind wahrscheinlich zwei Faktoren entscheidend: Die Sprachreihenfolge Französisch-Deutsch entspricht zunächst einmal einer Übersetzungsrichtung, die den Probanden sowohl im Rahmen ihres Studiums als auch in ihrem Alltag in einem französischsprachigen Umfeld häufig begegnet. Da die besonders schnelle Verarbeitung nur bei semantisch verwandten Wortpaaren auftrat, scheint auch die semantische Verwandtschaft der Wörter von Bedeutung zu sein. Warum dies so sein könnte, lässt sich anhand der Besonderheiten des Übersetzungsprozesses verdeutlichen: Der Ausgangstext stellt den Ausgangspunkt



Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung 

für den Übersetzungsprozess dar. Für den erfolgreichen und effizienten Ablauf dieses Prozesses zerlegt der Übersetzer den zu übersetzenden Text in Segmente, die er dann in mikrokontextuellen Handlungsketten aufarbeitet (Wilss 1998:€39). Eine entscheidende Form von Textsegmenten sind Wörter, die allerdings nach Wilss (1998:€ 93) der Zielsprache nicht in “Eins-zu-Eins-Entsprechungen” gegenüber stehen, sondern in einer “Eins-zu-Viele-Entsprechung” (Koller 2004). In diesem Bewusstsein und auf der Suche nach möglichst vielen zielsprachlichen Entsprechungen und damit einer möglichst großen Bandbreite an Übersetzungsmöglichkeiten stellt die Assoziation ein für den Übersetzer grundlegendes Suchverfahren dar (Risku 1998:€154). Das im Verlauf des Übersetzungsprozesses notwendige “Absuchen des semantischen Feldes” ist wahrscheinlich ein heuristisches Suchverfahren (Wilss 1998:€38), das, schon ehe der Verstehensvorgang des Ausgangstextes abgeschlossen ist, in einer Vorbereitung auf den Übersetzungsprozess beginnt. Der Übersetzer würde demzufolge, noch bevor er den Ausgangstext vollkommen verstanden hat, im Unterbewusstsein bereits verschiedene Übersetzungsmöglichkeiten in Betracht ziehen und dabei das semantische Feld “durchsuchen”. Diese Suche nach für den Übersetzungsprozess relevanten Wörtern als systematische Nutzbarmachung von Wissen hat sich bei unseren Probanden möglicherweise als Verhaltensnorm herausgebildet und ermöglicht ihnen ein deutsches Wort nach einem vorangegangenen französischen semantisch verwandten Wort überdurchschnittlich schnell zu verarbeiten. Eine derartige Verhaltensnorm würde nicht nur die besonders schnelle Reaktionszeit unserer Probanden erklären, sondern auch einen automatisierten Teilprozess innerhalb der Übersetzungsstrategie für eine bestimmte Sprachreihenfolge reflektieren. Darüber hinaus fanden sich bei unseren Probanden auch in der Performanzanalyse des Experiments Besonderheiten, die Hinweise auf für Übersetzer spezifische beschleunigte Prozesse liefern könnten. So konnten die Probanden mit Hilfe einer Verarbeitungsstrategie den Kontext eines französischen Wortes nutzen, um in der lexikalischen Entscheidung über ein folgendes semantisch verwandtes deutsches Wort eine höhere Performanz zu erreichen als bei zwei Wörtern ihrer Erstsprache. Die beschriebenen Ergebnisse unseres Experiments scheinen zu belegen, dass bilinguale Sprachverarbeitungsprozesse durch den ständigen Sprachwechsel während des Übersetzens beeinflusst und geformt werden und dass Übersetzer für sie spezifische kognitive Fähigkeiten herausbilden, welche die natürliche Gabe der Übersetzung ausprägen und festigen. Es ergibt sich somit das für die kognitive Expertise typische Bild eines Zusammenspiels aus allgemeinen und spezifischen Fähigkeiten (Ackermann 1986:€101ff). Zu diesen spezifischen Fähigkeiten, der translatorischen Expertise, gehört neben dem Umfang des Domänenwissens und höheren analytischen Fähigkeiten auch die Automatisierung der Verarbeitungsprozesse. Im Rahmen der Automatisierung von Teilprozessen und -strategien, die nötig sind, um Probleme der jeweiligen Domäne zu lösen, zeigt sich, dass Experten nicht nur Zugang zu mehr Wissen haben, sondern auch Wissen restrukturieren und

