TRANSLATION AND MEANING PART 10 Proceedings of the Łódź Session of the 5th International Maastricht - Łódź Duo Colloquium on “Translation and Meaning”, Held in Łódź, Poland, 16 – 19 September 2010 BARBARA LEWANDOWSKA-TOMASZCZYK Institute of English Studies University of Łódź, Poland and MARCEL THELEN Zuyd University of Applied Sciences Maastricht School of Translation and Interpreting Maastricht, The Netherlands editors Technical editor: Marcel Thelen Zuyd University of Applied Sciences Maastricht School of Translation and Interpreting II ŁÓDŹ SESSION ISBN 978-90-801039-9-3 NUR 812 © 2013 Zuyd University of Applied Sciences, Maastricht School of Translation and Interpreting, Maastricht (The Netherlands) All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) or other means without written permission from the publisher. This book was printed and bound by UPM, Universitaire Pers Maastricht. Source of Polish text on front cover: Jerzy Pilch – ‘Moje pierwsze samobójstwo’ Świat Książki Warszawa 2006. transl. by David Trick ‘My first suicide’. Open Lester Rochester, NY. 2012 Translation and Meaning Marcel Thelen and Barbara Lewandowska-Tomaszczyk, editors. 1. Proceedings of the Maastricht Session of the 1990 Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Maastricht, The Netherlands, 4-6 January 1990. Marcel Thelen and Barbara Lewandowska-Tomaszczyk, editors. ISBN 90-726140-3-8. 2. Proceedings of the Łódź Session of the 1990 Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Łódź, Poland, 20-22 September 1990. Barbara Lewandowska-Tomaszczyk and Marcel Thelen, editors. ISBN 90-801039-1-8. 3. Proceedings of the Maastricht Session of the 2nd International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Maastricht, The Netherlands, 19-22 April 1995. Marcel Thelen and Barbara Lewandowska-Tomaszczyk, editors. ISBN 90-801039-2-6. 4. Proceedings of the Łódź Session of the 2nd International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Łódź, Poland, 22-24 September 1995. Barbara Lewandowska-Tomaszczyk and Marcel Thelen, editors. ISBN 90-801039-3-4. 5. Proceedings of the Maastricht Session of the 3rd International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Maastricht, The Netherlands, 26-29 April 2000. Marcel Thelen and Barbara Lewandowska-Tomaszczyk, editors. ISBN 90-801039-4-2. 6. Proceedings of the Łódź Session of the 3rd International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Łódź, Poland, 22-24 September 2000. Barbara Lewandowska-Tomaszczyk and Marcel Thelen, editors. ISBN 90-801039-5-0. 7. Proceedings of the Maastricht Session of the 4th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Maastricht, The Netherlands, 18-21 May 2005. Marcel Thelen and Barbara Lewandowska-Tomaszczyk, editors. ISBN 978-90-801039-6-2 8. Proceedings of the Łódź Session of the 4th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Łódź, Poland, 23-25 September 2005. Barbara Lewandowska-Tomaszczyk and Marcel Thelen, editors. ISBN 978-90-801039-7-9. 9. Proceedings of the Maastricht Session of the 5th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Maastricht, The Netherlands, 19-22 May 2010. Marcel Thelen and Barbara Lewandowska-Tomaszczyk, editors. ISBN 978 90 801039 8-6 10. Proceedings of the Łódź Session of the 5th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”, Held in Łódź, Poland, 16-19 September 2010. Barbara Lewandowska-Tomaszczyk and Marcel Thelen, editors. ISBN 978 90 801039 9-3 Orders should be sent to: Dr. Marcel Thelen Maastricht School of Translation and Interpreting, Zuyd University of Applied Sciences P.O. Box 634, NL-6200 AP Maastricht Tel.: + 31 43 346 6471 / Fax: + 31 43 346 6649 / e-mail: marcel.thelen@zuyd.nl TRANSLATION AND MEANING, PART 10, 2013 III ACKNOWLEDGEMENTS With the last volume in the cycle Translation and Meaning, the editors wish to express their thanks to the many people with whom we have been cooperating for over twenty years both in the organization of the most successful “Translation and Meaning” colloquia in Maastricht and in Łódź and in the publication of the eleven volumes (i.e. 10 volumes in the Translation and Meaning series and 1 volume containing an anthology of selected papers), which present a record of our common interest and involvement in the field as well as the dynamically changing research and educational paradigms in Translation Studies. Words of thanks are addressed to those numerous participants from numerous different countries and cultures, speaking different languages and offering new insights, and to the plenary speakers who participated in our colloquia, and provided lasting inspiration to further research and creative practice in translation. We wish to express our appreciation of the continuing support we have experienced from the authorities of Zuyd University of Applied Sciences in Maastricht and the University of Łódź over the years of our cooperation. Special thanks are due to the colleagues in our departments and institutes and to the numerous students who have been involved in the organization of the colloquia. We are happy to see that the ideas originating from the meetings and publications are being carried forward and contribute to the advancement of translation theory and practice, and are confident that the series of colloquia will be continued in the future. Łódź Barbara Lewandowska-Tomaszczyk Chair of the Łódź Organising Committee Maastricht Marcel Thelen Chair of the Maastricht Organising Committee ŁÓDŹ SESSION IV PREFACE The present volume of proceedings, titled Translation and Meaning, Part 10, contains the papers, both plenary and non-plenary, of the Łódź Session of the 5th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”. The Łódź Session took place from 16 to 19 September 2010 in Łódź (Poland). It was preceded by the Maastricht Session from 19 to 22 May 2010 in Maastricht (The Netherlands). The first four Maastricht-Łódź Duo Colloquia took place in 1990, 1995, 2000 and 2005. Also then both the Maastricht Session and the Łódź Session took place in one and the same year, something that is one of the important characteristics of the Maastricht-Łódź Duo Colloquiua. All these Duo Colloquia focused on the way that meaning figures in translation (hence the title of “Translation and Meaning”), with special emphasis in Maastricht on practice, and on theory in Łódź. All Duo Colloquia were organised jointly by the Maastricht School of Translation and Interpreting of Zuyd University of Applied Sciences and the Institute of English of the University of Łódź. All proceedings were published as issues in the series of Translation and Meaning. Barbara Lewandowska-Tomaszczyk Chair of the Łódź Organising Committee Marcel Thelen Chair of the Maastricht Organising Committee TRANSLATION AND MEANING, PART 10, 2013, V -VII V TABLE OF CONTENTS Introduction BARBARA LEWANDOWSKA-TOMASZCZYK AND MARCEL THELEN 9 SECTION I: COMMUNICATION, SEMANTICS, INTERSEMIOTICS Bedeutung – Sinn – Funktion:was wird eigentlich übersetzt? CHRISTIANE NORD, Germany 17 Parameter variability in translational approximation BARBARA LEWANDOWSKA-TOMASZCZYK, Poland 27 Paintings re-seen. poems re-read. A few remarks on intersemiotic translation AGATA HOLOBUT, Poland 39 Translating cognitive blends in political discourse Poland 53 RAFAL PACHOLCZYK, Les paramètres socioculturels de la traduction de textes humoristiques MAGDALENA SZEFLINSKA-BARAN, Poland 71 Dolores de la traducción, or Lolita in translation JANUSZ WOBLEWSKI, Poland 79 SECTION II: TRANSLATION & LANGUAGES IN CONTRAST Interrogation, hypothèse, argumentation. Le français et l’anglais en contraste AGNES CELLE, France 93 Traduire des jeux de mots du français vers le polonais. Exemple de René Goscinny ANNICK CATALDI, Poland 109 “Translation language” in Polish versions of the Gospel of Mark – a case study ALEKSANDER GOMOLA, Poland 121 A corpus-driven study of translation universals: The case of English and Russian “Lolita” and reference corpora of literary texts ŁUKASZ GRABOWSKI, Poland 129 The semantics of Northern Sotho verbs as translated from English: Comments on the African languages WordNet project MAMPAKA LYDIA MOJAPELO, South Africa 143 Translation and forms of address: An experience in Tshivenḓa South Africa 157 MUNZHEDZI JAMES MAFELA, Calques in Polish translations of English press artricles: linguistic innovations or mistranslations? MIROSLAWA PODHAJECKA, Poland 165 VI ŁÓDŹ SESSION Génologie contrastive au service de la traduction des textes de spécialité Poland 181 BARBARA WAŁKIEWICZ, SECTION III: THE TRANSLATOR & TRANSLATOR EDUCATION Being a translator: how does it feel? Poland 197 JOANNA ALBIN, Non-formal elements in academic translator education: Why, how and what for? KONRAD KLIMKOWSKI, Poland 211 Kognitive Prozesse in der translatorisch gemittelten interkulturellen Kommunikation Poland 221 ANNA MAŁGORZEWICZ, Zur Wahl der Texte für die Bildung von Übersetzern Poland 231 MARCIN MICHOŃ, Einige Argumente für die Verwendung der Regenbogenpresse als Einstieg ins Übersetzungstraining MARIUSZ MILCZAREK, Poland 239 Teaching, training, educating? Terminological ambiguity in translator education theory PAULINA PIETRZAK, Poland 245 Translational competence and the native speaker PIUS TEN HACKEN, United Kingdom 253 SECTION IV: TERMINOLOGY The use of German terminology as a euphemizing strategy in Hebrew 1930-1970 Israel 265 NITSA BEN-ARI, Die Fallbearbeitung - Eine Möglichkeit des Einstiegs in die Deutsche Rechtssprache MAGDALENA DUŚ, Poland 275 Le lexique de spécialité dans l’apprentissage et dans la traduction Poland 281 JACEK FLORCZAK, Le latin juridique à l’épreuve de la traduction Poland 291 KSENIA GAŁUSKINA, Legal terms and meaningful units. A corpus-based study of the term discovery in the United States Supreme Court opinions STANISŁAW GOŹDŹ-ROSZKOWSKI, Gdańsk/Poland 299 The challenge of meaning construction: expert vs. non-expert processing Poland 307 EWA KOŚCIAŁKOWSKA-OKOŃSKA, Semantic ambiguity and the translation of medical terminology Slovakia 317 RENÁTA PANOCOVÁ, TRANSLATION AND MEANING, PART 10, 2013, V -VII VII Übersetzungen von Phraseologismen in der Automobilpresse KRZYSZTOF SAKOWSKI, Poland 325 Methods for problem solving in translation and terminology: WordNet and Frames vs. LexicalSystem-like Structures MARCEL THELEN, The Netherlands 331 Medical translation in an academic setting: demands on the training in terms of content and practice CORNELIA WERMUTH, Belgium 349 SECTION V: INTERPRETING & AUDIOVISUAL TRANSLATION Théorie du sens revisited. CLI in the target language output of simultaneous interpreters MARIA BAKTI, Hungary 363 Audiovisual transfer: where do we go from here? Poland 371 ŁUKASZ BOGUCKI, Construal manipulation in subtitling MIKOŁAJ DECKERT, Poland 377 Lost in translation – Mission (im)possible in voice-over translations? Poland 387 MAGDALENA MACHCINSKA-SZCZEPANIAK, Traduire le sens d’une production médiale Poland 393 TERESA TOMASZKIEWICZ, FINAL REMARKS Current translation discourses and the future BARBARA LEWANDOWSKA-TOMASZCZYK & MARCEL THELEN 409 Survey of papers included in Translation and Meaning, Parts 1 – 10 413 MARCEL THELEN AND BARBARA LEWANDOWSKA-TOMASZCZYK List of contributors MARCEL THELEN 429 Index MARCEL THELEN 433 VIII ŁÓDŹ SESSION TRANSLATION AND MEANING, PART 10, 2013, IX - XIII IX INTRODUCTION 1. The Łódź Session of the 5th International Maastricht-Łódź Duo Colloquium The Łódź Session of the 5th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning” was organized on 16-19 September 2010 by the Institute of English of the University of Łódź. The 2010 Łódź Session was geared towards the theory of translation, and dealt with theoretical aspects of the central theme of the Duo Colloquium and the sub-themes suggested for it. The central theme of all the colloquia organised in this series is Translation and Meaning, while the sub-themes for 2010 Łódź Session were: A) meaning in translation/interpreting from the point of view of: 1 (a) cognitive sciences (e.g. think-aloud protocols and other modes of reflection, prototypicality, etc.) (b) computer-assisted translation training, electronic teaching/learning environments, distance teaching/learning (c) corpus linguistics and language corpora (d) history of translation (e) lexicology, lexicography (f) local/national/international translation/interpreting programmes (g) machine translation, machine-aided translation, other computer aids, combinations of computer aids (h) new theoretical/practical developments (i) philosophy / the philosophy of language, etc. (j) pragmatics, semantics, semiotics (k) psycholinguistics, psychology (l) sociolinguistics/sociology, ~ and minority status, institutionalised multilingualism, interculturality, cultural identity (m) software localisation (n) subtitling, dubbing, sign language, voice-over (o) terminology, conceptuology, expert systems (e.g. term-word distinction, automated term extraction, etc.) (p) text linguistics (e.g. text typology, text analysis, etc.) (q) the interplay between theory and practice (r) translation theory, translation studies (s) various types of translating/interpreting 2 (a) the relation between translator/interpreter and client/customer (b) the student of translation/interpreting (c) the translation trainer/the trainer of interpreting (d) the translator/interpreter B) meaning in the translation/interpretation of: 1 (a) the language of economics / business (b) the language of law / politics (c) the language of media and the arts (d) the language of medicine / biology (e) the language of science / technology / IT (f) the language of tourism/leisure 2 (a) holy scriptures (b) literature (c) LSP vs. general language vs. phraseology vs. terminology, etc. ŁÓDŹ SESSION X (d) allusions, figurative language, jokes, metaphors, etc. C) quality assessment of translation/interpreting and modes/modalities of training D) The role of translation methods/strategies in: 1 the actual practice of translating/interpreting 2 the training of translation/interpreting This list was not exhaustive. The papers in the present volume of proceedings do not cover all of these sub-themes, and some papers deal with a combination of sub-themes. The main focus of the Łódź Session was on a number of relevant sub-themes of the main topic of the colloquium, i.e. translation and meaning, frequently from a perspective of language and translation theories. For these proceedings, the following sections were established: Section I: Section II: Section III: Section IV: Section V: Communication, Semantics, Intersemiotics Translation & Languages in Contrast The Translator & Translator Education Terminology Interpreting & Audiovisual Translation 2. Proceedings of the Łódź Session The Łódź Session took place from 16 to 19 September 2010 at the Łodź University Conference Centre in Łódź and attracted over 75 participants from 21 countries: Algeria, Argentina, Belgium, Brazil, France, Germany, Hong Kong, Hungary, Israel, Italy, Japan, Kuwait, Lithuania, Poland, Republic of South Africa, Russia, Slovakia, Spain, Taiwan, The Netherlands, and United Kingdom. 3. General description of the papers in these proceedings The volume we present to the reader is the tenth, and last, in the series of publications, composed primarily of the papers presented at the joint cyclical colloquia on Translation and Meaning, which were organized as a two-session event every five years at Zuyd University of Applied Sciences in Maastricht and the University of Łódź since 1990 until 2010. Thirty six contributions in the present volume are structured into five main parts and completed with final remarks by the volume editors to present a current picture of the scene of translation studies and sketch relevant advances in translation theory and practice. Section I of these proceedings (Communication, Semantics, Intersemiotics) comprises six papers. Christiane Nord’s (Germany) contribution “Bedeutung – Sinn – Funktion: was wird eigentlich übersetzt?” offers a fresh look at the relationship between sense, reference and function mapped from the Source to the Target language realities. The author foregrounds the role of the communication objective in the original and its re-formulation in the translation context, addressed to a different Addressee in the TL culture. The intended communicative function is taken to act as tertium comparationis in the context of translation. The author discusses possible uses of the four-function model in the translation practice and education. In her paper “Parameter variability in translational approximation”, Barbara Lewandowska-Tomaszczyk (Poland) develops the concept of communicative approximation in same-language communication and in translation based on the author’s idea of translational re-conceptualization as an outcome of some approximation processes, first proposed in her earlier publications. The present discussion offers a set of parameters which undergo modification between SL and TL texts and shows the scope and range of the processes of translational reconceptualization. The third paper in Part I “Paintings re-seen, poems re-read. A few remarks on intersemiotic translation” by Agata Holobut (Poland), takes up intersemiotic translation and the relationship between the visual and the verbal with reference to painting and poetry. The notion of TRANSLATION AND MEANING, PART 10, 2013, IX - XIII XI intersemiotic translation is re-defined and its relevance discussed both with respect to translation studies and to art and literary criticism. In his paper “Translating cognitive blends in political discourse”, Rafał Pacholczyk (Poland) uses the familiar Cognitive Linguistic concept of conceptual integration (blending) in the anaysis of Polish-English translation in political context, while Magdalena Szeflińska-Baran (Poland) focuses on modification of Attardo’s Verbal Humour Theory by introducing extralinguistic considerations to the field in the paper ”Les paramètres socioculturels de la traduction de textes humoristiques”. The paper “Dolores de la traducción, or Lolita in translation ” by Janusz Wróblewski (Poland) investigates some of the wordplay techniques as used in English to Polish translatiom of the novel. Section II of the volume (Translation & Languages in Contrast) contains seven contributions. The first three papers deal with the French language as a Source and/or Target language in the analysis. In the paper “Interrogation, hypothèse, argumentation.le français et l’anglais en contraste”, Agnès Celle proposes an account of the rhetorical function of interrogative and hypothetical clauses in the translations of both French-to-English and English-to-French academic discourse from a contrastive point of view. In the paper “Traduire des jeux de mots du français vers le polonais. Exemple de René Goscinny”, Annick Cataldi focuses on the study of the French-to-Polish translation of word play and idioms in selected comics series (Asterix and Little Nicholas) by René Goscinny. In the paper “Translation language” in Polish versions of the Gospel of Mark – a case study”, Aleksander Gomola presents a lesser known translation of the Gospel of Mark into Polish and makes an attempt to identify the translator’s interpretive contribution to the Target Language version. In his paper “A corpus-driven study of translation universals: the case of English and Russian “Lolita” and reference corpora of literary texts”, Łukasz Grabowski uses a corpus of literary texts. The author compares basic patterns of lexical use in the original and its translation in his attempt to identify data in support of translation universals. The next two papers focus on South African languages. With reference to an English-based African WordNet project, Mampaka Lydia Mojapelo discusses the semantics of Northern Sotho verbs as translated from English, while in the paper “Translation and forms of address: an experience in Tshivenḓa”, Munzhedzi James Mafela highlights the translation problems in certain forms of address in the Tshivenḓa-to-English translated texts, concentrating primarily on honorific pronouns. In her paper, “Calques in Polish translations of English press artricles: linguistic innovations or mistranslations?”, Mirosława Podhajecka (Poland) argues that linguistic innovations are often due to the uncritical adoption by modern language users of deviant expressions as instances of correct language. Barbara Walkiewicz (Poland) focuses on the place and function of genre in her paper “Génologie contrastive au service de la traduction des textes de spécialité” and treats it in terms of a cluster related by family resemblance in a discursive polysystem in particular cultures. Five papers in Section III of the present volume focus on the place and function the The Translator & Translator Education. Joanna Albin asks ‘How does it feel to be a translator?’ and presents the results of her study on the affective aspects of translator profession, pointing to the importance of their social conditioning. Anna Małgorzewicz focuses on the mental processes in translation. In the paper “Kognitive Prozesse in der translatorisch gemittelten Interkulturellen Kommunikation”, the author investigates the proposed mental respresentations of the Source and Target texts. In his paper “Non-formal elements in academic translator education: why, how and what for?”, Konrad Klimkowski proposes that the curriculum of academic translator education should introduce, as a standard practice, authentic real-life projects to the translation classroom. Marcin Michoń discusses the role of educational texts in the translator training classroom in his paper “Zur Wahl der Texte für die Bildung von Übersetzern”, while Mariusz Milczarek argues in his paper “Einige Argumente für die Verwendung der Regenbogenpresse als Einstieg ins Übersetzungstraining” for using less complex tasks in the context of translation education to ensure learners’ step-by-step road to success. In her paper “Teaching, training, educating? Terminological ambiguity in translator education theory”, Paulina Pietrzak (Poland) asks two important questions who translators are and where they come from and tries to investigate this area more fully. Pius ten Hacken (United Kingdom) discusses the XII ŁÓDŹ SESSION place of the Native Speaker and so-called Native Speaker Principle in translation practice and theory of the generative linguistic provenance in his contribution "Translational competence and the native speaker”. Section IV of the volume presents a discussion and analysis of Terminology and its links with translation theory and practice. The theme of terminology is first tackled in Nitsa Ben-Ari’s (Israel) paper “The use of German terminology as a euphemizing strategy in Hebrew 1930-1970”, in which the author presents the use of German terminology as a strategy in Hebrew non-fiction paramedical texts refering to erotic and sexual themes. Magdalena Duś’s (Poland) contribution enriches the discussion of legal language in her paper “Die Fallbearbeitung - Eine möglichkeit des Einstiegs in die deutsche Rechtssprache”. The author takes up the theme of German legal language as exemplified by case solutions in terms of problems encountered with regard to their translation. In Jacek Florczak’s (Poland) contribution “Le lexique de spécialité dans l’apprentissage et dans la traduction”, the author puts forward a model of contrastive analysis of a typology of semantic relations between Polish and French specialized terminology. Ksenia Gałuskina (Poland) further develops the issue of the specificity of legal discourse by concentrating in her paper “Le latin juridique à l’épreuve de la traduction” on the analysis of Latin expressions and maxims implemented in legal Polish and French in the context of comparative legal linguistics. In “Legal terms and meaningful units. A corpus-based study of the term discovery in the United States Supreme Court opinions”, Stanisław Gożdż-Roszkowski (Poland) demonstrates a methodology of term analysis by focusing on various types of legal terms textual recurrence, specifically the term discovery, central to the US legal practice. Ewa Kościałkowska-Okońska (Poland) focuses on the concepts of expert and non-expert competence in her paper “The challenge of meaning construction: expert vs. non-expert processing”. The author addresses primarily the role of motivation, experience and creativity as factors central to meaning construction and transfer. Renáta Panocová (Slovakia), discusses in her paper “Semantic ambiguity and the translation of medical terminology” the problem of finding the correct equivalent terms in translating medical texts. She argues that analysing lexical paradigmatic relations betweeen terms both in the source and target text is of crucial importance. The topic of specialized vocabulary is explored by Krzysztof Sakowski (Poland) in his paper “Übersetzungen von Phraseologismen in der Automobilpresse“, which presents overlaps between general language rules and specialized automobile terminology as used in journalistic prose. In “Methods for problem solving in translation and terminology: WordNet and Frames vs. Lexical-System-like Structures”, Marcel Thelen combines the issues of terminology translation with translator education in terms of guided tasks focusing on a discovery procedure for meaning determination. Finally, the last contribution in this Part by Maria-Cornelia Wermuth (Belgium), “Medical translation in an academic setting: demands on the training in terms of content and practice”, examines an MA programme offered at the Department of Applied Language Studies (Sub-Faculty of Linguistics, Lessius/KULeuven Belgium) and focuses on the development of medical translator competencies. Section V, finally, Interpreting & Audiovisual Translation, contains five papers. The first contribution “Thèorie du sens revisited. CLI in the target language output of simultaneous interpreters” by Maria Bakti (Hungary) discusses manifestations of interference or cross-linguistic influence in the disfluencies observed in the Target Language output of trainee and professional simultaneous interpreters working from English into their native Hungarian. The paper “Audiovisual transfer: where do we go from here?” by Łukasz Bogucki (Poland) takes up a timely issue of rapid advances in technology and their impact on the development of new practices in audiovisual translation in which current norms and conventions fade and are being replaced by new patterns. Mikołaj Deckert’s (Poland) paper “Construal manipulation in subtitling” argues for the essential role of the Cognitive Linguistics paradigm as a suitable paradigm to analyse translation, in this case in subtitling practice. Magdalena Machcińska-Szczepaniak (Poland) discusses meaning reconfiguration in voice-over translation in her papere “Lost in translation – Mission (im)possible in voice-over translations?”. In the closing text of this volume, Teresa Tomaszkiewicz (Poland) reflects on the concept of meaning TRANSLATION AND MEANING, PART 10, 2013, IX - XIII XIII translation, encompassing linguistic and extralinguistic elements, in her plenary talk “Traduire le sens d’une production médiale”, particularly with regard to the multiple receiver of audio-visual translation. The last volume in the series ends with our reflection on the current state and the future of translation theory and practice. We consider in this chapter (Barbara Lewandowska-Tomaszczyk & Marcel Thelen “Current translation discourses and the future”) the state of the art as we perceive it today and sketch a map of the rising and developing fields of translation studies which await further consideration, models, and theoretical accounts, together with an adequate programme for practical implementation in different types of translation and interpreting as well as in translator education. Barbara Lewandowska-Tomaszczyk University of Łódź Marcel Thelen Zuyd University of Applied Sciences, Maastricht 14 15 Section I: Communication, Semantics, Intersemiotics 16 TRANSLATION AND MEANING, PART 10, 2013, 17 - 25 17 BEDEUTUNG – SINN – FUNKTION: WAS WIRD EIGENTLICH ÜBERSETZT? Christiane Nord University of Applied Sciences Magdeburg, Germany Abstract: Im allgemeinen Verständnis geht es bei der Übersetzung darum, die Bedeutung oder den Sinn einer ausgangssprachlichen Äußerung in einer Zielsprache wiederzugeben. In dem folgenden Beitrag soll gezeigt werden, dass Bedeutung und Sinn sich in jedem Falle ändern, wenn eine Äußerung in einer anderen Sprache und in einer anderen Situation für andere Adressaten neu formuliert wird. Als Ausweg aus dem Dilemma wird vorgeschlagen, statt Bedeutung und Sinn die (intendierte) kommunikative Funktion als Tertium comparationis für Ausgangsund Zieltext zu wählen. Für die praktische Umsetzung dieses Vorschlags in Übersetzungsdidaktik und -praxis wurde ein Vier-Funktionen-Modell entwickelt, das an verschiedenen Beispielen verdeutlicht wird. 1. Einleitung: Bedeutung und Etymologie Das deutsche Wort Übersetzen ist eine Lehnbildung zu lat. transferre (vgl. en. translate aus dem dazugehörigen Partizip Perfekt translatum), während fr. traduire, es. traducir, it. tradurre etc. direkt aus dem lateinischen Verb traducere ("hinüberführen") entstanden sind. Vielleicht liegt es daran, dass besonders im Deutschen die etymologisch-bildhafte Bedeutung des "HinüberSetzens" immer noch präsent zu sein scheint, die in dem bekannten Wortspiel von Jakob Grimm (1785-1863) so unnachahmlich ausgedrückt ist: „übersetzen ist übersetzen, traducere navem. wer nun, zur seefahrt aufgelegt, ein schiff bemannen und mit vollem segel an das gestade jenseits führen kann, musz dennoch landen, wo andrer boden ist und andre luft streicht. ("ueber das pedantische in der deutschen sprache". 1847). Aber was wird denn eigentlich "über(ge)setzt"? Das ist im allgemeinen Verständnis des Wortes nicht so eindeutig. In Internetforen, wo heute "Volkes Stimme" ganz gut zu hören ist, wird vielfach zwischen "Bedeutung" und "Übersetzung" unterschieden, wenn beispielsweise jemand nach "einer guten deutschen Übersetzung für count down" fragt und erklärend hinzufügt: Beispiel 1: Volkes Stimme Die Bedeutung des Begriffes ist mir natürlich klar. Ich suche nach einer guten Übersetzung, wie zum Beispiel: Toaster und Weißbroterhitzer… Aber auch achtbare Wörterbücher führen die Bedeutung eines Wortes auf den Ursprung zurück: Beispiel 2: Wortbedeutung auf Typus-Ebene Frühstück: spätmhd. vrüestücke, meint eigtl. das in der Frühe gegessene Stück Brot (DudenHerkunftswörterbuch, Hervorhebung C.N.) Das in der Frühe gegessene Müsli oder Spiegelei wäre dann als Metonymie zu verstehen. Dagegen sind en. breakfast, es. desayuno, fr. (petit) déjeuner Negationen von lat. ieiunus, "nüchtern", meinen alao das, womit man sich (morgens?) "entnüchtert", das können ein Keks und eine Tasse Kaffee, aber auch gebratener Hering oder Spiegeleier mit Toast sein. Was wird denn dann "hinüber-gesetzt", wenn wir de. Frühstück mit en. breakfast übersetzen? 2. Bedeutung und Sinn in der interkulturellen Kommunikation Nach Reiß/Vermeer (1984: 30) ist die "Bedeutung" einer Äußerung unabhängig von einem aktuellen Sitautionsvorkommen auf der Typus-Ebene anzusiedeln, während sich der "Sinn" erst ergibt, wenn die Bedeutung in einen konkreten Situationskontext eingebettet wird, zu dem der Sprecher/Textpro- 18 CHRISTIANE NORD duzent mit seiner kommunikativen Absicht, Zeit, Ort und Medium der Interaktion und die intendierte Funktion der Äußerung gerechnet werden. Maßgeblich für die Bestimmung von Bedeutung und Sinn sind also das Signifiant, also die materielle Seite des sprachlichen Ausdrucks, der Begriff, d.h. die psychisch-mentalen Aspekte, die an der Herstellung von Konzepten beteiligt sind, Sprecher und Situationskontext, also der pragmatische Gebrauch des Zeichens, und Referenz, d.h. die Objekte, Eigenschaften und Sachverhalte der realen Welt (vgl. Bussmann 1983). Wir fragen jeweils danach, was mit ihnen bei der Übersetzung von einer Sprachkultur in die andere geschieht. (a) (b) (c) (d) Die materielle (lautliche oder graphematische) Seite des sprachlichen Ausdrucks: z.B. Tisch [ti∫] wird bei der Übersetzung in der Regel (außer bei Entlehnung) verändert, weil ein anderer Sprachcode verwendet wird (z.B. en. table, es. mesa, fr. table). Die bei der Begriffsbildung beteiligten psychisch-mentalen Aspekte sind bei der Rezeption des Zieltexts auf jeden Fall andere, weil die zielkulturellen Empfänger durch eine andere Kultur und deren Sichtweisen geprägt sind. Die pragmatischen Bedingungen der Textproduktion und –rezeption sind ebenfalls für den Zieltext andere als für den Ausgangstext. Sonderfall: Wenn der Zieltext in einer ausgangskulturellen Situation verwendet wird (wie etwa übersetzte Touristeninformationen) ist nur der Textproduzent ein anderer (= Translator), während die Rezeptionsbedingungen die gleichen sind. Bei den Objekten, Eigenschaften, Sachverhalten der realen Welt, auf die sich eine Äußerung bezieht, kann man zwar annehmen, dass sie "objektiv" dieselben sind, es ist jedoch kein Geheimnis, dass sie je nach kulturellem Horizont, Wertesystem und anderen kulturellen Bedingungen unterschiedlich "gesehen" werden. Oft haben sie unterschiedliche prototypische Formen (siehe das Beispiel Tisch), einen anderen Status oder eine andere Funktion. Bereits intrakulturell sind Funktionen und Bedeutungen von sprachlichen Einheiten je nach Kontext unterschiedlich zu interpretieren – um wie viel mehr wird dies in verschiedenen Kulturen der Fall sein. Beispiel 3: Polysemie Wiederum am Beispiel Tisch lässt sich bereits an den folgenden Beispielsätzen zeigen, wie wenig auf der Typus-Ebene von "einer" Bedeutung gesprochen werden kann: Lasst uns zu Tisch gehen. Warum setzt du dich zum Malen nicht an den Tisch? Wenn ich übersetze, liegen auf meinem Tisch alle möglichen Wörterbücher aufgeschlagen herum. Sie saßen alle im Wohnzimmer um den Tisch herum. Tischlein, deck dich! Sie lebten getrennt von Tisch und Bett. In Asien stehen die leckersten Sachen auf dem Tisch, aber darunter kann ich meine Beine immer irgendwie nicht unterbringen. Wenn Sinn und Bedeutung also nicht geeignet sind, über die Sprach- und Kulturbarriere "hinübergetragen" zu werden, muss nach einem anderen Tertium comparationis für Ausgangs- und Zieltext gesucht werden. Aus der Sicht der funktionalistisch orientierten Skopostheorie (Vermeer 1978, Reiss/Vermeer 1984, Nord 1997) soll dafür der Begriff der Funktion vorgeschlagen werden. 3. Der Funktionsbegriff aus skopostheoretischer Sicht Die von Hans J. Vermeer zuerst 1978 vorgestellte Skopostheorie (gr. skopos = "Ziel, Zweck") beruht auf einer Reihe von "Regeln", von denen die wichtigste die sogenannte Skoposregel ist: "Die Dominante jeder Translation ist deren Zweck." In eine Handlungsanweisung umgemünzt, bedeutet das: Übersetze so, dass dein Zieltext in der Zielkultur für die zielkulturellen Adressaten die gewünschte Ziel-Funktion erfüllen kann (vgl. Reiß/Vermeer 1984: 95ff.). TRANSLATION AND MEANING, PART 10, 2013, 17 - 25 19 Diese Ziel-Funktion wird durch den Übersetzungsauftrag definiert, dessen wesentlicher Bestandteil die Bestimmung der Ziel-Adressaten ist. Da sich die Ziel-Adressaten in jedem Fall von den Adressaten des Ausgangstexts durch ihre Prägung durch eine andere Kultur mit ihren Wertesystemen und Verhaltensnormen unterscheiden, kann für die Zieladressaten die Funktion des Zieltexts eine andere sein als die Funktion des Ausgangstexts für die ausgangskulturellen Adressaten. Um herauszufinden, ob – und wenn ja, wie – der Ausgangstext für die zielkulturellen Adressaten "funktionieren" kann, bzw. in welcher Wiese er verändert werden muss, damit er funktionieren kann, brauchen wir eine übersetzungsbezogene funktionale Textanalyse. Diese besteht in einem Vergleich der Ausgangstext-Funktionen mit den durch den Übersetzungsauftrag vorgegebenen ZT-Funktionen und zeigt auf, wo sich Übersetzungsprobleme ergeben, die vom Translator durch die entsprechenden Reproduktions- oder Adaptationsverfahren zu lösen sind. Um diese Funktionsanalyse für die Übersetzerausbildung, aber dann natürlich auch für die Praxis des Übersetzens, zu erleichtern, schlagen wir ein Vier-Funktionen-Modell vor, das aus vier Grundfunktionen mit jeweils einer Reihe von Unterfunktionen besteht. Es hat sich in der Übersetzungsdidaktik für alle Arten von Übersetzungsaufgaben, unabhängig von den jeweiligen Sprachen- und Kulturpaaren, aber auch unabhängig von Textsorten und Übersetzungsaufträgen, seit langem bewährt (vgl. vor allem Nord [1988]2009, aber auch Nord 1997 und später). 4. Das Vierfunktionen-Modell (Nord) Das Vierfunktionen-Modell beruht auf einer Kombination des Organonmodells der Sprachfunktionen von Karl Bühler (1934) mit dem kommunikationstheoretisch begründeten Modell der Sprachfunktionen von Roman Jakobson (1971). Aus dem Bühlerschen Modell stammen die referentielle, die expressive und die appellative Funktion, während dem Jakobsonschen Modell die sogenannte phatische Funktion entliehen wurde. Wir betrachten zunächst die Funktionen im Einzelnen mit jeweils einem Beispiel; diese Beispiele werden dann erneut aufgegriffen um zu zeigen, was mit den Funktionen beim Übergang von einer Kultur zur anderen geschieht, bzw. welche Bedingungen gegeben sein müssen, damit sie erfüllt werden können. Die phatische Funktion ist für den Kommunikationskanal zuständig, und zwar für seine Eröffnung (z.B. durch eine Begrüßung) und seine Schließung (z.B. durch eine Abschiedsformel) sowie dafür, dass er während der gesamten kommunikativen Interaktion geöffnet bleibt. Dazu können neben Pausenfüllern (z.B. ähm) auch direkte Ansprachen an die Adressaten (z.B. wie Sie sicher wissen), Metakommunikation (z.B. dieses Beispiel hinkt vielleicht ein wenig) oder nonverbale Verhaltensweisen wie Lächeln, Blickkontakt etc. eingesetzt werden. Wenn der Kontakt optimal gestaltet werden soll, ist es auch wichtig, die zwischen den Kommunikationspartnern bestehende statusbedingte soziale Beziehung (Lehrende – Lernende, Vorgesetzter – Untergebener, Freundin – Freund etc.) nicht nur zu festigen, sondern eventuell auch zu verändern (z.B. von distanziert zu freundschaftlich, oder auch umgekehrt!). Neben der reinen Kontaktfunktion, wie sie Jakobson beschreibt, würde ich daher der phatischen Funktion auch die Aufgabe, das soziale Rollenverhältnis zwischen den Kommunikationspartnern zu regeln. Das ist besonders etwa für die Wahl der angemessenen Anredeformen von Bedeutung. Beispiel 4: Einleitungssatz eines Touristentexts über Münchner Spezialitäten SPEZIALITÄTEN "Liebe geht durch den Magen." Dieser Spruch findet in München seine besondere Bestätigung. Denn es gilt als ein Teil der vielzitierten Münchner Gemütlichkeit, dass man hier auch zu essen und zu trinken versteht. Probieren Sie deshalb zuerst, was unter "Schmankerl" als Münchner Spezialität auf der Speisenkarte aufgeführt ist: Die Weißwurst… Bei deutschen Touristeninformationen findet sich sehr häufig ein Sprichwort, ein Zitat oder eine 20 CHRISTIANE NORD bekannte Redewendung als "Aufhänger" für das Thema. In diesem Text geht es ja nicht um die Liebe, sondern um Essen und Trinken, und die Redewendung dient dazu, zum Thema hinzuführen, hat also rein phatische Funktion. Die referentielle Funktion (bei Bühler: Darstellungsfunktion) ist für die Vermittlung von Informationen über Objekte und Phänomene der außersprachlichen Welt zuständig, also für die Denotate von sprachlichen oder nichtsprachlichen (z.B. durch Zeigen realisierten) Referenzen. Unterfunktionen sind unter anderem die Beschreibung von Gegenständen oder Vorgängen (z.B. in einer Produktdokumentation), die Anleitung (z.B. zum optimalen Gebrauch eines Geräts, zur Herstellung eines Gerichts in einem Rezept), das Berichten über Ereignisse (z.B. in der Zeitung, in einem Protokoll), das Erläutern (von Sachverhalten, Begründungen, Folgen) oder die Deklaration (z.B. Tauf- oder Trauformeln, Übernahme von Haftung oder Verantwortung in juristischen Texten). Beispiel 5: 600 Jahre Universität Heidelberg AUS TRADITION IN DIE ZUKUNFT "Aus Tradition in die Zukunft". So lautet das Leitmotiv des Jubiläumsjahres 1986, in dem die Ruperto Carola 600 Jahre alt wird. (Motto und erster Abschnitt eines Textes über die 600-JahrFeier der Universität Heidelberg) Mit dem lateinischen Namen Ruperto Carola wird hier auf die Universität Heidelberg Bezug genommen. Der Name findet sich zwar auf dem offiziellen Siegel der Universität, kommt aber im allgemeinen Sprachgebrauch höchstens in sehr offiziellen Verlautbarungen in der Form Ruprecht-Karls-Universität vor. Er geht auf die Gründer der Hochschule (die Kurfürsten Ruprecht und Karl) zurück. Für alte deutsche Universitäten sind solche latinisierten Namen relativ typisch, Neugründungen haben entweder den Namen einer bekannten Persönlichkeit (z.B. Justus-Liebig-Universität Gießen) oder heißen einfach Universität XY. Auch die Ruperto Carola wird in der Regel als Universität Heidelberg bezeichnet, auch etwa auf ihrer Homepage. Die expressive Funktion (bei Bühler: Ausdrucksfunktion) bezieht sich auf den Ausdruck von Emotionen oder Bewertungen des Senders, also auf die Konnotate der benutzten Zeichen. Als Unterfunktionen können Äußerungen des Bewertens (z.B. Werturteile, Ironie), der Gefühlsbekundung (z.B. in der subjektiv gefärbten Schilderung, in Erzählungen, in der Charakterisierung von Personen und Gegenständen) oder des Wollens und Wünschens angesehen werden, solange die letzteren sich nicht als Aufforderungen an andere Personen richten. Beispiel 6: Ein Buch über den Tod der Mutter Simone de Beauvoir betitelt ihr Buch, das von dem Tod ihrer Mutter und der etwas schwierigen Beziehung zu ihr handelt, mit Une mort très douce. Die Verbindung von Tod mit dem emotional geladenen Adjektiv süß drückt die Gefühle der Verfasserin aus. Die appellative Funktion schließlich ist darauf gerichtet, die Adressaten zu einer bestimmten Reaktion oder Anschlusshandlung zu veranlassen. Unterfunktionen sind die Aufforderung (in den verschiedenen Abstufungen von Empfehlen bis Fordern), die Erinnerung an zuvor Gewusstes (etwa in Form von Anspielungen oder Zitaten) und die Überredung. Letztere ist besonders interessant, da sie sich weniger direkter Appellformen (z.B. Imperative wie in Kauf dir diese DVD!) sondern vor allem indirekter Appellformen bedient, die rein äußerlich wie referentielle, expressive oder phatische Äußerungen daherkommen (vgl. Nord 2002). Beispiel 7: Indirekte Mittel der Überredung (a) Die ECHTE SALZBURGER MOZARTKUGEL von MIRABELL ist eine traditionsreiche, typisch österreichische Spezialität aus erlesenen Zutaten. Sie wird von erfahrenen Confiseriemeistern unter ständiger Qualitätskontrolle hergestellt. (b) Diese Lindt-Chocolade verdankt ihren unvergleichlichen Geschmack der Verwendung von Rohstoffen höchster Qualität ebenso wie einer seit Generationen gepflegten schweizeri- TRANSLATION AND MEANING, PART 10, 2013, 17 - 25 21 schen Herstellungstradition. (c) Kölln-Hafer wird nach besonders hohen Qualitätsmaßstäben ausgewählt und verarbeitet mit all dem Wissen und der Erfahrung, über die wir seit 1820 verfügen. Kölln Müslis werden unter Verwendung ausgesuchter und hochwertiger Rohstoffe hergestellt und unterliegen einer strengen Qualitätskontrolle. (d) JUSTINUS-Bestecke sind das Produkt sorgfältiger Fertigung. So werden nur Werkstoffe der besten Besteckstahlqualität verarbeitet. Hinzu kommt die fachmännische Erfahrung eines über 100-jährigen Unternehmens, damit Sie ein Leben lang Freude an Ihrem Besteck haben. Die vier Texte, alle aus Garantieversprechen verschiedener Produkthersteller entnommen, wiesen eine Gemeinsamkeit auf: Die Appellfunktion wird einerseits durch Referenzen auf die langjährige Tradition und Erfahrung der Firma, andererseits aber durch (positiv) bewertende Adjektive, vor allem im Elativ (besonders hohe, höchste, beste etc.) realisiert. In Text (d) wird darüber hinaus auch die phatische Funktion zu diesem Zweck eingesetzt (damit Sie… Freude haben). 5. Die Funktionen im interkulturellen Dialog Um uns darüber klar zu werden, was mit diesen vier Grundfunktionen (und ihren jeweiligen Unterfunktionen) geschieht, wenn sie in die Zielkultur übersetzt werden sollten, fassen wir zunächst kurz die Bedingungen zusammen, unter denen sie realisiert werden können. Wie die Beispiele zeigen, beruht die phatische Funktion vor allem auf der Konventionalität der verwendeten sprachlichen (und nichtsprachlichen) Verhaltensweisen. Sie kann in der Zielkultur also nur "funktionieren", wenn die Zieltextformulierung als konventionell im Sinne der Zielkultur erkannt wird. Nur in den seltenen (eher theoretischen) Fällen, in denen Ausgangs- und Zielkultur dieselben phatischen Konventionen haben, kann die phatische Äußerung des Ausgangstexts also durch reine Umkodierung in die Zielsprache die gewünschte Funktion erfüllen. In allen anderen Fällen würde eine Umkodierung der Bedeutung zu einem anderen, befremdlichen Effekt führen (was unter bestimmten Bedingungen durchaus vom Übersetzungsauftrag gefordert sein könnte) oder vielleicht sogar missverstanden werden. Wenn man allerdings sichergehen möchte, dass die Äußerung auch in der Zielkultur als phatisch erkannt und verstanden wird, muss sie an die zielkulturellen phatischen Konventionen angepasst werden. Betrachten wir dazu ein paar Übersetzungen von Beispiel 4 aus den verschiedensprachigen Versionen des Münchner Spezialitäten-Prospekts. Beispiel 4: Liebe geht durch den Magen (a) Englisch: "The way to a man’s heart is through his stomach," it is said, and this proverb is perhaps particularly true in Munich… Der Text ist in den 1980-er Jahren entstanden, als "inklusiver Sprachgebrauch" noch kein Thema war. Heute würde er allerdings die soziale Beziehung zu den Adressatinnen stören und müsste vielleicht zu "The way to people's hearts is through their stomachs," it is said… umformuliert werden. Dann aber könnte er die phatische Funktion durchaus erfüllen, auch wenn in englischsprachigen Touristeninformationen die Einleitung durch Sprichwörter oder Zitate nicht so gang und gäbe ist wie in deutschsprachigen. (b) Französisch: »L’amour passe par l'estomac« affirme un proverbe allemand... ist metaphatisch, da es über die phatische Funktion des Ausgangssprichworts informiert. Damit kann diese Formulierung die phatische Funktion nicht erfüllen, da sie – anstatt auf etwas Bekanntes zu verweisen – eine neue Information über die deutsche Kultur vermittelt. Sie ist also referentiell. Darüber hinaus darf bezweifelt werden, dass diese Information wirklich eine positive Beziehung zwischen Sender und Empfängern herstellt, da diese vielleicht von der Vorstellung, die Liebe gehe bei den Deutschen durch den Magen, eher unangenehm berührt werden. 