 Caroline Lehr

infolgedessen auf relevante Information schneller zugreifen können (Englund-Dimitrova 2005:€16ff). Sie können ihr Wissen effizienter anwenden und den kognitiven Kontrollaufwand, der nötig ist, um dieses Wissen zugänglich zu machen, reduzieren (Schmidt et al. 1988:€46ff). Wie automatisch Teilprozesse im Zuge einer Reduzierung des kognitiven Kontrollaufwands während des Übersetzungsprozesses ablaufen, beeinflusst die Effizienz des Prozesses und die Qualität der Übersetzung (Griffin und Danks in Danks et al. 1997:€ 171). Gerade der Prozess einer sprachübergreifenden Assoziation scheint in unserem Fall genutzt zu werden, um überdurchschnittlich schnell für den Übersetzungsprozess relevante Information verfügbar zu machen, um dadurch effizienter zu arbeiten und Freiraum für die Lösung weiterer Probleme, die sich beim Übersetzen stellen, zu schaffen. Neben den Aussagen über die sprachliche Expertise scheinen die Ergebnisse unseres Experiments daher auch Aussagen über die translatorische Expertise der Studenten und Vermutungen über das Verhältnis beider Expertisen zueinander zuzulassen.

6. Sprachliche und translatorische Expertise Was das Verhältnis zwischen sprachlicher und translatorischer Expertise betrifft, lassen die Ergebnisse des beschriebenen Experiments vermuten, dass die Automatisierung der Verarbeitungsprozesse, und damit auch die allgemeine kognitive Expertise des Übersetzers, zumindest teilweise in die sprachliche Expertise eingebettet sind. Zunächst einmal ausgehend von der Annahme, dass die sprachliche und die kognitive Expertise getrennt voneinander betrachtet werden müssen, kann die Trennung dieser beiden Expertisen anhand der von der PACTE Forschungsgruppe2 beschriebenen Subkompetenzen der Translationskompetenz theoretisch verdeutlicht werden. Unter den von PACTE definierten Subkompetenzen befinden sich die “bilinguale Subkompetenz”, die “strategische Subkompetenz” und die “psychophysiologische Komponente.”3 Während die bilinguale Subkompetenz, die für eine zweisprachige Kommunikation notwendigen Kenntnisse, darunter auch die rein lexikalischen Kenntnisse und die natürliche Gabe, zwischen den Sprachen zu wechseln, umfasst, beinhaltet die strategische Subkompetenz das prozedurale Wissen, das die Effizienz des Übersetzungsprozesses und die Problemlösung steuert. Dieser dem Übersetzer eigenen Subkompetenz wird eine entscheidende Rolle im Übersetzungsprozess zugeschrieben, da durch sie die Planung des Übersetzungsprozesses und die Ausarbeitung der Übersetzung selbst ermöglicht 2. Forschungsgruppe für empirische experimentelle Studien zur Kompetenz des Übersetzers der Universidad Autónoma de Barcelona. 3. Neben der aufgeführten “sous-compétence bilingue”, der “sous-compétence stratégique” und den “composants psychophysiologiques”, nennt PACTE die “sous-compétence extralinguistique”, die “sous-compétence de connaissances en traduction” und die “sous-compétence instrumentale” (PACTE 2007:€2f).



Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung 

werden. Zusätzlich zu den Subkompetenzen führt PACTE die psychophysiologische Komponente an, die für alle gegebenen kognitiven Aspekte wie das Gedächtnis oder die Wahrnehmung steht (PACTE 2007:€2f). PACTE postuliert nun, dass im Übersetzungsprozess die definierten Subkompetenzen ineinander greifen. Hierfür könnte sich ein konkretes Beispiel zeigen, setzt man die theoretische Unterteilung mit den zuvor beschriebenen experimentellen Ergebnissen in Verbindung. Da die Ergebnisse unseres Experiments darauf hinweisen, dass allgemeine für den bilingualen Sprachgebrauch notwendige Suchprozesse im mentalen Lexikon durch den Übersetzungsprozess geprägt und modifiziert werden, würde in diesem Fall die bilinguale Subkompetenz als Rahmen für die strategische Subkompetenz dienen. Es kann zusammenfassend gesagt werden, dass ausgehend von der Annahme, dass die sprachliche und die translatorische Expertise zunächst einmal getrennt voneinander betrachtet und analysiert werden müssen, unsere Ergebnisse allerdings Anhaltspunkte dafür zu liefern scheinen, dass diese beiden im kognitiven Bereich an gewissen Stellen direkt ineinander greifen. Auch in Zukunft könnte die Kognitionsforschung das in der Übersetzungswissenschaft so viel diskutierte Verhältnis zwischen sprachlicher und translatorischer Expertise neu beleuchten und Hinweise darauf liefern, inwiefern sich beide unterscheiden oder auch überschneiden.

7. Die Übersetzungswissenschaft zwischen Psycholinguistik und Expertiseforschung Die Kognitionsforschung öffnet der Übersetzungswissenschaft in vieler Hinsicht neue Möglichkeiten und insbesondere die Psycholinguistik nimmt für die Untersuchung des Übersetzungsprozesses eine tragende Rolle ein. Arbeitete die Psycholinguistik in der Vergangenheit experimentell hauptsächlich auf Wortebene (wie im zuvor beschriebenen Experiment), ist sie im Moment im Begriff die Versuche mit Äquivalenten verschiedener Sprachen auch auf die Textebene auszudehnen. Da diese Experimente mit sprachlichen Äquivalenten beim heutigen Stand der Forschung in den meisten Fällen zunächst einmal bestätigen, was in der Übersetzungswissenschaft bereits theoretisch formuliert wurde, kann der Beitrag der Übersetzungswissenschaft hier von groβer Bedeutung sein. Diese kann den Experimenten nicht nur in der Ausarbeitung sondern auch in der Auswertung einen theoretischen Rahmen mit ihrem Wissen über den Vorgang des Übersetzens bieten, vor allem wenn es sich um Experimente handelt, die auf Textebene stattfinden. Die von Hatzidaki und Pothos (2008) mit sprachlichen Äquivalenten auf Wort- und Textebene durchgeführten Experimente und ihre Auswertung belegen unter anderem, dass der übersetzungstheoretische Rahmen von der psycholinguistischen Forschung herangezogen und benötigt wird. Die Übersetzungswissenschaft könnte sich in dieser Hinsicht noch stärker einbringen. Im Gegenzug würden experimentelle Daten diversen Aspekten der Übersetzungstheorien ein bisher ungekanntes experimentelles Fundament liefern.

 Caroline Lehr

Ein verstärktes Einbringen in die experimentelle Forschung könnte nicht nur über den theoretischen Zusammenhang geschehen, sondern auch darin bestehen, Übersetzer, wie zuvor beschrieben, in psycholinguistische Experimente als Probanden zu integrieren. Diese können nicht nur als kompetente Zweisprachige dienen, sondern auch Aufschluss darüber geben, wie sprachliche Expertise und domänenspezifisches Expertenwissen zusammenspielen, sich unterscheiden oder auch überschneiden. Im Rahmen einer solchen interdisziplinären Forschung befände sich die Übersetzungswissenschaft dann nicht nur in einer idealen Position, um ihr theoretisches Wissen über Übersetzungsprozesse in die experimentelle Forschung einzubringen, sondern auch, um als Bindeglied zwischen der Psycholinguistik und der kognitiven Expertiseforschung zu fungieren.

Bibliographie Ackerman, P. 1986. “Individual differences in information processing: An investigation of intellectual abilities and task performance during practice.” Intelligence. 10:101–139. Abutalebi, J., Annoni, J.M., Zimine, I., Pegna, Al., Seghier, M., Lee-Jahnke, H., Lazeyras, S. and Khateb, A. 2008. “Language Control and Lexical Competition in Bilinguals. An event-related fMRI Study.” Cerebral Cortex 18 (79):1496–1505. Collins, A. and Loftus, E. 1975. “A spreading activation theory of semantic processing.” Psychological Review 82:407–428. Dietrich, R. 2002. Psycholinguistik. Stuttgart: J.B. Metzler. Englund-Dimitrova, B. 2005. Expertise and Explicitation in the Translation Process. Amsterdam/ Philadelphia: John Benjamins. Francis, W. 1999. “Cognitive integration of language and memory in bilinguals: Semantic representation.” Psychological Bulletin 125:193–222. Francis, W. 2005. “Bilingual semantic and conceptual representation.” In Handbook of Bilingualism: Psycholinguistic Approaches, J. Kroll, and A. De Groot (eds), 251–267. New York: Oxford University Press. Griffin, J. and Danks, J. 1997. “Reading and translation: A psycholinguistic perspective.” In Cognitive Processes in Translation and Interpreting, J. Danks, G. Shreve, S. Fountain and M. McBeath (eds), 161–175. Thousand Oaks/London/New Delhi: SAGE Publications. Hatzidaki, A. and Pothos, E. 2008. “Bilingual language representation and cognitive processes in translation.” Applied Psycholinguistics 29:125–150. Hoffmann, R. 1992. The Psychology of Expertise. New York: Springer. Keatley, C., Spinks, J. and De Gelder, B. 1994. “Asymmetrical cross-language priming effects.” Memory and Cognition 22:70–84. Koller, W. 2004. “Übersetzungskompetenz als Fähigkeit, Äquivalenz herzustellen.” In Translationskompetenz, E. Fleischmann, P.A. Schmitt and G. Wotjak (eds), 117–134. Tübingen: Stauffenburg. Kress, J. 1992. Psycholinguistics. Amsterdam/Philadelphia: John Benjamins. Kroll, J.F. and Stewart, E. 1994. “Category interference in translation and picture naming: Evidence for asymmetric connections between bilingual memory representations.” Journal of Memory and Language 33:149-174.