22 CHRISTIANE NORD (c) Spanisch: «El amor pasa por el estómago», es una [!] adagio que vale especialmente para Munich… enthält nicht nur einen Fehler (es muss un adagio heißen), sondern ist für spanischsprachige Empfänger inkohärent, weil der Text etwas behauptet, was für spanischsprachige Empfänger nicht stimmt: es gibt im Spanischen kein solches Sprichwort. Auch hier könnte man annehmen, dass die Vorstellung eher nicht dazu beiträgt, den spanischsprachigen LeserInnen ein positives Bild von "den Deutschen" zu vermitteln. Der gleiche Einwand gilt für die italienische und die portugiesische Fassung (Italienisch: «L’amore passa per lo stomaco». Questo detto trova a Monaco di Baviera la sua particolare conferma…; Portugiesisch: «O amor passa pelo estômago». Este provérbio encontra em Munique sua afirmação toda especial... ). In den Sprachkulturen, die kein entsprechendes Sprichwort haben, wäre durchaus eine Substitution durch ein ähnliches möglich, im Spanischen etwa einen bekannten Ausspruch Sancho Panzas aus Cervantes' Don Quijote: "Panza vacía, corazón sin alegría." Este adagio vale especialmente para Múnich... Und wenn auch diese Möglichkeit entfällt, wäre immer noch eine neue Formulierung denkbar, die auf die Situation der Empfänger Bezug nimmt und auf das Thema Essen und Trinken hinführt, z.B. Sightseeing macht hungrig und durstig! Das ist in München kein Problem... Es wird deutlich, dass die Übersetzung der Funktion anstatt von Bedeutung und Sinn auch helfen kann, scheinbar "unübersetzbare" Formulierungen funktionsgerecht zu übersetzen. Die referentielle Funktion beruht darauf, dass das allgemeine und kulturelle Vorwissen der Adressaten korrekt eingeschätzt wird. In der Textlinguistik erkennt man das durch die Verteilung von Thema (der als bekannt vorausgesetzten Information) und Rhema (der im Text neu vermittelten Information). Wenn in einem Text auf Realien der Ausgangskultur Bezug genommen wird, kann der Ausgangstextverfasser annehmen, dass diese für die ausgangskulturellen Empfänger bekannt sind. Das muss aber keineswegs auch für zielkulturelle Empfänger der Fall sein, wie die Übersetzungen von Beispiel 5 zeigen. Beispiel 5: Ruperto Carola Schon im Ausgangstext dürfte für manche (besonders jüngere!) Empfänger die Bezeichnung Ruperto Carola für die Universität Heidelberg ein Stolperstein sein, aber wenigstens bezieht sich der Kontext auf die ihnen vertraute Kultur. Das gilt für die Empfänger der Übersetzungen nicht in gleichem Maße. Daher hat der englische Übersetzer hier adressatengerecht Ruperto Carola mit "Heidelberg University" wiedergegeben. Damit ist der Text für englischsprachige Empfänger (die ja im Prinzip aus allen Teilen der Welt kommen und mit dem lateinischen Ausdruck eher wenig anfangen können) kohärent. Es geht ihnen auch keine Information verloren, denn das Konnotat ("Lateinische Namen weisen auf eine sehr alte deutsche Universität hin!") ist bereits durch die Tatsache ausgedrückt, dass hier eine 600-Jahr-Feier annonciert wird. Die französische und spanische Übersetzung behelfen sich mit dem Zusatz université bzw. universidad zu Ruperto Carola, lösen damit jedoch das Problem der Inkohärenz nicht ganz, denn der fremdartige Name für die bekannte Universität dürfte auch hier die Empfänger eher verwirren, zumal sie mit dem Denotat ("Gründer der Universität") nicht vertraut sind. Natürlich ließen sich weitere Zusatzinformationen einfügen, die dann die Motivierung des Namens aufklären. Das ist im vorliegenden Fall jedoch aus Platzgründen nicht möglich, da die Zieltexte in genau dasselbe Layout passen müssen wie der Ausgangstext. Abgesehen von diesen Überlegungen zur Übersetzung wäre es sicher eine gute Idee gewesen, bereits im Ausgangstexts den irritierenden lateinischen Namen zu vermeiden. Das hätten die ÜbersetzerInnen dann ja der Pressestelle der Universität, die den Text verfasst hat, vor der Drucklegung vorschlagen können. Manche Textdefekte bemerkt man eben nur als Übersetzer/in. Die expressive Funktion kommt in zwei Varianten vor. Sie kann explizit sein, wie in Beispiel 6 (doux ist ein Adjektiv, das ein positives Gefühl ausdrückt) oder auch implizit (wie in Beispiel 7, wo etwa TRANSLATION AND MEANING, PART 10, 2013, 17 - 25 23 die langjährige Tradition als Positivum angeführt wird). Explizite Expressivität stellt im Allgemeinen keine Übersetzungsprobleme, da sie von den Empfängern (bei entsprechender Sprachbeherrschung) erkannt und gedeutet werden kann, auch wenn diese vielleicht aus kulturellen Gründen andere Bewertungen vorziehen würden. Aber es ist ja das gute Recht, jedes Senders oder Textproduzenten, seine eigenen Gefühle und Bewertungen zu äußern. Eine implizite Bewertung dagegen braucht zum "Funktionieren" ein gemeinsames Wertesystem. In einer Kultur, in der Modernität über Tradition gestellt wird, könnte der Verweis auf die jahrhundertelange Erfahrung möglicherweise nicht als indirektes Appell-Instrument dienen. Darauf kommen wir in der Diskussion von Beispiel 7' zurück. Beispiel 6: Der süße Tod Der Beauvoir-Titel ist explizit expressiv, er drückt (im französischen Original) eine starke, persönliche, positive Emotion aus. Die englische Übersetzung des Titels (A Very Easy Death) ist ebenfalls explizit expressiv, drückt aber eher eine Bewertung oder zumindest eine eher schwache Emotion aus. Einen "leichten Tod" wünscht man sich, auch ein Arzt, der ja nicht persönlich involviert ist, könnte diese Auskunft geben. Der deutsche Titel ist besonders interessant: Ein sanfter Tod. Hier könnte man sowohl die bewertende als auch die emotionale Komponente erkennen. Durch die Klangwirkung mit drei dunklen Vokalen ([ai] – [a] – [o:]) und den dadurch entstehenden melancholischen Rhythmus wird allerdings die emotionale Komponente verstärkt. Dazu trägt wesentlich bei, dass auf eine Übersetzung des verstärkenden très verzichtet wurde: Die Formulierung Ein sehr sanfter Tod wäre durch die beiden Zischlaute –s- und den hellen Vokal in sehr wesentlich weniger emotional konnotiert. Das gleiche gilt in vielleicht noch stärkerem Maße für eine wörtliche Übersetzung mit Ein sehr süßer Tod, die geradezu einen vergnügten Eindruck machen würde. Für das Spanische gibt es zwei Übersetzungen. Die eine, Una muerte muy lenta, ist referentiell, da sie auf die Dauer des Sterbens referiert. Allerdings könnte man hier eine implizite, stark negative Bewertung oder Emotion assoziieren, denn ein "langsames Sterben" würden sich die meisten sicher nicht gerade wünschen. Die zweite Variante, Una muerte muy dulce, entspricht dagegen in Ton und Emotion genau dem Originaltitel. Die appellative Funktion muss die Adressaten dazu befähigen und bewegen, in der gewünschten Wiese zu reagieren. Dazu muss die Formulierung ihre Sensibilität und Ansprechbarkeit, ihr allgemeines und Kulturwissen und ihre Bereitschaft zur Kooperation aktivieren. Wenn die Empfänger nicht "mitspielen" können oder wollen, kann die appellative Funktion ihr Ziel nicht erreichen. Betrachten wir dazu einen dänischen Ausgangstext, der mit den oben unter Beispiel 7 genannten Textsegmenten vergleichbar ist. Es handelt sich ebenfalls um ein Garantie-Versprechen, das einer Pralinenpackung beiliegt. Beispiel 7: Werbung auf Dänisch Der erste Satz des Ausgangstexts lautet: Anthon Berg har altid bestræbt sig for at levere varer af fineste kvalitet, wörtlich übersetzt etwa: Anthon Berg hat sich immer bemüht, Produkte bester Qualität zu liefern. Diese Formulierung wäre im Deutschen aus verschiedenen Gründen nicht funktionsgerecht. "Anthon Berg" ist als Eigenname erkennbar, allerdings ist für deutsche Empfänger nicht ersichtlich, um wen oder was es sich dabei handelt. Der Ausdruck "hat sich immer bemüht…" erinnert an den versteckten Code eines Arbeitszeugnisses, in dem er darauf hinweist, dass die betreffende Person in dem Bemühen um eine bestimmte Verhaltensweise (Pünktlichkeit, Kooperationsfähigkeit, effektives Arbeiten etc.) gerade nicht erfolgreich war. Die Übersetzung von Bedeutung und Sinn des Ausgangstexts würde also geradezu kontraproduktiv wirken. Aus diesem Grunde ist die Übersetzung ins Deutsche (und die anderen Übersetzungen ins Englische, Französische, Spanische in ähnlicher Weise) folgendermaßen formuliert: Anthon Berg, Kopenhagen, seit 1884 berühmte Schokoladenfabrik, hat ihren Ruf auf Verwendung feinster Rohstoffe zusammen mit strengster Qualitätskontrolle und sorgfältiger Verpackung aufgebaut. 24 CHRISTIANE NORD Damit ist zunächst einmal die referentielle Funktion gesichert (im englischen und spanischen Text wird der Verweis auf Kopenhagen noch um die Angabe Dänemark erweitert); in der Form orientiert sich die Übersetzung an der konventionellen Form der Qualitätsgarantie, die wir bereits oben in Beispiel 7 gesehen haben, bei der Hinweise auf Alter und Berühmtheit der Firma sowie eine Reihe von Reizwörtern und Elativen eine wichtige Rolle spielen. Damit ist sichergestellt, dass die appellative Intention des Herstellers/Senders auch in den Zielkulturen ihren Zweck erfüllt. 6. Schlussfolgerungen Anstatt zu versuchen, beim Übersetzen Bedeutung oder Sinn trotz unterschiedlicher Bedingungen auf der Ausgangs- und Zielseite zu "bewahren", wie es die äquivalenzorientierte Übersetzungstheorie fordert (vgl. etwa die Definition von Nida/Taber 1969: 12), schlage ich aus skopostheoretischer Sicht vor, die (intendierte) kommunikative Funktion als Tertium comparationis für die Translation zu verwenden. "Funktion" ist eine verkürzende Redeweise für "Funktion / Funktionen / Funktionshierarchie". Auch sollten wir nicht von "Textfunktionen" reden, da Texte in aller Regel jeweils für verschiedene Funktionen bestimmt sind, die möglicherweise hierarchisch auf verschiedenen Rängen angesiedelt sind, sondern von "Funktionen in Texten". Die für bestimmte Textteile oder Segmente intendierten Funktionen werden durch (sprachliche und nichtsprachliche) "Funktionsmarker" angezeigt, die den Adressaten signalisieren, für welche Funktionen der Text gedacht ist. Das heißt aber nicht, dass er unbedingt auch für diese Funktionen verwendet werden muss. Nur wird er für andere Funktionen vielleicht nicht optimal geeignet sein. Ein Text ist ein "Informationsangebot", aus dem sich Rezipienten je nach ihren Bedürfnissen und Verstehensvoraussetzungen "bedienen". Texte haben keine Funktion, sondern sind für Funktionen intendiert. In der Rezeptionssituation werden ihnen von einem individuellen Rezipienten je und je bestimmte Funktionen zugeschrieben. Das bedeutet, dass man auch "Funktionen" beim Übersetzen nicht ohne weiteres "bewahren" kann, weil die Bedingungen zu ihrer Zuschreibung in Ausgangs- und Zielkultur unterschiedlich sind. Für die Erzielung derselben oder einer ähnlichen Funktion (= instrumentelle Übersetzung) sind oft mehr Transformationen nötig als für die Bewahrung der Bedeutung bei veränderter Funktion (z.B. Meta-Funktion). Daher muss für eine Translation auf der Grundlage einer Interpretation des Übersetzungsziels (Skopos) festgelegt werden, welche Funktionen das Translat in der Zielkultur für einen bestimmten Adressatenkreis haben kann bzw. soll. Diese können durchaus von den Funktionen abweichen, die der Ausgangstext in der Ausgangskultur für einen bestimmten Adressatenkreis haben kann oder tatsächlich hat/gehabt hat. Literaturhinweise Bühler, K. (1934), Sprachtheorie. Jena: Fischer. Bussmann, H. (1983), Lexikon der Sprachwissenschaft. Stuttgart: Kröner. Duden, Der Große (1963), Etymologie. Herkunftswörterbuch der deutschen Sprache. Mannheim: Dudenverlag. Grimm, J. (1847), "Ueber das pedantische in der deutschen sprache." In: Störig, H.-J. (Hrsg.) (1963): Das Problem des Übersetzens. Darmstadt: Wiss. Buchgesellschaft. Jakobson, R. (1971), "Linguistik und Poetik." In: Ihwe, Jens (Hrsg.): Literaturwissenschaft und Linguistik I, Frankfurt/M., 142-178. Nida, E.A. & Taber, C.R. (1969), The theory and practice of translation. Leiden: Brill. Nord, C. ([1988]2009), Textanalyse und Übersetzen. Theorie, Methode und didaktische Anwendung einer übersetzungsrelevanten Textanalyse. Tübingen: Stauffenburg (Sammlung Groos). Nord, C. (1997), Translating as a Purposeful Activity. Functionalist Approaches Explained. Manchester: St. Jerome. Nord, C. (2002), "Über-Reden durch An-Reden. Die phatische Funktion als Mittel zum Appell in englischen, spanischen und deutschen Werbetexten." Linguistica Antverpiensia NF 1/2002, 145-168. TRANSLATION AND MEANING, PART 10, 2013, 17 - 25 25 Reiß, K. & Vermeer, H.J. (1984), Grundlegung einer allgemeinen Translationstheorie. Tübingen: Niemeyer. Vermeer, H.J. (1978), Ein Rahmen für eine allgemeine Translationstheorie. Lebende Sprachen 23 (1978), 99102. 26 TRANSLATION AND MEANING, PART 10, 2013, 27 - 37 27 PARAMETER VARIABILITY IN TRANSLATIONAL APPROXIMATION Barbara Lewandowska-Tomaszczyk University of Łódź, Poland Abstract: The paper is a sequel to Lewandowska-Tomaszczyk (in print 1, in print 2), in which the concept of approximation in same-language communication and in translation was proposed and its typology presented. The discussion takes advantage of the concept of translational reconceptualization (Lewandowska-Tomaszczyk, 2010) as an outcome of some approximating processes. The present discussion locates the phenomenon of communicative approximation in translational contexts, presents a set of parameters (Lewandowska-Tomaszczyk ,2012) that are likely to undergo modification between SL and TL texts and shows the circumstances and range of these processes. 1. Approximation and re-conceptualization Semantic approximation (Lewandowska-Tomaszczyk, 2013; in press) is a cognitive and linguistic phenomenon, pervasive in interactional encounters, present both in monolingual communication and in translation. By semantic approximation I do not exclusively mean ambiguity, as ambiguity implies the lack of precise linguistic potential due to the absence, temporary or decontextualized, of fully determined meanings. My claim is rather that interactants, first, out of language typological, social, etc., pressure and conditioning, or secondly, because they choose to, use lexical, phrasal or syntactic constructions, which do not convey fully determined meanings. The results of approximation understood in this way are invariably the addressee’s re-conceptualization of meanings constructed and conveyed (Lewandowska-Tomaszczyk, 2010). The re-conceptualization is most clearly observed in the case of translation. 2. Mental imagery There exists a complex relation between the world and language via processes of mental imagery construction, which combines both linguistic and encyclopaedic knowledge, speakers’ and addressees’ intentions, their expectations and preferences as well as contextual parameters. Translation, as was argued before, involves a cycle of re-conceptualizations of a Source Language (SL) message, expressed eventually in the Target Language (TL). Re-conceptualization is a matter of degree. In translation, the message as constructed by an author is re-conceptualized by SL recipients, the translator and eventually, the TL audience. Language typological pressure, expressed by TL form, with its morphology, syntax, and context of use is one of the necessary factors in re-conceptualization cycles. Other parameters involve translation strategies, by which I understand the translator’s accessibility to a TL system of synonymic or paraphrase meanings, which portray, either in literal or inferential terms, the basic conceptual-semantic content, i.e. the content of a scene in which basic spatiotemporal properties, participants and their actions are identical to those invoked in SL. What is different in TL though, under both of the scenarios above, is the construal of the scenes (Langacker, 1987; 1991), governed by a number of imagery categories. As the linguistic message consists of both conceptual-semantic content and the way this content is construed, the important properties of construal such as granularity (i.e. a selection of the degree of schematicity), focusing (amplification of certain aspects of the scene to the disadvantage of some others), prominence (particular property salience), and perspective (point of view), all contribute to overall TL meaning. What can also be different is the construal of the scene participants who, by acquiring TL specific grammatical properties on the one hand, and additional ones, as selected by the translator, cause reconceptualization of the participants populating the scene. The linguistic type a given system represents is responsible to a large extent for the semantics conveyed by these forms via what is labeled profiling in cognitive linguistics (Langacker, 1987; 1991), 28 BARBARA LEWANDOWSKA-TOMASZCZYK which designates the selection of particular, linguistically conventionalized, properties of things, relations and events, which can be expressed in one language but not necessarily in another. 3. Communicative equivalence As was proposed in Lewandowska-Tomaszczyk (2013; in press) both translation and monolingual types of communication can be either symmetric or asymmetric. Fully symmetric, i.e., closely aligned, or parallel, communication is not frequent, similarly to non-ubiquity of fully equivalent translation. Asymmetric types of interaction with two opposing poles, one - coarse-grained approximative correspondence of generalized alignment, on the one hand, and asymmetric fine-grained alignment (correspondence), i.e., a particularization relation, on the other, are much more frequent and form larger bulks of authentic texts. Fig. (1): Translation typology on the basis of the Approximative Interactive Perspective (Lewandowska-Tomaszczyk ,2013:46) Discourse interactants use a variety of partially synonymous or paraphrasing forms to communicate similar meanings. In a number of cases, they use forms which, outside the context, would be ambiguous, thus not uniquely interpretable, and receive only a more constrained interpretation in the context of use. When, on the other hand, the language users select the same forms in their communication, the forms need not be absolutely synonymous either in monolingual communication, in intercultural contexts or in translation. Rather, they can be considered to represent a set of polysemous meanings, constrained only by the same linguistic sign. An example of the latter context can be instances of SL loanwords used in TL translation. In the cases of such communicative context, both SL speakers and, even to a larger extent, translators, choose to express not identical meanings but rather the second, third, or nth best, which may be fairly distant from SL original messages. Nevertheless, under such circumstances the users produce and accept a given TL version as the best approximation of what was proposed in SL. The sources of this interpretive multiplicity, rooted primarily in SL and TL users’ inferential power, i.e., an ability to complement a message with further (unexpressed) thought(s) or image(s), are associated with a particularizing knowledge frame or an Idealized Cognitive Model-based creation of a particular scene details of which fill in the uninhabited parts of relevant mental spaces. The interpretive diversity is also connected with symbolic structural and constructional properties which express language-specific meanings. Cases of loose synonymy and equivalence, related to an asymmetry between radial polysemies attached to those meanings (Lewandowska-Tomaszczyk, 2007), asymmetric relations between words and their definitions, structural and lexical ambiguities or subtle differences between similar syntactic constructions – are all subject to multiple interpretations in parallel TL texts. One type of parallel alignment are schematic structure-preserving representations. Schematicity present in models or maps, diagrams or tables is responsible for and accounts for a resemblance between a model and a thing modelled. Some translation scholars place substitution as a separate class of translational strategies or correspondences. However, substitution is invariably an approximation. It can be symmetrically aligned, if basic meaning components of a SL or a SL user’s meanings are preserved, or asymmetric, in cases when the translation adds entirely new inferences or when it simplifies the SL message. Even the symmetric – parallel – alignment though is never an identity. The very fact that SL and TL are TRANSLATION AND MEANING, PART 10, 2013, 27 - 37 29 different linguistic systems, accompanied by the TL syntactic pressure exerted on the semantic layer of the SL message, results in approximations of different types. One of the types of particularisation is synonymous with precisation and is observed in self- or other-party repairs as when a speaker says this shows different sides of the matter or rather shows it in a different light. This process involves both conceptual depth particularization as the speaker reflects on the meaning and goes deeper into the original context as well as width particularization as the meaning is extended to create new conceptual links in metaphoric mapping. In the example above what is used is an elaborated type of particularized alignment in terms of a metaphoric mapping of the Target Domain material ‘illumination of matter from different angles and perspectives’ onto, an already metaphorical, Source Domain, representing a certain problem in terms of its reification (only physical things have sides). The relevance of such correspondence types to cross-linguistic comparisons and translation in particular is observed when a translator replaces for the headword one of the adjacent units, displaying a closer or more distant semantic similarity (e.g. for the concept of democracy, one of its constituent notions free will is used). What is worth noting is the fact that such substitutions are also very frequent in monolingual communication. It is certainly not suggested here that the headword and the same cluster components are absolute synonyms, but rather that they tend to function as such in authentic language contexts. The phenomenon which has to be emphasized here is the fact that interactants use notoriously approximative senses of different types in their exchanges, however the meanings employed are considered sufficient for the context in which they are used (see Gärdenfors, 2000 for details). In such contexts language users tend to be satisfied with a fairly coarse-grained level of semantic specificity in their exchanges. The processes like those also take place in translation. Translators move freely around the SL/TL conceptual identity axis but they are nevertheless constrained by its value at the same time. Language-typology requirements dictate a syntactico-semantic way to express what is going on in the outside reality. Coercive pressures discriminating between, say, the English This is my car, This car is mine versus This car belongs to me, I am the owner of this car, etc. are precisely such examples, not fully synonymous (each with a set of different presuppositions and implications), nevertheless frequently used as approximative pragmatic synonyms. Such typological and structural differences between SL and TL shape an alternative portrayal of the outside world in the original and its translated version. The closer the languages are typologically, the closer SL and TL images can be assumed. The farther away the systems are on the typological criteria the less of the original message is likely to be retained in the translation. 4. Meaning emergence in translation What can be inferred from the above analysis is that a translated text signals sets of new or modified meanings, emerging in the translation process. Meaning emergence takes place when a target meaning structure interpretation is not (entirely) predictable from the input parts from which it is built1. In other words, emergentism, to put it shortly, is seen when, as a consequence of the interaction of different elements of a system, a new (type of) meaning arises, at a higher level of self-organization. In translated texts, emergent meanings arise due to SL and TL differences in a number of language areas: 1. Prototypes and extensions 2. Collocational phrases 1 This interpretation is similar then to that of constructions in construction grammar and that of a blended entity in the meaning integration paradigm. 30 BARBARA LEWANDOWSKA-TOMASZCZYK 3. Idiomatic phrases 4. Non-compositional phrases 5. Inferential meanings 6. Implicit senses 7. Encyclopaedic meaning and cultural references 8. Language/Culture contacts: Loanwords 9. Language games Table (1): Loci of emergent structures in translated texts Emergence of TL meanings implies the inaccessibility of the totality of meaning from the input linguistic material and experiencing either a fairly unexpected ‘semantic leap’ (Coulson, 2001), expressed in mental frame-shifting, knowledge frame schematization or else its particularization and enrichment. 5. Language typological bootstrapping in translation The present section illustrates a translation analysis exercise, based on a short article by an International Herald Tribune journalist, Ruiyan Xu (2010), who reflected on the relationship between the Chinese original semantics of the phrase baidu as used in the name of a web search company there (BAIDU.com). The Chinese phrase baidu in its original character form is rich with linguistic and historical connotations. The Chinese characters bi and dù mean ‘hundreds of ways’. The phrase is borrowed from a 10th century Chinese poem about a man searching for a woman on a busy festivity day. Metaphorically, it is interpreted as a search for clarity amid chaos. However, transliterated into the Latin alphabet, it loses all its implicit meaning and poetic allusions. Another important semantic layer entirely lost relates to the divisibility present in Chinese and absent in English and other languages, between the phonetic and semantic interpretations of the character. The sound bai means one hundred but it can also designate a cypress, or the verb to arrange. Baidu, without diacritics, can mean ‘a failed attempt to poison’ or ‘making a religion of gambling’. All these nuances, cultural references and polysemic/homonymic potential disappear in the Roman alphabet.  The English speaker then, not proficient in Chinese, acquires very little of all the original levels of the sense from the translation. He or she may try to identify componential meanings of the name, such as by, bye, do, etc., obviously distinct from the original. Not even approximating nuances of the original are accessible to foreign users from the translation alone. Some additional insight can perhaps be achieved from the logo, an iconic representation of a wild animal’s (not uniquely panda’s) paw trace, which can activate associations with an animal long wandering for miles in desolate forests. Such an image is only loosely connected with the rich original message intended by the SL authors. 6. English-Polish translational approximations exemplified In the section to follow an analysis of selected English-to-Polish and Polish-to-English samples from various sources will be presented. The samples show a number of the parameter variants and will be exemplified on the materials drawn from the Lodz University PELCRA parallel resources. Most of the samples, which is a rule rather than an exception, include not a single parameter modified in TL TRANSLATION AND MEANING, PART 10, 2013, 27 - 37 31 but rather a number of complex typological and translational options available in either of the two languages. 6.1 Sub-types of translational approximations A vast majority of the reconceptualization outcomes, which result from the SL to TL approximative communication will be due to the TL grammatical pressure and coercion, some others result from the translator’s conscious selection strategies. The English to Polish examples to follow (1) - (4) represent basically TL syntax-induced approximations. The other translation samples (B - F) do not obviously lack the syntactic coercion parameter either. The outcomes in both cases result in the change in the scene construal2: A. TL SYNTACTIC COERCION APPROXIMATIONS (1) SYNTAX parallel alignment Eng. indefinite reference translated into Pol. Noun final sentential position Eng. Passive > Active Eng. Theme Verb receive > Pol. Agent Verb come One day a letter was received Pewnego dnia nadszedł list lit. ‘One day walked (the perfectivising prefix nad ‘over’ points to the centrality of the Agent in the scene) letter’ (2) SYNTAX particularisaton alignment Eng. Lexicalized verbal meaning mount > Pol. Analytical verbal meaning (Verb - satellite) przychodził na górę Eng. Verb taking > Pol. biegł przeskakując [running jumping] It was his custom to mount straight to the nursery, taking about three degrees of the staircase at once Zwykle przychodził [came] od razu na górę [upstairs] do dziecinnego pokoju, przy czym biegł szybko [ADDITION – He was also running quickly], przeskakując [jumping over] po trzy schody naraz. B. 2 (3) SYNTAX parallel alignment Eng. Phrasal Verb > Pol. Reflexive Verb Madame appealed to me Wówczas Madame zwróciła się [turned herself] do mnie (4) SYNTAX parallel and generalized alignments Eng. Passive > Pol. Reflexive Eng. Analytic structure (be + Adj) > Pol. Synthetic structure (Reflexive Ponoun + Nominalization) POLYSEMIC cluster Eng. prompt [szybkość] > Pol. żwawość [liveliness] I had been saved by being prompt. Uratowałem się [I saved myself] dzięki [thanks] swej [my] żwawości [liveliness]. LEXIS/SYNTACTIC COERCION (5) LEXIS particularization Eng. Negation lexicalized > Pol. negation syntactic implicit The order of presentation of the examples signifies the SL version (first sentence or phrase) and a corresponding TL version (the sentence/phrase following) 32 BARBARA LEWANDOWSKA-TOMASZCZYK He failed to do it Nie zrobił tego C. CONCEPTUAL STRUCTURE/MEANING METAPHOR (6) METAPHOR generalized alignment Pol. Source Domain lecieć [fly] > Eng. Source Domain roam lecąc wyobraźnią w przyszłość lit. ‘flying with imagination into the future allowing his imagination to roam into the future (7) METAPHOR generalized and parallel alignments Duma rozsadzała mu piersi [exploded his breast]– nie niska [low] duma Pride swelled his breast – not ignoble pride The metaphor in (7) represents a symmetrical bilingual alignment on the interpretation level, even though the shift in the Source Domain from the process of swelling to its final outcome of exploding is worth pointing at as the two metaphors can be perceived as portraying different phases of the same scenario, in fact a metonymic relation holding between them. The Adjective niska ‘low’, has also a metaphoric (physical) derivation and possesses a wider range of applications in Polish than the English ignoble, so the process can be considered particularized alignment. (8) METAPHOR generalized alignment Polysemic (emotion) cluster I lęk [anxiety] przed tym, co mogło i miało nastąpić, przytaił [made secret] mu się w tej chwili w sercu, a natomiast zalała je radość i duma And fear before what might and was to follow at that moment was subdued in his heart, and joy and pride overflowed it. In polysemic clusters, a concept similar enough to a given SL notion (lęk ‘anxiety’), from the same conceptual domain (fear emotions in this case) is selected as a TL equivalent (cf. LewandowskaTomaszczyk, 2012). MEANING [SYNONYMY] (9) SYNONYMY generalized alignment Polysemic cluster sam żywiąc się od tygodnia prawie wyłącznie durrą wykradaną [stolen from] wielbłądom. He sustained himself for a week almost exclusively upon durra taken from the camels (10) SYNONYMY generalized alignment Polysemic [Emotion] cluster Mimo żalu [longing, sadness, grief] za Mikołajem Inspite of my grief for Mikolai (11) SYNONYMY generalized alignment Polysemic [Axiological] cluster He was not only useless as an officer and a bad influence amongst the men Nie dość, że był złym oficerem [bad/evil officer] i dawał gorszący przykład [lit. gave indignation/shame-inducing example] swym ludziom TRANSLATION AND MEANING, PART 10, 2013, 27 - 37 33 The translator chooses to use a polysemic-cluster translation strategy in (11) above, which implicationally retains a schematic frame of the same conceptual content. D. TERMINOLOGY (12) TERMINOLOGY parallel and particularization alignments (a) Te pułki kawalerii angielskiej, które za czasów Napoleona potykały się z jazdą polską, chlubią się tym istotnie do dziś dnia i każdy oficer mówiąc o swym pułku nie omieszka nigdy powiedzieć: 'My biliśmy się z Polakami.' Those regiments of English cavalry which during the times of Napoleon met the Polish cavalry actually pride themselves with that fact at the present time, and every officer speaking of his regiment never fails to say, 'We fought with the Poles. Both English cavalry and Polish kawaleria have a wider range of use at present (denoting the part of an army that can use both horses as in older times and fast armoured vehicles for fighting3) than the Polish jazda (fighting or parading on horses), which is employed with reference to historical periods of time. Cavalry > kawaleria then is an example of a fully parallel approximative alignment, while jazda > cavalry represents a generalized approximative type. (12) TERMINOLOGY parallel and particularization alignments (a) Do you know what the difference is between the computing card and timesharing? Wiecie jak jest różnica między kartami do programowania [lit. ‘cards for computing] i pracą na komputerze z podziałem czasu?’ [lit. work on computer with time division] The Polish equivalents of the English computer terms in this TL version tend to show a less lexicalized status than the original English terms. The former are thus more semantically transparent and less idiomatic. E. STYLE (13) STYLE/PRAGMATICS generalized alignment Pol. formal style > Eng. archaisms Pragmatic effects aligned – formality dlatego zapanowałem nad sobą i odpowiedziałem sucho: — Mylisz się [You are mistaken], Haniu, jeśli sądzisz, że ty jesteś powodem. So [I] mastered myself and answered dryly: ‘Thou art mistaken, Hania, in thinking that thou art the cause ‘ F. 3 PRAGMATICS (14) PRAGMATICS generalized alignment Pol. English loan > Eng. the same English word retained Po czym dodał z dumą: - Sam pan wielki powiedział, że Kali jest donkey Compare the data from the BNC and NKJP with reference to an extended meaning of the modern cavalry: Leslie Hore-Belisha, who as Minister obliged the cavalry to take tanks in 1937 A modern, European-style cavalry unit of more than two thousand muskets, adapted to Japanese warfare. (BNC) Pierwszy Regiment Kawalerii, czyli lekkie czołgi, wozy bojowe i zwiadowcze (NKJP) 'First Cavalry Regiment, i.e. light tanks, combat and reconnaisance vehicles'. 34 BARBARA LEWANDOWSKA-TOMASZCZYK After which he added with pride: 'The great master himself said that Kali is a donkey.' Even though the English version achieves a schematic form of generalized alignment when contrasted with the Polish SL version, the humourous element similar to the Polish original will not necessarily be identified by the English reader. In both languages the proverbial comparison głupi jak osioł ‘as dumb as a donkey’ is used. In the Polish version, the reader proficient in English will recognize the joke but a monolingual reader will be left with a worry and a misinterpreted message (in a similar frame of mind as Kali was in fact). The English version represents an approximation by particularization (‘donkey’ is disambiguated and particularized). However, what can be missing is the humourous effect for the TL reader. 6.2 Quantitative subtypes There are basically three technical types of TL approximations, mostly visible in interpreting practice. Two of these, additions and elaborations, are clear cases of particulate approximations. The third – intrusions can either be part of explanatory strategies4 or does not present material relevant to translation strategies. Elaboration [MA student data, University of Lodz, COBEL programme] (15) SL: Professor, the floor is yours. I: Profesor D. będzie przewodniczył tej sesji. [Profesor D. will chair the session]. Profesor D. Panie profesorze udzielam panu głosu. Bardzo proszę. [Here you are] Intrusion (16) SL: Jak Państwo wiedzą mamy trzy prezentacje <nie widziałaś tego mojego pisaka nie? gdzieś go posiałem> yyy (..) jestem bardzo zadowolony z tego, że mogę przewodniczyć tej sesji. ‘As you can see we have three presentations, <you haven't seen my marker, have you? I've left it somewhere . yyy (...)> I am very pleased I can chair this session.’ An interesting case in both translation and interpreting, even though more frequent in interpreting, are TL language pointers, which include phrases such as so-called and its variants. They can either show translators’ distancing themselves from the materials presented (translator position pointers) but can also point to the content not expected to be familiar to TL addresses (novelty pointers), or else they can also function as pause fillers. 7. Parameters of translational approximation Approximative cases can be seen as instances varying along a number of parameters located in each language level, as could be seen in the typologies presented above. Each parameter is assigned a certain tolerance space and a tolerance threshold, whose crossing may result in miscommunication, even though the upper bound of the tolerance spaces will never be easy to uniquely identify (Gärdenfors, 2000; Lewandowska-Tomaszczyk, in press). The parameters can both refer to Qualitative as well as to Quantitative measures (Lewandowska-Tomaszczyk, 2012). All the examples analyzed above represent qualitative parameters. Example (17) on the other hand shows the numerical relations of SL 4 Prototypical scenarios of intrusive explanatory strategies are instances of community interpreting, in some cases also court interpreting, which extend an original SL message to elaborate on it and explain details cf. Interpreted Communicative Event (Carmen Valero Garcéz, Ann Martin Crossing Borders in Community Interpreting: definitions and dilemmas Benjamins 2008). TRANSLATION AND MEANING, PART 10, 2013, 27 - 37 35 and TL versions, presented in terms of the number of sentences and their ordering (in paranthesis) as well as the number of additions and deletions in the TL version. (17) [syntax: 1 sentence]in the form of reflexive snapshots staged in different ways [1]] and which drilled deep into [5] the strata of society [2], the currently alltoo-obvious[4] proliferations [4a] of an apparently [deleted 7] archaic postCommunist capitalism 6] were felicitously brought into focus.[3] [syntax: 2 sentences] ‘owe refleksyjne migawki. Mimo [8 added] różnorodności form, które przybierały,[1] wwiercając się [present participle for a relative clause] głęboko w [5] tkankę społeczną [2], bardzo trafnie naświetliły [3] i ukazały [3] zupełnie oczywiste z dzisiejszego punktu widzenia [4] manifestacje [4a] [7 deleted] archaicznego kapitalizmu postkomunistycznego’ 6] [Stadion /8] Some of the substitutions such as [4a] in (17) might be considered direct SL-to-TL interference cases even though the form manifestations was not in fact used in the SL version and the Polish form manifestacje is hardly ever used in the sense equivalent to the English original. The quantitative parameters cover frequencies of particular lexical items, e.g., the form capitalism is used 446 times in English and the equivalent kapitalizm* - 826 times in a comparable Polish sampler (ca 20 mln words each), which may point to varied lexical (conceptual) preferences of the English and Polish authors in similar periods of time (80ies-90ies of the 20th century). 8. Conclusions The resemblance between a SL message sent and a TL message received presents a variety of forms and shapes. In most of the cases complex types of approximation take part in the processes of SL text reconceptualization. The translator is a mediator and has to accommodate to two parties: a S speaker/ author and a TL addressee. The outcome is a blended SL and TL meaning in a TL form, in which the input components are not easily visible or identifiable without further analysis (Fauconnier and Turner, 1996). What is conventionally observed in translation are weakly commensurable categorical hierarchies in SL and TL and, rooted in them, a dynamic displacement of senses (cf. Lewandowska-Tomaszczyk, 1987)5. Translation, as was shown above, enhances numerous approximation operations and varied displacements. The concept of language typology or SL grammar coercion has been used with reference to linguistic typological bootstrapping, independent of the personal preferences of the translator, and resides in the phonological, morphological, syntactic, semantic and pragmatic layers of the language. Other shades of meaning, on the other hand, are settled in pragmatic, discourse and contextual conventions, i.e., modes of conventional human behaviour, while still others, which do deserve the label ‘strategies’, are an outcome of a conscious choice and the translator’s decision. The multidimensional typology of approximation types presented in this paper covers a variety of cases representing different degrees of similarity between a SL and the SL message. 5 As shown in Lewandowska-Tomaszczyk (1987), a cross-linguistic displacement of senses is a typical phenomenon in comparing languages. For instance, the verb go in English, a superordinate category for numerous verbs of movement, has no direct equivalent in Polish at the same categorial level. In Polish the concept of going can be considered to have equivalents implemented by two more specific verbs iść ‘walk’ and jechać ‘move by/in a vehicle’, which, in turn, would correspond to more specific English verbs such as ride, drive, etc. 36 BARBARA LEWANDOWSKA-TOMASZCZYK A similarity between the source and the target texts lies in the majority of cases in the basic conceptual content, which is nevertheless modified in the way it is portrayed, and enriched with the subjective interpretation, which is left for each language user to construct or uncover.6 The processes of approximation are clearly cognitive in nature and they primarily describe processing aspects of linguistic behaviour involving SL and TL. At the same time, they involve the final product of translation, i.e. they have a direct impact on the representational properties of linguistic performance. That means that in order to cater for both sets of aspects, what is needed is a battery of new or modified research methods, combining cognitive corpus linguistics (Lewandowska-Tomaszczyk & Dziwirek, 2009), bilingualism studies (Pavlenko, 2008), and psychological experimentation for the developing cognitive theory of translation (Lewandowska-Tomaszczyk, 1999; Halverson, 2010) to gain solid ground in science. References Coulson, S. (2001), Conceptual Blending. Frame-Shifting and Conceptual Blending in Meaning Construction. Cambridge: Cambridge University Press. Fauconnier, G., and M. Turner (1996), “Blending as a central process of grammar”. In: Adele E. Goldberg (ed.), Conceptual Structure, Discourse, and Language. CA: CSLI Publications, 113-131. Garcéz, C. V. & A. Martin (2008), Crossing Borders in Community Interpreting: definitions and dilemmas. Amsterdam: Benjamins. Gärdenfors, P. (2000), Conceptual Spaces: The Geometry of Thought. Cambridge, MA: MIT Press. Gutt, Ernst - A. (1991), Translation and Relevance: Cognition and Context. Oxford: Basil Blackwell. Halverson, S. L. (2010), “Cognitive translation studies: Developments in theory and method”. In: Gregory M. Shreve & Erik Angelone (eds.), Translation and Cognition. Amsterdam/Philadelphia: Benjamins, 349-369. Langacker, R. W. (1987/1991), Foundations of Cognitive Grammar, vols.1 and 2. Stanford: Stanford University Press. Lewandowska-Tomaszczyk, B. (1987), Conceptual Structure, Linguistic Meaning, and Verbal Interaction. Łódź: Łódź University Press. Lewandowska-Tomaszczyk, B. (1999), “A Cognitive-interactional model of cross-linguistic analysis: New perspectives on ‘tertium comparationis’ and the concept of equivalence”. In: B. LewandowskaTomaszczyk (ed.), Cognitive Perspectives on Language. Frankfurt am Main: Peter Lang, 53-76. Lewandowska-Tomaszczyk, B. (2007), ”Polysemy, prototypes, and radial categories”. Oxford Handbook of Cognitive Linguistics. Oxford, Oxford University Press, 139-169. Lewandowska-Tomaszczyk, B. (2010), “Re-conceptualization and Emergence of Discourse Meaning as a Theory of Translation”. In: Barbara Lewandowska-Tomaszczyk & Marcel Thelen (eds), Meaning in Translation. Frankfurt a. Main: Peter Lang, 105-147. Lewandowska-Tomaszczyk, B. (2012),"Explicit and tacit - An interplay of the qualitative and quantitative approaches to translation". In: Michael Oakes and Meng Ji (eds), Quantitative Methods in Corpus-Based Translation Studies. Amsterdam: Benjamins, 3-33. Lewandowska-Tomaszczyk, B. (2013), “Approximative Communication in Translation”. In: Marcel Thelen & Barbara Lewandowska-Tomaszczyk (eds.), Translation and Meaning, Part 9. Proceedings of the Maastricht Session of the 5th International Maastricht-Łódź Duo Colloquium on “Translation and Meaning”. Maastricht: Maastricht School of Translation and Interreting, Zuyd Universty of Applied Sciences, 45-51. Lewandowska-Tomaszczyk, B. (in press), “Approximative spaces and the tolerance threshold in communication”. International Journal of Cognitive Linguistics. Lewandowska-Tomaszczyk, B. & K. Dziwirek (eds.) (2009), Studies in Cognitive Corpus Linguistics. Frankfurt a. Main: Peter Lang. Pavlenko, A. (2008), “Emotion and Emotion-laden Words in the Bilingual Lexicon”. In: Language and Cognition 11 (2): 147–164. Cambridge University Press. Available at: http://journals.cambridge.org. Xu, R., (2010),“Search Engine of the Somg Dynasty”, International Herald Tribune, Monday May 17, 2010. 