Psycholinguistik, Übersetzungswissenschaft und Expertiseforschung  Lee-Jahnke, H. 2005. “New cognitive approaches in process-oriented translation training.” In Meta – Processus et Cheminements en traduction et interprétation, H. Lee-Jahnke (ed.), 50/2: 359–378. PACTE. “Une recherche empirique expérimentale sur la compétence de traduction.” URL: http://www.fti.uab.es/pacte/publicacions/web_rennepdf (14.12.2007). Rickheit, G., Sichelschmidt, L. and Strohner, H. 2002. Psycholinguistik. Tübingen: Stauffenburg. Rickheit, G. and Strohner H. 1993. Grundlagen der kognitiven Sprachverarbeitung. Tübingen/ Basel: Francke. Risku, H. 1998. Translatorische Kompetenz – Kognitive Grundlagen des Übersetzens als Expertentätigkeit. Tübingen: Stauffenburg. Schmidt, F., Hunter, J., Outerbridge, A. and Goff, S. 1988. “Joint relation of experience and ability with job performance: test of three hypothesis.” Journal of Applied Psychology 73:46–57. Schwarz, M. 1992. Einführung in die kognitive Linguistik. Tübingen: Francke. Ullrich, S. and Wilke, T. 2006. “Was ist Interdisziplinarität? Eine Einführung.” In Interdisziplinarität als Wissenschaftsparadigma. URL: www.interdisziplinaritaet.eu (7.12.2007). Wilss, W. 1998. Kognition und Übersetzen. Zur Theorie und Praxis der menschlichen und der maschinellen Übersetzung. Tübingen: Narr Verlag.

Interpreting Studies and psycholinguistics A possible synergy effect Agnieszka Chmiel

Adam Mickiewicz University, Poznań, Poland

Cognitive information processing has long been an area of interest for Interpreting Studies scholars. This paper discusses the interdisciplinary relation between Interpreting Studies (IS) and psycholinguistics as a source of a possible synergy effect and explains why Interpreting Studies matters not only to interpreters and interpreting researchers, but also to psycholinguists. First, the most significant contributions of psycholinguistics to interpreting research are identified, including in particular theoretical memory models and experimental methodology in mental lexicon studies. This is followed by an overview of the potential contribution of IS to psycholinguistics and bilingual studies. Conference interpreting is a unique case of bilingual/multilingual use of languages with more frequent code switching and greater inhibitory demands as compared to the non-interpreting use of language. Therefore, ‘bidirectional’ (A into B and B into A) and ‘unidirectional’ (C into A) interpreters and trainee interpreters are interesting experimental populations for mental lexicon studies and other cognitive studies. Finally, a progress report on COGSIMO, a research project aiming at leveraging the synergy of psycholinguistics and IS, will be presented. The project seeks to examine psycholinguistic processes and the skills of conference interpreting to establish efficient aptitude testing procedures for interpreter trainees based on cognitive predictors of success. Keywords: psycholinguistics, interpreting studies, mental lexicon, cognitive studies, cognates, verbal fluency, bilingualism, directionality

Introduction Cognitive information processing has long been an area of interest for Interpreting Studies scholars. Research by Gerver (1976); Moser-Mercer et al.€(2000); Gile (1995) and others has shed more light on the intricacies of cognitive processes in interpreting and led to new models (focusing on skills and efforts). This paper discusses the

 Agnieszka Chmiel

interdisciplinary relation between IS and psycholinguistics in terms of possible synergy. The theme of the 2007 EST Congress, “Why Translation Studies matters”, will thus be answered by showing that cognitive research within IS can be beneficial not only to interpreting researchers, but also to psycholinguistics. This will be followed by the analysis of possible benefits and a review of the latest research results in the area. Finally, a progress report on an ongoing longitudinal cognitive study of interpreters and trainees will be presented. In this paper, I argue that IS matters to interpreters, interpreting trainees and Interpreting Studies scholars – both “pure” researchers and practisearchers, i.e. practicing interpreters cum researchers (Gile 1994:€156), but also that, as a scholarly discipline, it has a lot to offer to researchers specialising in memory, mental lexicon and bilingualism. Interpreters are an interesting population to study because they represent a special case of bilingualism. In this paper, we are not simply trying to reiterate Gile’s call for interdisciplinary research in IS and a contribution by experts in psycholinguistics (as reviewed in Pöchhacker 2004:€72). By engaging in interpreting research, psycholinguists can not only contribute, but also benefit, and it will be shown how.

1. How psycholinguistics contributes to IS It is beyond the scope of the present paper to present an overview of all psycholinguistic research trends in IS. Suffice it to say that even early pioneers in Interpreting Studies considered it important to examine the process of interpreting from a cognitive point of view (cf.€Gerver 1976; Seleskovitch 1978). The sections below will focus on benefits of selected psycholinguistic studies in IS. The most valuable contributions of psycholinguistics to IS include its theoretical approaches and methodology for cognitionoriented empirical studies. We shall focus on memory and the mental lexicon since these two areas of psycholinguistic research seem to be especially dear to the interpreting researchers’ hearts.