6 Consult Gutt (1991) on “interpretive resemblance, the sharing of thoughts between the intended interpretations, which in fact forms a continuum, between full and zero resemblance”. TRANSLATION AND MEANING, PART 10, 2013, 27 - 37 37 Corpora BNC British National Corpus. A Corpus of English Conversation [CEC] (1980). Lund: C.W.K. Gleerup, eds. Jan Svartvik and Randolph Quirk. NKJP [www.nkjp.pl] National Corpus of Polish. Polish-English and English-Polish PELCRA parallel corpora (University of Lodz). 38 TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 39 PAINTINGS RE-SEEN. POEMS RE-READ. A FEW REMARKS ON INTERSEMIOTIC TRANSLATION Agata Holobut Jagiellonian University, Krakow, Poland Abstract: An interesting piece of witticism circles among critic-shy musicians and performance artists: “Writing about music is like dancing about architecture” (Naylor and Ball, 2005:57). Ironically, scholars familiar with Roman Jakobson’s concept of intersemiotic translation might be tempted to take this biting comment at face value. Indeed, both activities seem to illustrate the phenomenon, striving to express orthographically/choreographically what has originally been conveyed by auditory/spatial means. In my paper I revisit the Jakobsonian notion of intersemiotic translation and examine its relevance for translation studies and for other related disciplines, such as art and literary criticism. First, I address the questions of whether and under what circumstances it is justifiable to treat such hybrid art forms as dance inspired by architecture or poetry inspired by painting in terms of translation. Following the American scholar Claus Clüver, I argue that the concept of intersemiotic translation can be treated as an interpretive device, which reveals implicit structural correspondences between different works of art and sheds new light on intertextual relations between them. To describe these correspondences more precisely, I refer to the notion of linguistic imagery known from Ronald W. Langacker’s cognitive grammar. I subsequently analyse an ekhprastic poem (i.e., a text inspired by a visual work of art) in terms of translation from visual into the verbal in order to demonstrate that such an interpretive approach allows for a better understanding of both the painting and the poem, and that Cognitive Linguistics indeed offers useful tools in their contrastive analysis. 1. The definition of intersemiotic translation The concept of intersemiotic translation was introduced by Roman Jakobson in his seminal paper “On Linguistic Aspects of Translation” (1959). Referring to C.S. Peirce’s notion of semiosis, Jakobson defined translation as interpretation of signs by means of different signs, accounting not only for operations among natural languages, but also within or among different sign systems (Eco and Nergaard, 1998: 219). Thus, apart from interlingual translation (translation proper), which involves interpreting signs of one natural language by means of another, he also distinguished intraligual translation (paraphrase), i.e., interpreting signs of one natural language by means of other signs of the same natural language and intersemiotic translation (transmutation), i.e., interpreting verbal by means of nonverbal signs (Jakobson, 1959/1971: 261). With time, the latter concept was stretched to embrace the reverse operations, i.e., verbalisations of non-verbal signs (Clüver, 1989; Petrilli, 2003; Pisarkowa, 1998; Tomaszkiewicz, 2006; Wysłouch, 1994). Hence, in Susan Basnett’s classical Translation Studies, it is already defined as a “creative transposition from one system of signs into another” (1992: 15). In his approach to translation as a “subspecies of interpretation” (Eco and Nergaard, 1998: 219), Jakobson endowed the term with “extralinguistic horizons” (Frank et al., 2004: 55), demonstrating that translation studies and cultural semiotics can effectively cross-fertilise each other in their analysis of various forms of communication. This might have encouraged the translatologists to approach a wide range of cultural phenomena, such as restyling, rephrasing, book illustration, film and stage adaptation, ekphrastic literature, illustrative music or the art of museum exhibition in terms of translation. Yet, however appreciative of Jakobson’s innovative ideas (Torop, 2004: 595), translation scholars still focused mainly on interlingual relationships, leaving the investigation of paraphrase and transmutation to lexicologists, linguists, art and literary critics and experts in comparative literature, who work within their own methodologies. As Teresa Tomaszkiewicz points out, even today no monograph is available that would examine the phenomenon of intersemiotic translation from the point of view of linguistics or translation studies (2008: 65). Why? 40 AGATA HOLOBUT Jakobson seems to have endowed the term translation with figurative rather than technical meaning (Eco, 2001: 69). In an interview for Poetics Today, he even remarked: “No translation of a work of art can be a real translation. It is a transposition into different media” (1980: 89). Now, what is unreal about transmutaton? The notion certainly fails to meet “the ideal standards traditionally brought to the evaluation of interlingual translation” (Clüver, 1989: 59). Those ideal standards presuppose translation to involve intentional (i.e., conscious/faithful/equivalent) reproduction and replacement of the original in the recipient culture. By its very nature, transmutation seems to involve neither and consequently poses certain methodological problems. Let us enumerate some of them. As the American comparative scholar Claus Clüver observes in his brilliant essay “On Intersemiotic Transposition”, transmutation never acts as a substitute for the source text in the target context (1989: 57) and rarely does it re-express the original meaning that would be otherwise unavailable to the recipients. Thus, neither John Everett Millais’ portrait of Ophelia nor Franco Zeffirelli’s adaptation of Hamlet are expected to replace Shakespeare’s play. Even the interlingual subtitles or audiodescription accompanying the film do not substitute for the polysemiotic source, but they complement it. Transmutation always preserves a symbiotic relationship with the original, emphasising its existence and demanding the recipient’s contrastive response (ibid.). Because of this symbiotic relationship, transmutation rarely aims at self-sufficient reproduction of the original. It acts as its companion, competitor or descendant rather than a look-alike. Hence, it is either impractical or difficult to evaluate it in terms of equivalence. As the source and target are not cut from the same cloth, how to compare their weave? Finally, Clüver observes, transmutation is also rarely recognised as such by its authors (1989: 69). Directors, book illustrators, screenwriters, choreographers, poets or composers who transpose various works of art into their own media cannot be expected to understand their activity in terms of translation; they have every right to classify it as adaptation, illustration, caricature, staging, accompaniment, concretisation, simplification, amplification, elaboration, variation, etc. Labelling their efforts as intersemiotic translation seems therefore arbitrary. Indeed, only audiovisual translation (e.g., film and television translation or game localisation) offers us “authorised” examples of deliberate intersemiotic translation. Apparently, these methodological problems must have been discouraging for many translation scholars. Consequently, as Frank, Kittel and Greiner point out, “the discussion of intermedial transcodification and intersemiotic translation has hardly begun” and “approaches to practical translation examples are still rare” (2004: 56). However, recent developments in translation studies have aroused a renewed interest in the phenomenon. Both theoretical and descriptive branches of the discipline are ready to explore the concept. Having accepted the fact translation is inextricably linked with projection, transformation and manipulation, contemporary translatologists are better equipped to investigate the complex relations between messages formulated in different codes. Besides, they are bound to investigate them, as the dynamic technological development has forced modern-day translators to become professional transposers and engage in multimodal, multimedial communication. As Gunilla Anderman observes, having recognized the need for paraphrase and transmutation Jakobson “presciently anticipated recently-debated issues and developments in present-day translation studies” (2007: 48). In the face of the growing need for film and television translation, transmutation is apparently becoming “a subject of avid attention” (ibid.), opening up “valuable ideas for further research” (Frank et al., 2004: 56). 2. The application of intersemiotic translation Thus, Roman Jakobson’s concept of intersemiotic translation has become an indispensable theoretical tool for scholars interested in audiovisual translation. Yet the question remains how useful it is for scholars interested in less obvious examples of inter-art relationships, for example that between TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 41 painting and poetry. Hence, in my article I wish to examine the relevance of Jakobson’s notion for the analysis of a ekphrastic poetry, which recreates verbally visual works of art. I will present a case study of William Carlos Williams’ poem “Classic Scene”, analysing it in terms of intersemiotic translation of Charles Sheeler’s painting “Classic Landscape”, in order to find out, whether such an interpretive approach sheds new light on both works of art. Defined as “the verbal representation of graphic representation” (Heffernan, 1991: 299), the term ekphrasis has not enjoyed much popularity among literary scholars, although it dates from about the third century A.D. Literary portrayals of visual artefacts are at east as old as Homer’s description of the shield of Achilles in the 18th book of Iliad (Heffernan, 1991: 298) and they vary from detailed likenesses to allusive charcoal sketches. According to the literary critic Michael Davidson, they can be broadly divided into two groups: poems about paintings, which “pause at a reflexive distance from the work of art” and describe it and painterly poems, which “activate strategies of composition equivalent to but not dependent on the painting” and hence – recreate its aesthetic (quoted in Heffernan, 1991: 299). Considering the ancient tradition and heterogeneity of ekphrastic poems, we might wonder whether and under what circumstances we are allowed to approach them in terms of intersemiotic translation. Which texts qualify as transpositions? Which do not belong to this category? An interesting answer to these question has been offered by Claus Clüver in the aforementioned essay “On Intersemiotic Transposition”. The scholar argues that it is pointless to introduce similar distinctions, because transposition should be treated as an epistemological rather than ontological category. The status of “intersemiotic translation” is not an essential property of particular texts of culture. It is attributable them on the basis of the associations they evoke with other works of art, which we expect to have inspired them. Thus, the notion of transposition originates “in the eye of the beholder” and serves an analytical tool applicable to diverse texts of culture, regardless of their authors’ original intentions. (1989: 70). If we decide to interpret a work of art as a transposition of another, we immediately become sensitive to the possibilities and restrictions inherent in their respective media (Clüver, 1989: 70). According to Clüver, this stems from the fact that transposition does not aims to replace, but to reinterpret the original. Hence, it reveals its real value only in juxtaposition with the original and it performs essentially a self-reflexive function, encouraging the interpreters to notice and contrast the formal and conceptual peculiarities of different codes. As the scholar remarks, in this respect transposition resembles poetry translation: it can be best appreciated “by readers who need it least” (1989: 69). Clüver defines intersemiotic transpositions as signs “permitting the construction of meaning that is very similar to a meaning constructible from a sign in the other sign system” (1989: 83). Thus, if we decide to analyse an ekphrastic poem as a transposition of a painting, we need to specify the meaning it permits us to construct and compare it with the meaning constructible from the painting. If they are similar, we can conclude that a text is a successful translation of a visual work of art. If they are dissimilar, we can either view the transposition as unsuccessful, or revise our interpretive strategy. Yet, in order to evaluate this postulated “similarity of meaning”, we need to find reliable criteria to characterise the semantic make-up of poems and paintings and explain, how they resemble each other, to what degree they are equivalent. Unfortunately, such flexible criteria of comparison are very difficult to find (Clüver, 1989: 71). 3. The analysis and evaluation of intersemiotic translation Hence, in my analysis of an ekhprastic poem in terms of translation, I would like to overcome these difficulties by referring to Polish scholar Elżbieta Tabakowska’s cognitive perspective on translation equivalence, presented in her seminal monograph, Cognitive Linguistics and Poetics of Translation 42 AGATA HOLOBUT (1993). I would like to show that the theoretical and descriptive approach to translation explained in the book is equally valid in the analysis of transmutation. Tabakowska demonstrates that contemporary post-structuralist linguistic concepts can be valuable for contemporary post-structuralist translation theory and criticism. She argues that translatologists interested in a contrastive analysis of source and target texts need precise descriptive tools to compare their semantic make-up and explain the potential similarities and differences between them. These analytical tools can be found within the framework of Cognitive Lingusitics, an interdisciplinary approach to language inspired by recent developments in psychology and neuroscience, which gained popularity in the 1970s, first in the United States and then in Europe. Following a renowned cognitive linguist, Ronald W. Langacker, Tabakowska identifies meaning with conceptualisation or cognitive processing, which “subsumes both new and well-established concepts, sensory, emotive and kinaesthetic experience, as well as cultural, physical and linguistic context” (Langacker, 1991: 2). At the heart of cognitive approach to language lies the constructivist premise that we do not passively register reality, but actively structure it while thinking and expressing our thoughts. As Langacker explains, we have the ability to “construe” (i.e., structure, understand and express) the conceived situations “in alternate ways – by means of alternate images” (Langacker, 1987: 110), which we reflect by verbal means. Thus, a given objective situation can be conceptualised from various perspectives, with various attitudes and at various levels of specificity, focusing on selected aspects and disregarding others. Langacker calls these parameters dimensions of imagery (1991), distinguishing selection, perspective (point of view, orientation, directionality, deixis and figure/ground alignment,) and abstraction. Other scholars have enumerated other parameters that characterize our cognitive experience, such as scanning, comparison or metaphorical projection (Croft and Cruse, 2004: 44). All these parameters have their counterparts in the grammar of a language, which acts as “a repertoire of conventionalized images” (Tabakowska, 1993: 31). Speakers express their conceptualizations, making appropriate choices from this repertoire. In order to explain the meaning of a given utterance, we need to specify, how particular dimensions of imagery organise its conceptual content. In order to compare the meaning of source and target texts, we need to compare the images they impose on their conceptual content. Obviously, different languages symbolise chosen aspects of cognitive experience in different ways. Hence, equivalence is never absolute – it is scalar and it can be defined as “correspondence of images” (Tabakowska, 1993: 74). Tabakowska’s cognitive approach to translation equivalence seems universal enough to apply both to interlingual and intersemiotic translation. Not only verbal, but also visual, aural and kinetic images may correspond to each other, embodying similar perspectives, portraying similar aspects of experience with a similar degree of specificity. If cognitive experience is characterized by such parameters as perspective, selection, abstraction, scanning and metaphorical projection, then we can compare the way they are reflected in verbal, visual, kinetic or auditory images and find out what meaning they allow us to construct. Thus, Langacker’s dimensions of imagery can serve as a tertium comapartionis in the analysis and evaluation of various relationships between source and target texts of culture, such as choreography that reflects architectural styles or poems that recreate paintings. In my analysis of ekphrastic poetry, I approach both poems and paintings as expressions of a particular scene construal. The painter captures in his canvas a conceived situation. He structures it in a chosen way, establishing vantage point, orientation, figure/ground alignment, scale, scope and grain, allowing these parameters to conspire (Tabakowska, 1993) to achieve the intended effect. Obviously, this effect dependents largely upon the artist’s technique and style. An analogous situation takes place in the world of poetry. Here, the author uses verbal means to symbolise his conceptualisation. If the poet wishes to verbally “transpose” a work of visual arts, he may exploit the analogies between TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 43 the perception of a given painterly scene and its linguistic construal. Successful intersemiotic translation consists in the reconstruction of a visual image in all its characteristic parameters. Let us now apply these criteria to describe the relationship between Charles Sheeler’s painting “Classic Landscape” (1931) and his friend, William Carlos Williams’ poem “Classic Scene” (1937). First, we shall characterise briefly Sheeler’s painterly oeuvre and the scene construal he captured in his “Classical Landscape”. Subsequently, we shall view Williams’ poem in similar terms, hoping to find out if and on what grounds it “permits the construction of meaning” that is “very similar” to the meaning constructible from Sheeler’s picture (Clüver, 1989: 83). 4. Precisionism Charles Sheeler (1883–1965), the American painter and photographer, was one of the most prominent representatives of Precisionism, or Cubist Realism, an artistic movement that flourished in the United States in the inter-War period. Begotten in the times of social and economic upheaval, it reflected the industrialisation of American landscape with understated reverence. Sharing the early Modern fascination with technology and progress, visible in the choice of themes (machines, industrial architecture), Precisionism upheld five basic premises. First of all, it assumed that both photographic and painterly image should be as precise as possible. It ought to focus on objects, which are equivalents for human emotion. As the literary critic Peter Schmidt explains, “the tangible thing, with its unique texture” could in their view “articulate the intangible most precisely” (Schmidt, 1980: 387). Capturing the “intangible” on their canvas, the painters wished to remain invisible, as “the ideal artist is as selfless as a lens” and the emotions s/he preserves should be universal too, independent of particular time, place, or personality (ibidem). Particular time and place, however, was not altogether unimportant for the Precisionists, as they wanted to create essentially American, contemporary art, emancipated from the European influences. Thus, as art historian Dickran Tashjian explains, Precisionism was “the art of urban and industrial America, whose external forms, geometric and hardedged, seemed to require a commensurate visual style” (1978: 83). The unique visual style of Charles Sheeler, recognizable in the “Classic Landscape”, fused his experience as a painter with that of a photographer. Having received formal education in industrial drawing and applied art, Sheeler was initially drawn to Impressionism, but he soon abandoned it for realism, in an effort not to betray his presence in the work (Sheeler, 2010). He “distrusted the artist’s role in actively shaping his materials and themes” (Tashjian, 1978: 77) and struggled for years to “develop a formalism that could withstand his laconic deference for the object” (ibid.). Hence, after completing the academic training, he took up commercial photography, focusing on architectural subjects, and started to specialise in “straight” photography, aimed at dispassionate objectivism. His photographic career skyrocketed in 1927 when Ford Motor Company commissioned him to photograph their new River Rouge Plant near Detroit. It was then the largest industrial complex in the world, which employed seventy-five thousand workers and the photographs of the building were used in the company’s advertising campaign. Moreover, Sheeler used them too as templates for a series of oil paintings depicting the plant, which he created over the next six years (Kelly, 2000). 5. “Classic Landscape” as scene construal “Classic Landscape” (Figure 1) belongs to this series. Painted in 1931, it presents the part of the plant responsible for producing cement from the by-products of the manufacturing process. Its central section depicts a silos, which stored the cement before it was shipped for sale (Sheeler, 2010). Let us characterise the semantic make-up of the painting, using Ronald W. Langacker’s dimensions of imagery as a template. 44 AGATA HOLOBUT Figure (1): Charles Sheeler, Classic Landscape, 1931. Collection of Barney A. Ebsworth 5.1 Selection Conceiving of a scene, the conceptualiser always needs to decide which elements to select for explicit attention and how to organise them. Langacker’s concept of selection accounts for this parameter, crucial for both visual and verbal pictures. It determines “which facets of the scene are being dealt with” (1987: 117) and which cognitive domains constitute an image (Tabakowska, 1993: 33). Which cognitive domains does Sheeler’s “Classic Landscape” activate? The scene exudes non-organic, inhuman stasis and balance. Its main participants include anonymous industrial buildings: a tawny silos with a metal structure in front (possibly a gantry crane); two white buildings, one of them seemingly administrative, and seven factory chimneys. These architectural objects are enclosed by two darker forms presented in the foreground: those of cement heaps to the left and the railway track extending diagonally to the right, ending with a freight car at a far distance. These recognizable elements are set against a grey, clouded sky. The composition seems well-balanced and complete, although a part of the cement heap and a stretch of rail tracks actually remain outside the picture frame, revealing the limits of the observer’s visual field. 