1.1

Theoretical approaches – memory models

The most influential memory model in Interpreting Studies is the multi-component working memory model developed by Baddeley and Logie (1999) and the long-term memory model by Squire and Zola (1996). Working memory can be further subdivided into components specialising in separate types and modes of cognitive processing: the phonological loop, the visuo-spatial sketchpad and the central executive. Long-term memory handles declarative and procedural knowledge. The former is further divided into episodic memory (recollection of personal experiences) and semantic memory (encyclopaedic knowledge about the world). The latter is unconscious and responsible for skills and habits (procedural memory) (Squire and Zola 1996).



Interpreting Studies and psycholinguistics 

Chmiel applied these theoretical memory models to review experimental cognitive studies featuring interpreters as subjects and to explain the complex activity of conference interpreting (2004). She concluded that the parallel involvement of multiple memory systems is a success factor in interpreting. For instance, auditory memory supports the interpreter in consecutive interpreting since the speaker’s input can be ‘played back’ while providing a target language rendition. In general, declarative memory is used as a source of background knowledge. Semantic memory related to text structures, current political events, or facts learned at a conference the previous day etc. is tapped into when decoding the input. With a familiar speaker, interpreters will fall back on their episodic memory and past experiences with that speaker. The intermodal nature of interpreting entails the activation of separate neurocognitive resources (temporal lobe areas for linguistic processing and occipital lobe areas for visual processing). Paradoxically, some concurrently performed tasks may also share neuronal modules in the brain and thus require less attentional resources than the sum of resources engaged by those tasks individually. Obviously, one has to remember that parallel processing requires good management and coordination mechanisms which tax working memory. Nevertheless, parallel processing remains more powerful than serial processing of a single system with no specialised modules. The intermodal nature of interpreting (audio and visual input, oral and visual output) also entails the involvement of various processing modules, such as imagery, which proves an extremely powerful mnemonic. The use of well-established symbols in note-taking may involve the visuo-spatial sketchpad (VSSP) only and leave the phonological loop processing capacity for other tasks. Notes serve not only as a memory prompt. Because symbols tend to be alingual, source-language interference is attenuated (Gernsbacher and Shlesinger 1997). The multimodal nature of interpreting forces interpreters to rely on channels which handle different physical signals (spoken words vs. written notes), which, according to Cowan (1999) is easier than attending to channels of the same modality. Other studies on intermodal aspects of interpreting brought equally interesting results. Agrifoglio (2004) conducted an error analysis in sight translation and compared it to consecutive interpreting (CI) and simultaneous interpreting (SI). Sight translation brought more errors of expression, while CI/SI led to more errors of meaning, which suggests that visual interference is stronger than audio interference (Agrifoglio 2004:€61). Lambert (2004) conducted a similar study with sight translation, simultaneous interpretation and simultaneous interpretation with text. She evaluated only performance rates without differentiating between types of errors and found that the text available through the visual modality lifts the burden off the Memory Effort as defined in Gile’s Effort Models (1995:€91). Interpreters have various strategies at their disposal to perform the same task. Usually, they use the less cognitively demanding strategy and only resort to the less economical ones when encountering problems. The phonological loop is involved only in more demanding processing, e.g.€in off-line analysis of linguistically complex

 Agnieszka Chmiel

sentences or difficult lexical items. Additionally, the subvocal rehearsal process is at the interpreter’s disposal when putting down more difficult words and non-contextual information with a high risk of errors (e.g.€in proper names) in the listening phase of CI. Phonological information is not necessary for semantic access, therefore the loop may actually be bypassed in sight translation to save cognitive resources. Similarly, while reading, interpreters may use a simpler direct strategy (no phonological processing) or a more effortful grapheme-phoneme conversion when reading unfamiliar words. This more demanding strategy does not overburden the phonological loop since it is assumed not to involve deriving phonology from print via the prelexical route (Baddeley and Gathercole 1993). Paradoxically, limitations of cognitive resources may sometimes be beneficial to interpreters. For instance, it seems plausible that in the second stage of CI (involving note-reading and target-language production) articulatory suppression stemming from the interpreter’s own output inhibits source-language interference by channelling the processing through visual stimuli (symbols) that are less likely to lead to interference than verbal source-language stimuli. Similarly, in SI the interpreter’s oral output may serve as an inhibitor of source-language interference. The limitations of attentional and memory resources seem to make the interpreter focus on the most relevant information to convey. In fast delivery, such a ‘summary’ strategy is communicatively the best and the most effective. As the above overview shows, working memory and long-term memory models from cognitive psychology help explain various phenomena and shed more light on cognitive processing performed by conference interpreters.

1.2

Experimental methodology – memory and the mental lexicon

Apart from these theoretical models, one of the most relevant contributions of psycholinguistics to cognitive IS is its methodology. An array of experimental tasks (including digit span, reading span and list recall tasks for working memory research and lexical decision tasks, verbal fluency, word completion and priming procedures for mental lexicon studies) offers data comparable across populations. Results are measurable, reliable and devoid of subjective bias characteristic of some other IS research methods such as introspection-based interviews or field observation. Despite some limitations, this experimental methodology enables identification and manipulation of variables and lends itself easily to statistical analysis. Although not all these tasks are directly part of interpreting per se, they may serve to measure specific relevant subskills. The following section is an overview of empirical studies focusing on the mental lexicon of interpreters. Benefits of psycholinguistically oriented experimental research to Interpreting Studies will be highlighted. Moser-Mercer is one of the greatest proponents of cognitive studies in IS. She applied psycholinguistics to IS as early as in the late 1970s (Moser 1978) and has frequently stressed the importance of interdisciplinary research in IS (Moser-Mercer



Interpreting Studies and psycholinguistics 

1997:€ 195). Gile, another psycholinguistically oriented researcher, claims that psycholinguists work “with more precision, logic and depth than practisearchers” (1994:€156). Indeed, the methodology of cognitive studies involves an impressive range of experimental methods with well defined, isolated and controlled variables. One of the latest psycholinguistic contributions to IS is a series of experiments conducted by Christoffels in cooperation with De Groot and Kroll (2004, 2005, 2006). All three authors are cognitive scientists with a strong background in psycholinguistics and notice differences between terminology used in IS and “the standard terminology in psycholinguistic studies on bilingual control” (De Groot and Christoffels 2006:€199). However, terminological differences are by no means a hindrance to interdisciplinary research efforts. In her 2004 study, Christoffels concentrated on lexical retrieval and working memory, understood as “two possible subskills of SI” (2004:€ 60). Her subjects were twenty four students with no background in interpreting. Christoffels hypothesised that if these two subskills were important for natural interpreting, there should be intra-subject correlations between their performance in lexical retrieval and working memory tasks and their performance in simultaneous interpreting (Christoffels 2004:€60). To test lexical retrieval efficiency, a word translation task and a picture naming task were administered to the subjects. Students were asked to give translation equivalents of words and name objects appearing on the screen as quickly as possible. The stimuli were controlled for word frequency and cognate status to avoid skewed results. Reaction times were measured by a sound-activated switch. Such a setting allowed for the collection of precise data. Working memory was examined by means of a reading-span task and a digit span task. Both methods are well established in memory studies. In the former, subjects are requested to remember last words of sentences presented in series. In the latter, they have to repeat increasingly longer digit sequences. The average digit span is “the magical number seven, plus or minus two” (Miller 1956). Again, words used in the reading-span task were controlled for length and frequency. The simultaneous interpreting task was a traditional experimental task in IS and did not include any methodology specific to cognitive studies. Christoffels conducted a statistical analysis, including a correlational analysis, to examine correlations between lexical retrieval and memory tasks and the interpreting task. She found that faster reaction times in a word retrieval task and reading span were associated with better interpreting performance, while digit span and picture naming were not good predictors of the natural ability to interpret simultaneously (Christoffels 2004:€ 68). This study could lead to recommendations for aptitude testing in conference interpreting schools. To uncover the intricacies of skilled linguistic control performed by professional interpreters, Christoffels, De Groot and Kroll (2006) administered a battery of tasks similar to the above-described study (picture naming, word translation, reading span, speaking span, word span) to three populations of bilinguals: ordinary bilinguals

 Agnieszka Chmiel

without interpreting experience, professional interpreters and language teachers. The hypothesis was as follows: “If interpreting is a specific skill that does not affect the more basic components of language processing, then all of the participants should perform similarly on simple language processing tasks in their first and second languages” (Christoffels et al.€2006:€326). Whereas interpreters differed significantly on all measured subskills from non-interpreting bilinguals, the nature of discrepancies between interpreters and language teachers viewed as proficient users of both of their languages varied. Interpreters did not outperform teachers in word retrieval, but they turned out to have a better working memory. The authors concluded that word retrieval was not uniquely related to SI and was not enhanced by conference interpreting experience (2006:€341). These results are at a variance with findings by Bajo et al.€(2000), who discovered that interpreters outperformed other non-interpreting professionals with high second language competence in such tasks as comprehension, lexical decision, categorization and suppression, thus suggesting that interpreters, due to their training and experience, show more efficient lexical and semantic access alongside better working memory. However, these discrepancies might be due to a different research methodology involving different tasks (production vs. comprehension), different control groups (teachers vs. other non-interpreting professionals) and different sample sizes. It would be interesting to see if similar differences in lexical processing and memory tasks could be found when comparing interpreters and translators. The studies by Christoffels and her associates focused, among other things, on word retrieval as a function of the mental lexicon. How the mental lexicon of bilinguals and polyglots is organized is one key research question in psycholinguistics. Interpreters could be an extremely valuable group of subjects for studies in this area. More specific benefits will be discussed in the next section.