5.2 Relative salience and figure/ground organisation When observing a scene, the conceptualiser “turns his attention to its particular facets, and not all elements within either immediate or maximal scope are given equal prominence” (Tabakowska, 1993: 46). Consequently, a substructure within a scene that is perceived as “standing out” from the remaining “ground” is called a “figure,” a term borrowed by cognitive linguists from Gestalt Psychology. According to Ronald Langacker’s definition, figure is “accorded special prominence as the pivotal entity around which the scene is organised and for which it provides a setting” (1987: 120). Obviously, the choice of the most salient portion within a scene depends on the conceptualiser, yet there are factors which render some choices more natural than others. Thus, “a relatively compact region that contrasts sharply with its surroundings shows the strong tendency to be selected as the figure” (ibidem). If an entity changes its position relative to the other elements of the scene, it is also more likely to be foregrounded as a figure. Which objects in Sheeler’s “Classic Landscape” might classify as figures? It might be the red silos, because of its central location and the eye-catching colour, which contrasts sharply with the monochromatic background. It is also the rail tracks, because of their dynamic diagonal positioning and the dark shadows they cast, taking up the rhythm of the silos pillars. TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 45 With its entire composition, the image seems to give special prominence to spatial relations between particular elements of the plant. Sheeler entwines the buildings in an elaborate web of parallels and meridians. Thus, he brings to the fore the geometrical order recognizable in industrial architecture, which he often used as a basis of his composition (Tashjian, 1978: 81). Another salient aspect of Sheeler’s scene is the volume and texture of objects. The buildings and rail tracks appear as ideal, smooth, regular cubes of clearly recognizable materials: cement, brick and metal. The landscape appears as realistic and idealised at the same time, its austere coarseness softened by inherent symmetry, regularity and balance. By profiling spatial relations, volume, texture and geometrical order, Sheeler endowed the landscape with timeless dignity, which might account for the title of the picture: “Classic Landscape”. The art critic Franklin Kelly comments on the ambiguity of the adjective. On the one hand, the landscape is “classic”, i.e., typical and standard, the Ford Motor Company being the originator and paragon of standardisation. On the other hand, it is dignified and harmonious, thus exalting the American industrial achievements. It brings to mind architectural heritage of ancient Greece and Rome with the “temple-like” silos and the “pediment-like” roofs (Kelly, 2000). As Dickran Tashjian observes, “balance, harmony and reason contribute to this classic landscape”, where geometrical relationships of the buildings promote “the meaning of an industrial order” (1978: 84). 5.3 Specificity Another parameter that helps us to characterise the semantic make-up of Sheeler’s “Classic Landscape” is the level of specificity, pertaining to the “fineness of detail” with which particular aspects of the scene are viewed (Langacker, 1987: 132). Precisionist paintings deliberately strived at “photographic”, rather than “painterly” surface, achieving sharp universal focus (Schmidt, 1980: 385). Yet this particular image, however “fine-grained”, is not overloaded with excessive detail. In Tashjian’s words, it “achieves a necessary balance of detail and design to stand on its own” (1978: 81). The shape and structure of particular buildings is clearly recognizable, yet their function remains a mystery. We can only identify them as industrial. As Kelly remarks, Sheeler chooses a “rather anonymous scene” instead of one clearly suggestive of the production of automobiles. This implies that he was interested in “making a generalized portrait of the landscape of industry” (Kelly, 2000). Thus, by manipulating the parameters of selection and abstraction, he managed to capture “an industrial complex” rather than “the industrial complex”; an idealized type rather than a concrete token of the plant. 5.4 Perspective And finally, we should consider the most complex aspect of Sheeler’s scene construal: the perspective it embodies. Ronald Langacker defines this parameter as “the position from which a scene is viewed, with consequences for the relative prominence of its participants” (1987: 117), yet it can be explained more broadly as the conceptualiser’s location in spatial, temporal, epistemic and cultural context (Croft and Cruse, 2004: 58). The schematic term subsumes more elaborate parameters, such as viewpoint, orientation, deixis (which can be analysed in spatiotemporal, epistemic and empathic terms) and finally – subjectivity of conceptualisation, i.e., the degree to which the conceptualiser reveals his/her presence in the scene (ibidem). Let us define these concepts and apply them to the description of Sheeler’s painting. As regards the viewpoint preserved in “Classic Landscape”, the anonymous observer watches the buildings from a distance, focusing on their balanced spatial arrangement. As only inanimate participants are present, the scene appears to be static. There are two exceptions to the rule: the smoking chimney and the diagonal rail tracks, which give the impression of dynamism, encouraging the viewer to trace their progress into the distance. 46 AGATA HOLOBUT Another aspect unique for Sheeler’s composition is deixis, which involves “using the elements of the subject’s situatedness”, i.e., spatial, temporal, epistemic, cultural context, to “designate something in the scene” (Croft and Cruse, 2004: 59). How does Sheeler shape the observer’s relationship with the observed scene? Does his “here and now”, his cognitive, psychological and axiological involvement permeate the image? As for spatial deixis, the observer reveals his physical situatedness by capturing only a fragment of the plant’s visual expanse. He reveals the limits of his visual field by presenting the rail tracks and the cement heap as incomplete; allowing them to extend outside the picture frame. As concerns temporal deixis, despite the directness of vision apparent by his painting, Sheeler seems to avoid explicit reference to the “moment of perception” (corresponding roughly to the notion of the “speech event”). By emphasising the permanent visual order, the ideal stasis of architectural cubes, he does not deliberately anchor the scene in the transient “here and now” of the observer. He presents it as both concurrent with the act of perception and timeless. The only changeable element in the painting is the smoking chimney captured in the middle of its activity, which interrupts the stagnation of the landscape. The most important feature of Sheeler’s perspective is the impersonal, matter-of fact, objective attitude of the observer, determined to register and not to evaluate. Langacker accounts for this aspect of scene construal in terms of subjectivity/objectivity, i.e., the manner of conceptualisers’ selfpresentation in the construed scene. The conceptualiser is maximally subjective if s/he remains “offstage” as an “implicit, unselfconscious subject of conception”, who loses all awareness of him/ herself, being “totally absorbed in apprehending the on-stage situation” (Langacker, 2006: 18). The conceptualiser becomes more objective if s/he enters the scene as its explicit participant. Sheeler construes his “Classic Landscape” as subjectively as possible. The viewer seems completely absorbed in the observed scene, reluctant to reveal his existence, following the Precisionists’ belief that “the artist’s own presence must be impersonal, even invisible” (Schmidt, 1980: 387). He is almost transparent, careful not to evaluate the reality but to record it with taciturn precision. As it turns out, however, vision always involves interpretation. Sheeler’s sensitivity to the geometrical order of architecture results in the “overstructuring” of the scene. He puts excessive emphasis on the harmony and beauty of the landscape, visible only to an observer “mesmerised by a commonsense visual reality” (Tashjian, 1978: 77). By focussing on objectivity, Sheeler thematises it, exposes its conscious mechanism. In a sense, he creates a metapicture of his own objective vision. As Franklin Kelley remarks, his “stark interpretations of architectural, industrial, and technological subjects” actually “reflect his love of precise, geometric forms and his strong sense of abstract design” (2000). 6. “Classic Scene” as scene construal Having characterised Charles Sheeler’s “Classic Landscape”, let us confront it with William Carlos Williams’s “Classic Scene” (1937), in order to evaluate how successfully it transposes the painting. As we have mentioned, this is obviously an arbitrary interpretive choice. The poet does not overtly mention the painting as a direct source of inspiration. Yet, critics often emphasise the correspondence between the two works of art (c.f. Dijkstra, 1969; Tashjian, 1978; Schmidt, 1980), treating the text as a “homage to Sheeler’s “Classic Landscape,” (Schmidt, 1980: 393). Williams had every reason to be a successful translator of Sheeler’s works both on personal and artistic grounds. As for the former, he befriended the artist in 1923 and wrote two insightful essays devoted to Sheeler’s aesthetics: an introduction to his MOMA exhibition catalogue in 1939 and an article for Art in America in 1954 (Costello, 1979). As for the latter, among his many fascinations with visual arts, Williams took particular interest in Precisionism, applying its premises in his objectivist writing, “enlarging”, “complicating” and “giving voice” to the painters’ ideas in order to renew the American literary tradition (Schmidt, 1980: 384). He also shared many of Sheeler’s individual aesthetic beliefs. Both artists wanted their work to be “taken objectively rather than subjectively, TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 47 impersonally rather than emotionally” (Tashjian, 1978: 84). As we shall see, the correspondences between “Classic Landscape” and “Classic Scene” (Williams, 1986: 444) go beyond their authors’ programmatic emotional detachment. To quote Tashjian, what is particularly significant for their relationship is the fact that “the poem does not depend upon the painting for its imagery”. Instead, they are “bound together by their aesthetic presuppositions” (1978: 84). Let us check, how these common aesthetic presuppositions are expressed in particular dimensions of linguistic imagery: A power-house in the shape of a red brick chair 90 feet high on the seat of which sit the figures of two metal stacks ̶ aluminum ̶ commanding an area of squalid shacks side by side – from one of which buff smoke streams while under a grey sky the other remains passive today − 6.1 Selection As far as selection is concerned, Williams chooses for his “Classic Scene” different participants than Sheeler. Instead of a red silos, he presents a red power-house and squalid shacks, which are absent from the painting. But still, he conjures up an image strangely suggestive of “Classic Landscape”. Why? First of all, he evokes the same cognitive structures as the painter. He activates the domain of industrialization (represented by such concepts as a power-house and metal stacks). He focuses specifically on materials and substances: brick, metal, aluminium, smoke. He also alludes to the domain of visual perception, evoking shape (in the shape of a chair), size (90 feet high), number (two) and colour (red, grey, buff). This clearly reflects Sheeler’s poetics, his “respect for number, edge, arrangement, and material” (Schmidt, 1980: 394). 6.2 Relative salience and figure/ground organisation As far as figure/ground alignment is concerned, Williams seems to share Sheeler’s compositional sensitivity. The powerhouse, and subsequently the metal stacks perform the role of figures, while the other elements of the scene, such as the shacks and the sky, serve as the ground. The poem’s most suggestive “Sheelery” quality stems from Williams’ decision to profile spatial relationships, establishing distinct geometrical order upon the scene. The physical location of objects is explicitly described: the figures sit on the seat, stacks command the area side by side, a smoke streams from one of the chimneys, while the other is localised under a grey sky. Besides, the spatial arrangement is also imposed graphically. Williams makes extensive use of dashes, which iconically represent Sheeler’s “emphasis upon line” (Tashjian, 1978: 81). 48 AGATA HOLOBUT 6.3 Specificity Moreover, the poet forces us to re-experience the process of “constructing” or “building” the picture, which was characteristic of Sheeler’s visual poetics. Fascinated with the ideal of a machine composed of functional parts, the painter believed that a picture should also be “built” of indispensable, functional parts (Tashjian, 1978: 84). By careful manipulation of selection and specificity, Williams analogously “builds” his poetic picture and encourages the readers to perform the same procedure. In almost every line he introduces a new image of an architectural object: firstly schematic, then more and more precise. Each image enters into a spatial relationship with another. First, we see a powerhouse – more and more specifically with each consecutive line. Then it blends with another image, that of a chair, which again gains clarity, revealing its colour, material and size. Subsequently, metal stacks appear. This time, their position, number and material seem more important than their actual identity and function. First, they reveal themselves as anthropomorphic silhouettes. Then, their exact location is specified. Only afterwards does the observer identify them as stacks and specify the material they are made of, i.e., aluminium. The next stanza is again predominated by spatial concepts. The mysterious squalid shacks form an area, and the stacks are localized side by side. Then the vision focuses on one of the chimneys and its relationship with the smoke. We follow it, turning our attention to the sky and the second chimney under it. Thus, Williams emphasises the significance of spatial composition in his “Classic Scene”. He also imitates the interplay of generalisation and specificity, characteristic of Sheeler’s composition. Our cognitive observations correspond closely to the remarks of scholars working within different methodologies. As the literary critic Peter Schmidt observes, Williams “picks a different setting, but captures Sheeler's spirit by carefully recording the height of the smokestacks on the powerhouse, the industrial colors (red, buff, and grey), the construction materials (brick and aluminum), and the precise relation of the buildings to each other” (1980: 393). 6.4 Perspective Now, let us concentrate on the complex parameter of perspective, embodied in Williams’ poem. The text clearly profiles unchangeable entities and states rather than processes and hence gives the impression of timelessness and stasis, similar to Sheeler’s classic vision. This stems from Williams’ clever manipulation of the parameters of spatial and temporal deixis, as well as subjectification. According to Tashjian, because “Classic Scene” is not even a complete sentence, but a descriptive phrase qualifying a powerhouse, it is “laconic and abstracted, precisely focused on a building in the landscape” (1978: 84). The sense of abstraction results from Williams’ careful epistemic grounding of the scene. He profiles entities and conceals processes. The entities he describes are consistently indefinite. Let us mark the use of indefinite articles in the portrayal of a powerhouse, an area or a grey sky. Williams’ choice of indefinite reference can be interpreted in at least two ways, both of which correspond to Sheeler’s scene construal. On the one hand, we might argue that the objects are presented as indefinite, because they are mentioned for the first time. Williams does not want his readers to enter the scene in medias res. He provides them with raw, unprocessed experience. A powerhouse, an area, a grey sky are presented as previously unknown, absent from the speaker’s and readers’ spheres of knowledge. Reading the poem imitates direct perception of reality, which needs to be captured and structured. This aspect of scene construal imitates Sheeler’s photographic realism. Another interpretation of indefinite reference, equally valid, alludes to the “classic”, timeless quality of Sheeler’s painting. Definiteness anchors the scene in the spatiotemporal circumstances of the speech event. Indefiniteness, however, allows Williams to present the scene as timeless and abstract. Indefinite reference performs in English defining rather than identifying function (Tabakowska, 1999: 416), TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 49 it “triggers” encyclopaedic knowledge about particular concepts. Thus, a powerhouse, an area of shacks, a grey sky are images evocative of generic types rather than concrete specimens. This, paradoxically, also brings to mind Sheeler’s landscape, which transforms the River Rouge plant into a universal symbol of industrial beauty. As Peter Schmidt remarked, “the brute confidence in Sheeler’s paintings of industry” is “impersonal” and “universal” (1980: 387). What is more, Williams also reflects Sheeler’s paradoxical fusion of experiential immediacy and universal timelessness with his ingenious use of temporal deixis. As mentioned, the poem portrays predominantly a network of objects intertwined in spatial (hence – a-temporal) relations. It strikes the reader as static and verbless. However, on closer inspection we realise that the image actually involves a number of states (remaining, sitting) and processes (commanding, streaming). These are presented in Present Simple tense, which can be interpreted in two ways: either as a statement of unchangeable truth, or as a portrayal of an action concurrent with the act of its description. Thus, by saying that smoke streams, Williams either implies this is a an unchangeable state preserved in our structural knowledge (which would render the scene static and generalised), or that the full instantiation of the process of streaming takes place simultaneously with the act of speaking. This, by contrast, signals extreme immediacy of experience and directness of vision, reminiscent of an eye-witness’ account. Again, the former interpretation reinforces the classicism, while the latter – the realism of Sheeler’s painting. On the whole, Williams avoids spatiotemporal deictic signals that would reveal the conceptualiser’s presence. Interestingly, he breaks this rule in the last line of the poem, where he suddenly grounds the entire scene in the conceptualiser’s (i.e., speaker’s/reader’s) deictic today. Apparently, he wishes the Precisionist belief in contact and directness of vision to predominate over the classic abstraction of the scene. He also avoids axiological signals that would reveal the speaker’s personality and renders the scene maximally “subjective”. As Peter Schmidt explains, “in Williams’ Precisionist lyrics, the poet usually speaks with an impersonal voice, the voice of Everyman, and the poem’s action is suspended in time, to be repeated within the reader each time it is read” (1980: 392). Yet again, despite its subjective construal, the scene also bears the mark of the conceptualiser’s cognitive and emotional involvement. It is visible in the personification of the stacks, which sit on the seat and command the area. It is also apparent in the classification of shacks as squalid, which sharpens the contrast between inanimate, industrial harmony and human, organic chaos, but also implies an unusual degree of emotional involvement on the part of the speaker. Indeed, the critics interpret the personification as evocative of Sheeler’s enthusiasm for industrialization and progress (Schmidt, 1980: 393). Yet, the image of squalid shacks smacks more of another painter, Charles Demuth and his “half-ironic attitude” which “goes against the mood” of industrial apotheosis prevalent in Sheeler’s painting (ibid.). Thus, in his transposition, Williams approached the “design” of Sheeler’s landscape the way Sheeler approached the “design” of reality. He combined a copyist’s respect with a connoisseur’s appraisal. He recreated the artist’s unique scene construal, enriching it with his own insights and reflections. Thus, both Sheeler’s and Williams’ Precisionist experience reflects every translator’s dilemma. It is impossible to strive at equivalence without recognizing the “design” of the original. “Design”, however, is tantamount to “the liberating action of the imagination” (Costello, 1979). By unleashing it, we lose the status of a “selfless lens” and get ourselves entangled in “artist’s fabrications” (to paraphrase Williams’ famous commentary on Sheeler, quoted in Tashjian. 1978: 81). 7. Conclusion In my article I explored Claus Clüver’s suggestion that a successful poetic transposition of a painting “permits the construction of meaning that is very similar” to the meaning constructible from a visual 50 AGATA HOLOBUT sign (1989: 68). In order to evaluate this similarity of meaning, I turned to Elżbieta Tabakowska’s (1993) conception of cognitive poetics of translation. I argued that mental imagery finds reflection both in verbal and visual media. Its particular dimensions shape the semantic make-up of paintings and poems alike, serving as a good tertium comparationis in their contrastive analysis. To illustrate these claims, I subsequently analysed “Classic Scene” as intersemiotic translation of “Classic Landscape” only to find out that William Carlos Williams produced a brilliant transposition of Sheeler’s work, reconstructing the original combination of perspective, specificity and selection recognizable the painting. The text clearly classifies as a painterly poem, which uses strategies of composition that are “equivalent to but not dependent on the painting” (quoted in Heffernan, 1991: 299) and hence – manages to recreate a characteristic painterly aesthetic. Interestingly, this small-scale analysis of an ekphrastic poem in terms of transposition allowed me to “re-see” the painting and “re-read” the text from a new perspective. Comparing the painting with the poem, I started to look for visual exponents of deixis, axiological commitment and modality – a rather unorthodox interpretation, which I would not have arrived at outside the translatological framework. Comparing the poem with the painting, I became sensitive to their geometric composition and spatial organisation, which would have otherwise escaped my notice. This confirms Claus Clüver’s observation that “studying the poem by means of translation … will lead to insights that cannot be had any other way, and that are earned by a creative engagement that is different from our engagement with a critical analysis” (1989: 69-70). Figures Figure 1. Charles Sheeler (1931). A Classic Landscape. Collection of Barney A. Ebsworth. Available at: http://www.nga.gov/fcgi-bin/timage_f?object=105879.0&oimage=0&c= (20th December 2010). References Anderman, G. (2007), “Linguistics and Translation”. In: Kuhiwczak, P. and Littau, K. (eds). A Companion to Translation Studies. Clevedon/Buffalo/Toronto: Multilingual Matters. Basnett-Maguire, S. (1992), Translation Studies. London: Routledge. Clüver, C. (1989), “On Intersemiotic Transposition”. Poetics Today 10: 1 (Spring 1989). 