2. How Interpreting Studies may contribute to psycholinguistics Synergy between psycholinguistics and Interpreting Studies is possible since both disciplines can contribute to and benefit from interdisciplinary research involving both. This section will focus on potential contributions of interpreting research to psycholinguistics, in particular to studies on bilingualism and the mental lexicon. The first and foremost advantage is obviously related to the specificity of interpreters as subjects in studies. They represent a special case of bilingualism since the way they use their working languages is exceptional.€ Non-interpreting bilinguals usually operate in either monolingual or bilingual mode (Grosjean 2001) depending on their interlocutors. Interpreters use a specific bilingual mode with either active or inactive input and output mechanisms in each language (Grosjean 2001:€18). Some researchers have even posited the existence of a distinct ‘interpreting mode’ (Heltai, personal communication). The fact that interpreters constantly ‘juggle’ with words from their working languages



Interpreting Studies and psycholinguistics 

should have some specific influence on the organisation of languages, and mental lexicons in particular, in their mind. The approach to the mental lexicon advocated in this paper is a connectionist one. The mental lexicon is seen as a giant web with nodes linked by connections of varying weights. Long-term interpreting experience may restructure the mental lexicon in a specific way. As De Groot and Christoffels claimed: […] any translation act will become reflected in a memory trace that connects the two terms of the translation; the more often the same two terms (words or longer phrases) co-occur in a translation act, the stronger the memory connection between them will be (De Groot and Christoffels 2006:€198).

Therefore, interpreters may prove extremely useful in studies on the structure and flexibility of the mental lexicon. Longitudinal research involving interpreting trainees and professionals from the outset of their career, could monitor changes in interlingual connections occurring over time and with practice. Intervals between checkpoints could be as short as two or three years because this is normally the duration of an interpreting course at university. During that time, students are exposed to many hours of intensive practice, which does not leave their bilingual processing unchanged. Additionally, it is possible to create a fairly homogenous group of experimental subjects including interpreters with the same working languages, a similar number of years in the profession, similar second-language acquisition histories, etc. Thus, unlike other interesting subjects such as aphasics, interpreters give researchers the possiblity of experimenting with groups as opposed to case studies. As mentioned above, interpreters are interesting because they offer a special case of bilingualism. Interpreters working in a single direction from one or several source languages into a single target language (‘unidirectional interpreters’) might be even more interesting because they represent a special case of interpreting. According to established practice in some countries (e.g.€Switzerland) and professional standards in various international institutions (the European Commission, the European Parliament), interpreters are usually allowed to work into their native language only (with the exception of some having languages from New Member States, such as Slovene or Polish, where retour is possible in the EU institutions). Thus, interpreters with one A language and four or five C languages are not uncommon. This professional standard is a very specific setup of bilingual, or rather multilingual functioning. In terms of Grosjean’s model (2001), the output mechanism will be permanently activated in language A only, while the input mechanism will be active in languages A and C. In the connectionist perspective adopted here, it might be posited that interlingual lexical links in the unidirectional interpreter’s lexicon will be asymmetrical – which makes for an interesting object for psycholinguistic studies. (Trainee interpreters can also be examined in a longitudinal study to see the asymmetry creation process at work.) Other potential contributions of IS to psycholinguistics include experimental tasks, error analysis and skill development. Interpreting offers an array of techniques

 Agnieszka Chmiel

that can be used in bilingual and polyglot research. These include simultaneous interpreting (used by Christoffels in her 2004 study), sight translation and shadowing. These experimental tasks, as performed by non-interpreting bilinguals, can help shed more light on language processing. Error analysis in interpreting-related tasks is also beneficial, especially as regards various aspects of language processing under cognitive load (code-switching, inhibition, self-monitoring, inflectional errors, interference). Due to the cognitive complexity of interpreting, it is easy to elicit various errors for analysis of language processing under saturation conditions. Finally, the development of interpreting skills can be easily observed in longitudinal studies of trainees. Therefore, the effects of intensive experience under specific cognitive conditions on language processing and the structure of the mental lexicon can be observed.

3. COGSIMO – objectives and work-in-progress report COGSIMO is the author’s post-doctoral research project that aims at leveraging the synergy of psycholinguistics and Interpreting Studies. The project is to explore psycholinguistic processing and conference interpreting skills with a view to establishing efficient aptitude testing procedures for interpreter trainees on the basis of cognitive predictors of success. The project offers an interdisciplinary approach to the investigation of simultaneous interpreting skills since it applies advanced psycholinguistic experimental methodology to examine language processing in three populations – ‘unidirectional’ and bidirectional professional conference interpreters and interpreting trainees. The latter group will be examined longitudinally – at the beginning and at the end of their training – to identify the effects of practice on their cognitive/linguistic makeup and bilingual mental lexicon and see which cognitive variables predict their success in training and which linguistic skills develop considerably as a result of their activities during training. It is also important to use ‘unidirectional’ interpreters (such as staff interpreters employed by DG Interpretation in Brussels) and ‘bidirectional’ interpreters (active on the Polish market) as subjects. Many directionality studies compare the performance of the same interpreters in both directions. Just as ‘bidirectional’ conference interpreters are a special case of bilinguals (since they actively use both of their working languages in the same context), EU staff interpreters are a special case of conference interpreters (due to their unidirectionality). Their mental lexicons might be structured differently. COGSIMO will hopefully offer some insight into the interlingual links between the interpreter’s working languages. The experimental design includes processing of cognates (management of lexical links, inhibition of true and false cognates) following visual (sight translation) and verbal input (CI), masked semantic priming, lexical retrieval, word translation tasks and aptitute tests (digit span, word span, recall after suppression and semantic verbal fluency). COGSIMO will contribute both to psycholinguistics (data on interlingual links and lexical processing of unique subjects) and to Interpreting Studies (revision



Interpreting Studies and psycholinguistics 

of existing information-processing models, training recommendations). Through identifying predictors of success in interpreter training, the project will also offer streamlined aptitude testing tasks and online self-assessment tools for candidates to interpreting programmes. Two pilot studies have been conducted so far. One of them focused on the processing of cognates in various modalities (Chmiel 2007a). The other examined semantic verbal fluency of interpreter trainees (Chmiel 2007b).

3.1

Processing of cognates

The study included a group of 10 professional interpreters and 25 interpreting trainees who were asked to perform sight translation (visual input) and simultaneous interpreting (audio input) of texts with true and false cognates into their A and B language. It was assumed that true cognates would be beneficial to interpreters as phonological similarity facilitates word retrieval in the production stage of interpreting, whereas false friends would be troublesome as they require extra mental effort for inhibition of their false cognate equivalents (Sanchez-Casas and Garcia-Albea 2005). The organisation of the mental lexicon was taken to be based on a connectionist principle, which assumes that relevant interlingual lexical links get activated while irrelevant links get suppressed. It was hypothesised that professional interpreters would manage cognates more efficiently than students and that visual input would generate more interference due to the lack of deverbalisation. Additionally, it was postulated that for true cognates professionals would use non-cognate TL counterparts more frequently than students, which would be due to increased inhibition to avoid transfer and conscious resistance to linguistic interference (Seleskovitch 1978). These hypotheses were partially corroborated in the study. We confirmed better management of cognates by professionals, including more successful inhibition of false cognates. Professionals did not use noncognate TL counterparts for true cognates more frequently than students. Thus, they did not avoid positive transfer for acceptable words of phrases. There was no pronounced difference between audio and visual processing except for true cognates in sight translation, which shows good cognate management skills and resistance to visual interference. De Groot and Christoffels stated that “bilingualism may turn bilinguals into experts in inhibitory control” (2006:€194). This study showed that interpreters might be even better at inhibitory control due to the specific linguistic setup of interpreting. The study also revealed possible reasons for inefficient processing of false cognates by interpreting trainees. Due to insufficiently strong interlingual links between translation equivalents wrong links between false cognates were not sufficiently suppressed. Moreover, inhibition was less effective due to the lack of attentional resources (saturation with other efforts, spillover effects – Gile 1995). Sometimes, unsuccessful interpreting resulted from the students’ insufficient language proficiency. The experiment confirms the idea that interpreters and interpreting trainees are an interesting population in bilingual research. By comparing experienced professionals

 Agnieszka Chmiel

to novices, it is possible to observe the process by which connections in the mental lexicon are built and strengthened. Interlingual connections between true and false cognates are a unique type of lexical links in the interpreters’ mental lexicons. Due to their specific nature they can offer additional insight into the organisation of the bilingual’s mental lexicon and open further possibilities for psycholinguistic investigation.

3.2

Semantic verbal fluency

The other pilot study was a longitudinal experiment with students of conference interpreting as subjects (Chmiel 2007b). Semantic verbal fluency was understood as lexical accessibility for production as measured by the quantity of words produced within a certain time and usually within a restricted semantic category. It indicates how fast and easily words and sentences are generated (Moser-Mercer et al.€2000:€123). MoserMercer and her team conducted a similar study comparing the performance of professional interpreters and trainees. They discovered no major differences in verbal fluency performance between the two groups. In the study by Chmiel, interpreter trainees of the Conference Interpreting Programme at the Adam Mickiewicz University in Poznań participated in two experimental trials. The winter trial was held after 40 weeks of training (approx. 870 hrs). Students were retested after an additional 20 weeks of training (480 hrs). The average results for the summer trial were better than the winter trial results and the difference was statistically significant (t=2.19, p