55-90. Croft, W. and Cruse, D. A. (2004), Cognitive Linguistics. Cambridge: Cambridge UP. Costello, B. (1979), “William Carlos Williams in the World of Painters”. Boston Review June/July 1979. Available at: http://bostonreview.net/BR04.6/costello.html (19th December 2010). Eco, U. and Nergaard, S. (1998),“Semiotic approaches”. In: Baker, M. (ed). Routledge Encyclopaedia of Translation Studies. London/New York: Routledge. 218-222. Frank, A.P., Kittel, H. and N. Greiner (2004), Übersetzung: ein internationales Handbuch zur Übersetzungsforschung, Vol 1. Berlin: Walter de Gruyter. Heffernan, J.A.W. (1991), “Ekphrasis and Representation”. New Literary History, Vol. 22, No. 2. 297-316. Jakobson, R. [1959] (1971), “On Linguistic Aspects of Translation". In: Selected Writings. Vol. 2. The Hague: Mouton. 260-66. Kelly, F. (2000), Art for the Nation Exhibition Catalogue. National Gallery of Art. Washington D.C. available at: http://www.nga.gov/fcgi-bin/tinfo_f?object=105596 (20th December 2010). Langacker, R.W. (1987), Foundations of Cognitive Grammar. Vol.!. Theoretical Prerequisites. Stanford: Stanford UP. Langacker, R.W. (1991), Concept, Image and Symbol: The Cognitive Basis of Grammar. Berlin: Mouton de Gruyter. Langacker, R.W. (2006), "Subjectification, grammaticalization, and conceptual archetypes”. In: A. Athanasiadou et al (eds.). Subjectification. Various Paths to Subjectivity, Mouton de Gruyter: Berlin/New York. Naylor, M. and Ball, R. (2005), Form Follows Idea: An Introduction to Design Poetics. London: Black Dog Publishing. TRANSLATION AND MEANING, PART 10, 2013, 39 - 51 51 Petrilli, S. (ed). (2003), Translation Translation. Amsterdam/New York: Rodopi. Petrilli, S. (2003), “Translation and Semiosis. Introduction”. In: Petrilli, S. (ed). Translation Translation. Amsterdam/New York: Rodopi. 17-39. Petrilli, S. (2003), “The intersemiotic Character of Translation”. In: Petrilli, S. (ed). Translation Translation. Amsterdam/New York: Rodopi. 41-55. Pisarkowa, K. (1998), Pragmatyka przekładu. Przypadki poetyckie. Kraków: IJP PAN. Schmidt, P. (1980), “Some Versions of Modernist Pastoral: Williams and the Precisionists”. Contemporary Literature, Vol. 21, No. 3. 383-406. Sheeler, C. (2010), An excerpt from an interactive companion to videodisc American Art from the National Gallery of Art. published by National Gallery of Art in Washington D.C. Available at: http://www.nga.gov/cgi-bin/tbio?tperson=2745&type=a (20th December 2010) Tabakowska, E. (1993), Cognitive Linguistics and Poetics of Translation. Tübingen: Gunter Narr Verlag. Tabakowska, E. (1999), “Linguistic Expression of Perceptual Relationships. Iconicity as a Principle of Text Organisation (A Case Study)”. In: Nänny, M. and Fischer, O. (eds). Form Miming Meaning: Iconicity in Language and Literature. Amsterdam/Philadelphia: John Benjamins. 409-422. Tashjian, D. (1978), William Carlos Williams and the American Scene, 1920 – 1940. New York: Whitney Museum of American Art. Tomaszkiewicz, T. (2006), Przekład audiowizualny. Warszawa: PWN. Williams, W.C. (1986), The Collected Poems of William Carlos Williams. Volume 1: 1909-1939. Litz, W. and MacGowan, C. (eds). New York: New Direction Books. Wysłouch, S. (2001), Literatura i semiotyka. Warszawa: Wydawnictwo Naukowe PWN. 52 TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 53 TRANSLATING COGNITIVE BLENDS IN POLITICAL DISCOURSE Rafał Pacholczyk University of Łódź, Poland Abstract: The purpose of this thesis is to introduce the notion of conceptual integration (blending) into the field of translation and test its applications within various contexts. The paper will consist of three main sections. In the first part I will demonstrate the theoretical background behind the cognitive approach to translation and how it factors into conceptual integration. In the second part, I will provide the theory of blending. In the last section I will incorporate examples from political corpora, both in English and Polish, in order to test the theory in relation to translation. 1. Translation theory and the cognitivist Cognitive linguistics can provide useful tools for text analysis within the descriptive branch of translation. It enables a modern and holistic view of language viewed as a culturally grounded phenomenon. As such it provides a more thorough analysis of S and T L texts by revealing implications that otherwise may be hidden. This in turn could enable for a more thorough and more informative conclusions than those drawn from a descriptive research. With regard to translator practice, the cognitive approach could also be of great importance to translators in two stages: analyzing and reading the ST and ultimately in reconstructing its meaning potential for the TT. The first case by analysis of the ST in cognitive terms could reveal the richness of textual meanings and use them accordingly. For the second instance, the knowledge of semantic implications of TL grammatical structures and lexical items would help the translator in choosing the best solutions and constructions carrying meanings as similar to the ST as possible. The cognitive approach also attempts to reestablish linguistics as an important part of interdisciplinary framework of translation studies. It is due to the difference of cognitive linguistics from other models in the context of other phenomena and regards it as culturally and expediently grounded. This is of course a similar attitude to the interdisciplinary trends present in modern translation studies today, which also provide such tendencies in cross linguistic transfer. Traditionally, linguistics and literary studies have been regarded as separate branches of study. Nowadays, the division is perceived as artificial and there have been many attempts to bridge the two disciplines together. Among them is Linguistics for students of literature by Elizabeth Closs and Mary Louise Pratt (1980) and Linqustics and the study of literature edited by Theo D’ahen. In his essay “A question of interpretation and overview of some recent developments of stylistics” Ronald Carter notes that suggests to treat literature as a sort of discourse (1986: 7-26). His views could be regarded as more prototype ideas of “cognitive approaches to literature”. In Poland the cognitive approach is discussed by Dorota Korwin Piotrkowska in her study Telling the world. Cognitive Analysis of literary texts (2006). In her work she discusses the cognitive approach within literary studies, further reinforcing the interdisciplinary trend. Another example is the work of Elżbieta Tabakowska Cognitive linguistics and poetics of translation (1993) where she also approaches the application of cognitive linguistics and translation. In support of her arguments she provides concepts common to contemporary literary studies and linguistics. 54 RAFAL PACHOLCZYK According to Tabakowska’s views the three disciplines that is literary, translation and cognitive studies are correlated and they compromise unifying systems. She argues that ignoring one state of research could lead to shortcomings as seen in previous linguistic approaches. For instance, future studies in cognitive translation theory could overlook how innovations from cognitive linguistics could be beneficial to literary studies. In Korwin-Piotrowska’s work, she notes that cognitive stylistics allows us to “tell the world” with words and after that we can study how the translator “retells” it. Cognitive linguistics has been developed as a response to restriction of previous theories of language, particularly structural and Chomsky’s linguistics. As Langacker states, one of the leading theorists behind the CL theory, remodeling the previous problems as well as other dilemmas requires” a fresh perspective, an innovative approach permitting us to attack traditional problems in new and productive ways” and “demands of linguists a radical conceptual reorientation” (1987: 2). However cognitivism is not an entirely new discipline, quite the opposite in fact. As, once again acknowledged by Langacker “the insights of traditional grammar are deeper and its contributions greater, than its critics tend to realize” (Langacker 1968: 9). Thus this theory builds upon previous approaches in a critical way, verifying their assumptions in a critical manner with regards to neighboring disciplines. The cognitive model notes that “language is neither self-contained nor describable without essential reference to cognitive processing” (Langacker 1991: 1) this of course undermines some of the basic assumptions of structuralism such as the autonomy of linguistics as a discipline and the determinacy of meaning. In turn cognitive linguists view language as an interpretation of reality. What is more they view language as part of other cognitive abilities, not as an isolated entity. As Fauconnier notes “In contrast to this sharply autonomous view of language structure, cognitive linguistics has resurrected an older tradition. In that tradition, language is in the service of constructing and communicating meaning, and it is for the linguist and cognitive scientist a window into the mind”. He also proposes the idea of the structure of human cognition: “Language is only the tip of a spectacular cognitive iceberg, and when we engage in any language activity, be it mundane or artistically creative, we draw unconsciously on vast cognitive resources, call up innumerable models and frames, set up multiple connections, coordinate large arrays of information, and engage in creative mappings, transfers, and elaborations. This is what language is about and what language is for.” (http://cogweb.ucla.edu/ 1999). What is more language itself is also treated in a holistic level. Cognitivists argue against treating language components such as syntax and semantics separately. Instead they propose that semantics influences syntax and our mental choices are picked from an inventory of grammatical constructions. Cognitivists also argue that there is no objectivism of meaning. Instead they support subjectivity in that they equate meaning with mental experience and state that there is no one correct “view of the world” (Lakoff 1987: 9). This of course leads to the idea that there is no perfect translation either. As Tabakowska notes language is not a “perfect mechanism to produce words and sentences but who dicerm in language the essence of humanity, uniqueness and personality of a user, unpredictability and originality and cultural involvement, as the factors determining the way we use language, which we both create and depend on” (1993: 5). Cognitivists also advocate the prototype approach, based on the ideas developed in the early 1970’s by Elanor Rosch. She calls into question the typical theories of categorization where instead of searching for common attributes she proposed the notion of “family resemblance” between category members. In her view, categories have fuzzy boundaries and are not perceived as homogenous and TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 55 defined in terms of distinctive features or residing within strict boundaries of category membership, but structured around a prototype (Taylor 1992: 42-46). An example of this is a noun “bird”, which has more common members like “sparrow” and less common like “ostrich”. In this sense each category is defined with relation to its prototype, perceived as a more representative member or as a “schematic representation of the conceptual core of a category” (Taylor 1992: 59). This view was later developed by Langacker who explains the ideas of schemas as “an abstract characterization that is fully compatible with all the members of the category it defines” (1987: 31). These notions are compatible and reflect the ideas present in cognitive processes like image schemas and mental images. The prototypical view is also applicable in translation, where it can substitute protypical lexical items or grammatical structures for a marginal one which may in turn result in markedness not present within ST. This approach account for exceptions, marginal cases, indeterminacy, and free concepts. It thus deals with central and less frequent instances, which can be often unaccounted for in generative grammar; as it mostly deals with prototypical cases. A closely related mechanism that is basic level categorization is also many applications for translators. In most cases, according to research those examples that are found in the middle of hierarchies are psychologically most generic and generate the largest number of attributes. For instance, the verb “run” has several subordinate terms such a s:” limp” ,”stride”, “go” and so on. Above that, this verb is a member of the category “move” which is very general and can be understood within basic level instantiations. According to Linke, this approach to categorization “provides researchers with valuable insights concerning the asymmetry of linguistic systems as regards abstraction levels” (2008: 124). Similar concepts, as Langacker stresses are perceived not as accumulations of attributes or semantic primitives but against cognitive domains which they subsequently activate. To better illustrate the concept he provides the example of “knuckle” which is characterized in the context of “finger” (Langacker 1987: 147-148). He argues that most linguistic units need several domains which together are referred to as a ‘matrix’. With regards to images, Langacker stresses that they “manifest our capacity to structure or construe the content of a domain in different ways” (1991: 5). Taking the cognitive framework into account it can be noted that each conceptualizer may associate distinct attributes to a given concept, besides the most prototypical ones. In other words cognitivism postulates an encyclopedic view of human knowledge, an accumulation of collection that is dependent on our own personal experience. In turn, certain facts are less common to some users than to others. In Langacker’s words: “there are no sharp divisions between domains belonging to the meaning of the concept that are linguistically relevant and extralinqustic ones, which undermine the distinction between semantics and pragmatics and supports the encyclopedic view of semantics” (Langacker 1987: 154-166). As a consequence symbolization is present at all stages of a linguistic system, which is treated in a hoillistic way. This model thus becomes useful to translation theories. This stance was also supported by George Steiner who notes that “every language structures and organizes reality in its own manner and thereby determines the components of reality that are peculiar to this given language” (1984: 44) What this means is that conceptualization provides not only the mental processes of an individual but also the socio-cultural context of a given community. Thus this important to the translator who, as Linke notes, can be perceived as a “second conceptulizer” but his perception concern “the textual world that invokes an image in the translator’s mind that was already conceptualized by the ST” (2008:129). As it was stated above some basic premises that constitute translation theory are also present with cognitive linguistics. Among them the interdetemrinacy of meaning and its dependence on our expe- 56 RAFAL PACHOLCZYK rience. This in turn proposes fuzzy boundaries to concepts and in turn the prototypical approach to categorization. With regards to literary studies it also shares some common attributes such as dependence on the reader’s knowledge of texts and their influence on the interpretation of a given literary work. This is related to the, discussed above, notion of the encyclopedic view of human knowledge which is also assumed to exist by cognitive linguists. Another important factor is the fact that cognitive grammar is concerned with conventional character of linguistic structures which helps it in being more useful than other approaches based on plurality of meanings. This becomes most important with regards to translation theory. In short, thanks to its analogies in approach cognitive linguistics can be used to analyzing translation works. What is more, according to Linke , this theory can compensate for the shortcomings of other translation theories. That is to say, the cognitive model uses the prototypical interpretation which is never final, but aware of the inferences from its linguistic expressions (Linke 2008: 131). 2. Translating metaphors Metaphor has been discussed in various contexts. In terms of its cognitive approach the most influential work has been published by Lakoff and Johnson “Metaphors we live by” (1980). In terms of translation however, it has been disregarded for many years. It wasn’t until Newark’s theory was put into light when the idea of translating metaphors has been viewed in a more cohesive manner. Conceptual metaphor has its roots in the cognitive imagery theory discussed above. Similarly this concept is the basis for the cognitive blending theory which is the main area of study in my thesis. That is why I find it important to discuss the concepts underlying the understanding of this theory and its implications towards translation. From the traditional point of view metaphors have been perceived as devices of extraordinary, special language; sort of tools for poets to use. It has also been perceived as exclusive to language, to words as opposed to thoughts and actions. For this reason it has been disregarded by most people. In the 1980’s , both George Lakoff (a cognitive linqust) and Mark Johnson (a philosopher from the University of Oregon) proposed a new approach. In their own words they perceive metaphor not as an elusive device but rather one “pervasive in everyday life, not just in language but in thought and action. Our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature.” (1980: 8). The issue of metaphor translation has been touched upon in the latter part of the twentieth century. Among them Eugene Nida (1969), Rolf Kloepfer (1967 in Dagut 1976), Katharina Reiss (1971 in Dagut 1976), Menachem Dagut (1976) and Jean-Paul Vinay and Jean Darbelnet (1995). Most of these studies have come to a number of conclusions that is: 1. Metaphors cannot be translated 2. Metaphors can be translated literary 3. Some metaphors can be translated while others cannot. Out of the three views the third one seemed the most plausible. The advocate of this notion was Dagut. It was later supported by Peter Newmark and Raymond van den Broeck. His theory on the translatability of metaphors was based on the idea of common ground between the SL and TL cultures. He viewed metaphors as ”individual flash of imaginative insight” (1976: 22) and such they were highly culture specific. He argued that the main purpose of a metaphor is to shock its readers creating an aesthetic impact. He argued that if this effect couldn’t be retained in the TL then the metaphor was untranslatable. Many authors agreed that sometimes the translation the image from the SL to TL was not possible (for instance when the image is unknown in the TL). Therefore a number TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 57 of alternative solutions have been proposed. For instance van den Broeck proposes the fallowing solutions (1981: 77): 1. Translation ‘sensu stricto’ (i.e., transfer of both SL tenor and SL vehicle into TL). 2. Substitution (i.e., replacement of SL vehicle by a different TL vehicle with more or less the same tenor). 3. Paraphrase (i.e., rendering a SL metaphor by a non-metaphorical expression in the TL). Raymond van den Broeck proposes these sollutions as theoretical possibilities. In light of DTS theory he sees the task of translating metaphors as explaining and describing identified solution not in showing how metaphors should be translated (Shäffner 2003: 3). In contrast to this approach Newmark proposed his theory as a set of guidelines for translating metaphors and translator training. His work cathegorised metaphors into five types: dead, cliche´, stock, recent, and original. The methods can be summarized as follows (Schäffner 2003: 4): 1. Reproducing the same image in the TL, e.g., golden hair—goldenes Haar. 2. Replacing the image in the SL with a standard TL image which does not clash with the TL culture, e.g., other fish to fry—d’autres chats a´ fouetter. 3. Translating metaphor by simile, retaining the image, e.g., Ces zones cryptuaire ou´ s’ e´labore la beaute´.—The crypt-like areas where beauty is manufactured. According to Newmark, this procedure can modify the shock of the metaphor. 4. Translating metaphor (or simile) by simile plus sense (or occasionally a metaphor plus sense), e.g., tout un vocabulaire molie´resque—a whole repertoire of medical quackery such as Molie`re might have used. 5. Converting metaphor to sense, e.g., sein Brot verdienen—to earn one’s living. This procedure is recommended when the TL image is too broad in sense or not appropriate to the register. However, emotive aspects may get lost. 6. Deletion, if the metaphor is redundant. 7. Using the same metaphor combined with sense, in order to enforce the image. The cognitive approach to metaphor as seen in Metaphors we live by is a more recent endeavor (mostly developed in the 1990’s) with regards to translation theory. As it was mentioned in the previous section, metaphors are perceived not as decorative tools but basic resources for human thought and processes. They are means of understanding one domain of experience in terms of another. The source domain is mapped to the target domain and the components of the source domain are transferred to the target domain. In translating metaphor, conceptualizations are also important. The question of translatability of a given metaphor is linked to the conceptual systems present in source and target cultures. 3. Conceptual integration In the following segment I will discuss the notion of conceptual integration, which compromises the use of mental spaces, and their correspondences. A mental space contains a partial representation of entities and their relations in a given scenario. As perceived by the speaker such a scenario invokes elements representing the entities and frames showcasing relations between them. Mental spaces can be modified with the change of the topic of discourse. One of the theories sees mental spaces working on neural level, available in active memory but is also built up by activating structures form long term memory. For instance in a sentence “When I was 11 I used to read comics” the person built an identity connector between him now and when he was 11 despite the differences. Further, I will also discuss the implications of blending on translation as well as provide analysis of the process. The theory, as outlined by Faconnier and Turner is best described as “a set of non-compositional processes in which the imaginative capacities of meaning construction are invoked to produce emergent structure (Fauconnier and Turner, 1998). The process is used for complex feats of reasoning as well as to cognitive models which are very simple. 58 RAFAL PACHOLCZYK To better illustrate the process consider an example. In office, two workers are bored. One of them throws a crumbed piece of paper into a trash can. After a few moments, the two men begin shooting paper “balls” into a trash “basket”. As a result a game of “basketball” is born. This process can be seen as a basic example of conceptual blending. In order to for the game to be understood, it must involve knowledge structures from different domains. In the theory of blending established frames (inputs) are combined and in turn they create a hybrid frame (a blend). The blend is comprised structures from different inputs but it also creates its own unique ones. In the example, the two input domains are the trash disposal and basketball. The resulting structure incorporates bits from both spaces as well as an emergent structure of trashcan basketball – whose properties are different from that of both input spaces. Mental spaces are used to represent a number of aspects of the conceptual blending theory. For one, the partial nature of their representations comprised from elements, their higher-order relations and the resulting elaboration which is an important part of background knowledge. What is more, the access principle, which allows the transfer of a term from one space to another, establishes mental spaces as good framework for metaphoric language (Coulson 2001: 117). In order to understand the properties of trashcan basketball, from the previous example, the relationship between established domains and novel concepts must be apprehended. For such purposes Fauconnier and Turner have argued that the process of conceptual integration must take place in conceptual integration networks (1998). A conceptual integration network comprises mental spaces structured with frames which the speaker constructs from contextual information and background knowledge. The prototypical network is constructed from four mental spaces: two input spaces , a generic space and a blend. The model can be represented by the use of a table or in graphic form (Fauconnier 2001: 2): Figure (1): Conceptual blending network Such a network is established for on-line, dynamical cognitive processes people construe for the purpose of thought and action. The elements of this process, as mentioned above are: • • Mental spaces – represented by circles in the illustration. Cross-space mappings of counterpart connections – represented by solid lines. These connections can be of different kinds: between frames and roles in frames, identity transformation, metaphoric connections. TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 • • • 59 Generic space – this is the structure that, as the process of integration unfolds, belongs to both input spaces. In any moment during the construction of the process, the input space can map to any of the inputs. A given element from a generic space maps to its counterparts in the other spaces. Blend – structure from both input spaces is transferred to the third space – called a blend. A blend contains the structure form the generic space as well as more specific structure that is impossible for the inputs. Selective projection – the projection from the inputs is, in most cases, partial. Not all elements are transferred. There are, however, more simple as well as more complex networks than the prototypical one discussed above. For instance, the simplest kind is the single framing network. It consists from two input spaces and a blend. The end result of conceptual integration involves three processes – composition , completion and elaboration. Each of them can produce emergent structure. The three elements can be described as follows: • • • • Composition – the process of attributing a relation from one space to elements from other input spaces. Fusion is one kind of such composition. Counterparts can be brought into the blend as separate elements or as a fused element. In some cases the elements can be distinct entities. The emergent structure arises from contextual accommodation of a predicate from one domain to apply to elements in a different domain (Coulson 2001: 122). Completion – occurs when structure projected from inputs matches with the information from long term memory. It is done mostly subconsciously. The composed structure becomes complete with the other structure. The prototypical subtype of recruitment is pattern completion. The process is closely related to elaboration (Fauconnier 2001: 5). Elaboration – evokes novel structures in the blend in response to given information. The novel structure can be activated by mental stimulation or on relation with exiting blended models. New concepts and logic arise through the process of elaboration. The blend can be “elaborated” on indefinitely, and blended spaces can be extremely elaborated (Coulson 2001: 123 and Fauconnier 2001: 6). Emergent structure - Composition, completion, and elaboration lead to emergent structure in the blend; the blend contains structure that is not copied from the inputs (Fauconnier 2001). The process of conceptual blending can be perceived at many levels of analysis: in morphological structure of single words, concepts, sentences and even conceptual domains. 4. Translating cognitive blends A blend, as discussed in the previous chapter, consists of a few components. Most notably mental spaces. According to Langacker they are a referential structure that proposes a conceptual organization between the described situation and the linguistic structures that are used to describe it (Langacker, 1993). They contain a partial representation of entities and their relations in a given scenario. This invokes elements representing the entities and frames showcasing relations between them. Mental spaces can be modified with the change of the topic of discourse. If all elements in a given space become known, entrenched they become a frame. In turn meaning construction employs mapping cognitive models from one space into another. The process which helps in evoking certain representation upon hearing a given word, and in turn transports elements form one space to another is referred to as the access principle. Another component of the CBT, mentioned previously, is a frame and the process of frame sifting. When a mental space become entrenched and can become a frame. When people access long term memory they used said frames to provide the information. Frames can be seen as representations 60 RAFAL PACHOLCZYK with default values and slot/ filler structures. The process of frame-shifting enables such frames to transfer participants, structures and other data from one such constructed frame to another entity. Finally, the process of conceptual integration (blending) consists of all those elements. The basic conceptual integration network uses mental spaces, structured with frames to create a blend. The most prototypical network is a 4 space model with a generic space, 2 input spaces and a blend space. The generic space stores information common to both input spaces, the 2 input spaces store information store their exclusive information invoked by a given scenario and the blend which stores some of the information present in the other spaces as well as structures of its own. This process is called selective projection. Additionally the end result of structuring a blend involves the processes of composition, completion and elaboration, which I have also discussed in the previous section. With regards to translation, A. McElhalon (1997: 84) proposes to add an additional space called the translation space which could be linked to the blend in the SL network. Similarly, Mandelblit extends the idea and notes that in the interpretation: “The language receptor (hearer/reader) receives as an input the linguistic expression (the linguistic 'blend' generated by the speaker) and attempts to reconstruct the linguistic and conceptual blending operation performed by the speaker. A correct mapping reconstruction (as well as additional elaboration and semantic "pattern completion" of the blend) would (ideally) lead to the generation of conceptual constructs in the mind of the language receptor which are quite similar to the constructs from which the speaker generated the sentence. These conceptual constructs constitute the "interpretation" of the sentence” (Mandelblit 1997: 190). In fact, he proposes that a translation process is composed of two blending processes: one, which he calls the reverse blending operation, is involved in the interpretation of the source sentence and the other involved in the generation of the target text. When translating a sentence (from SL to TL), the translator analyzes the outcome of two independent blending processes. One blend involved in the creation of the source text (done by the source text author) and the other in the generation of the target text (done by the translator). Each of the two blends can reflect a different mapping and integration operation. Therefore each blend can highlight different aspects of language. In turn, the analysis should be concerned with the link between the two blending configurations, which motivate the generation of source and target forms. The basic graphical representation of the process is provided below (McElhalon, 1997: 194). TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 61 Figure (2): Translation as the outcome of two blending operations (McElhalon 1997:194). From the cognitive point of view, the translator’s first task is to interpret the SL input. The SL sentence is in itself a blend created by the author of the text. The interpreter employs an operation where he examines the particular mapping that the author used to construct the sentence. This process is seen by McElhalon as reverse blending:” reconstruction aims at restoring the original conceptual structure which was linguistically "compressed" by the author into a single clause.” (McElhalon, 1997: 195). The act of interpretation of a given text typically involves incorporation of additional information based on “common sense” knowledge and prototypical scenarios. Any present lexical items provide content of partial aspects. For instance any participants in a given scenario or predicates. The next step is to create a translation. The main aim here is to transfer the meaning of the SL structure into the TL structure. This involves finding a structure which would invoke the same kind of response in the TL as it does in the SL. From the cognitive blending point of view this involves “involves finding a syntactic pattern in the target language whose associated semantic schema and "rhetorical function" best correlates with the conceptual event structure to be communicated (transferred), and the rhetorical effect the author of the source text intended to convey (as reflected in the choice of grammatical construction by the source text author)” (McElhalon, 1997: 195). At this point, in order to create a fitting target structure, a new linguistic blending process must take place. While independent from the 1st one it still uses some of its elements. That is to say, some elements from the blend in the source sentence are mapped into the slots of the second integrated structure. This whole process can be seen as an interaction of blending processes. From the cognitive point of view a successful translationfirst requires a “de-integration” of the source structure into conceptual and linguistic units and then “re-blending” of some of those structures in the TL construction. Any divergences in translation may thus result from differences present in blending structures chosen to represent the same event scenario. My analysis of translationexamples will be focused on any cognitive linguistic operations involved in the process. This will be based on the cognitive blending theory discussed in the previous section. 62 RAFAL PACHOLCZYK My goal will be to analyze the blending operations which underline the change of form one language into another. My main questions of interest will be: how can this change of form be accomplished? what kind of conceptual and linguistic operations are involved in creation of the TT? Can a blend be successfully translated?. For this purpose I will provide example from three types of corpora I have prepared. The topics will be political speeches , religious and humor (jokes). 5. Translating cognitive blends in political discourse - an analysis In this sub-section, I will provide some examples of cognitive blends present in political discourse. The examples will be taken from a mini- political corpus that I have created for the purpose of this study. This corpus incorporates speeches raging from the early XX century to the beginnings of the XXI century. It is comprised of about twenty thousand words for each of the two languages. My main point of analysis will be the resulting blend and whether the translator was successful in analyzing the SL structure. This study is based on the theory proposed in the previous section by McElhalon (1997: 87). The first example, is taken from Winston Churchill’s speech given at Futon on March 5 1946. The speech was mainly concerned with the perception on the Communist East. First, the SL blend example: 1) From Stettin in the Baltic to Trieste in the Adriatic an iron curtain has descended across the Continent. Behind that line lie all the capitals of the ancient states of Central and Eastern Europe. Warsaw, Berlin, Prague, Vienna, Budapest, Belgrade, Bucharest and Sofia; all these famous cities and the populations around them lie in what I must call the Soviet sphere, and all are subject, in one form or another, not only to Soviet influence but to a very high and in some cases increasing measure of control from Moscow. Here the SL author presents a blend incorporating two distinct scenarios. The generic space of PLACE is common to both inputs. It has its elements and participants. The beginning of the sentence presents the mental space of GEOGRAPHIC MAP with its elements (countries) and participants (capitals).What is more by introducing the expression of “Soviet Sphere” the author creates another element of shape. On the other hand, the well-known expression, “iron curtain” creates a mental space of STAGE with the participants (actors) and the location (stage) and the typical shape (flat). The blend in this example incorporates the two mental spaces and by the process of elaboration takes participants from input space 1 (countries) and elements from input space 2 (stage). Additionally, the incorporation of “influence” and “control” create the mental space of an ACT. In the first mental space it has negative connotations while in the second they are positive. In the blend, again by process of elaboration and completion, the structure remains negative. The resulting representation of the blend can be seen in the table below. Generic space PLACE Elements Participants Effected entities Shape (element) Act (Participants, Effected entities, Result) Input space 1 GEOGRAPHICAL MAP Elements Steltin, Baltic, Triste, Adriatic, Moskow Warsaw, Berlin, Prague, Vienna, Budapest, Belgrade, Bucharest and Sofia Sphere (none) Act (Moskow. Warsaw, Berlin, Prague, Vienna, Budapest…, Control) Input space 2 STAGE Elements Actors Viewers, listeners Square (curtain) Act (Actors, Viewers, Entertainment) Table (1): The “iron curtain” blend Blend GEOGRPAHICAL STAGE Elements Steltin, Baltic, Triste, Adriatic, Moskow Warsaw, Berlin, Prague, Vienna, Budapest, Belgrade, Bucharest and Sofia Sphere (curtain) Act (Moskow. Warsaw, Berlin, Prague, Vienna, Budapest…, Control) TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 63 Now let us look at the TL translation of the fragment in TL: A) Od Szczecina nad Bałtykiem do Triestu nad Adriatykiem opuściła się żelazna kurtyna w poprzek kontynentu. Poza ta linia znajdują się wszystkie stolice byłych państw środkowej i wschodniej Europy: Warszawa, Praga, Wiedeń, Budapeszt, Bukareszt i Sofia - wszystkie te sławne miasta i zamieszkująca wokół nich ludność leżą, że tak się wyrażę, w sferze radzieckiej i wszystkie, w takiej czy w innej formie, podlegają nie tylko wpływom radzieckim, ale kontroli z Moskwy w bardzo wysokim, niekiedy rosnącym stopniu. Thanks to the universality of the THEATRE space neither its participants nor elements are subject to any changes. The same situation can be seen in the GEAGRAPICAL MAP space. The situation present in this space is universal to both countries and as such incorporates the same elements. From the cognitive point of view both input spaces in the S and T language produce a semantic schema which invokes the same kind of response in both structures. In turn the two conceptual event structures present in the original correlate to the two structures present in the translation. As a result the blend produces the same kind of connotations as the SL one and can be seen as an example of successful translation. The next example comes from a speech given by Martin Luther King in his famous speech “I have a dream” delivered at Washington DC on August 28 1963. The main topic was racial equality and was perceived as a defining moment for the American Civil Rights movement. The extract comes from the beginning part of King’s speech where he talks about the need to fight for rights: 1) Now is the time to make real the promises of democracy. Now is the time to rise from the dark and desolate valley of segregation to the sunlit path of racial justice Here the second sentence is of particular interest. The blend is produced from the basic four space model. The expression “rise from the valley” creates the mental space of JOURNEY. In this case we have the participants (we), the starting point (valley), its characteristics (dark and desolate ) the type of journey (rise) and the end point (path) with its characteristics (sunlight). The second mental space, made apparent by the first sentence, is that of OTHER REIFICATIONS. The participants are the same. The elements i.e. the starting point (racial segregation) and the end result are different (racial justice). The generic space in this example can be considered to be a PATH. The participants would be travelers, the starting point of A and the ending point of B. The final effect of the path would incorporate its participants and the effect it has on them. In the case of the first input space the effect of the journey would be reaching the destination point. The second input space, however, would produce the effect of (change of perception). The graphical representation can be seen below. Generic space Path Elements Participants Starting point (type) Input space 1 Journey Elements The Afro-Americans Valley (dark, desolate) Ending point (type) Path (sunlight) Result (the effected, the effect) Destination (travelers, end of journey) Input space 2 Other reifications Elements The Afro-Americans Racial segregation (injustice) Racial equality (justice) Racial equality (AfroAmericans, equality) Blend Racial journey Elements The Afro-Americans Racial segregation (dark, desolate) Racial equality (sunlight) Racial equality (AfroAmericans, equality) Table (2): King’s “I have a dream” speech blend. Next, we will look at the Polish translation of the fragment. This translation was done in 2006 by a webpage editor: 64 RAFAL PACHOLCZYK B) Teraz jest czas by uczynić prawdziwą obietnicę demokracji; teraz jest czas by wyjść z ciemnego i izolowanego padołu segregacji na nasłonecznioną drogę rasowej sprawiedliwości; Back translation: Now is the time to make true the promise of democracy. It is time to go out of the dark and isolated segregation vale into the path of racial justice In this blend, we once again share some of the same elements with the TL structure. The participants stay the same. However the use of “padoł” (vale), coupled with dark and isolated, creates a more negative connotation then the use of vale. Additionally, the use of “segregation” can create both the same element, but it can also create the element of CREATION and CONSTRUCTION. The NKJP (National Polish Language Corpora) produced more results (about 500/1000) of “segregation” as regards to “racial segregation” (only 5/1000). This proves that the term “segregation” used in the TL is more likely to produce the element of CREATION and CONSTRUCTION than Racial Segregation. As such, this mental space has a more positive connotation than the one used in the SL. There are also other differing factors, which however produce the same equivalent effect. The type of journey is different (go out as opposed to rise), however it still retains the same equivalent effect of going from point A to point B. Furthermore, without knowing the rest of the speech, “rasowej sprawiediwości” (racial equality) can either mean “equality of races” or “true justice”. As such it can either create the element seen in SL blend or JUSTICE without the racial undertones. Still, the effect of a journey to racial equality is retained to a point (Table 3). Generic space Path Elements Participants Starting point (type) Input space 1 Journey Elements The Afro-Americans Vale (dark and desolate) Ending point (type) Path (sunlight) Result (the effected, the effect) Destination (travelers, end of journey) Input space 2 Other reifications Elements The Afro-Americans Creation and construction (segregation) Racial justice (equality) Racial justice (AfroAmericans, equality) Blend Racial journey Elements The Afro-Americans Segregation (dark, desolate) Racial equality (sunlight) Racial equality (AfroAmericans, equality) Table (3): Polish translation of the “I have a dream” fragment. The third example is a speech delivered by prime minister Tony Blair to the Polish Stock Exchange on the 06 October of 2000. It was delivered in Warsaw in the presence of Czech, Slovenia and Polish Prime Ministers and Foreign Affairs Minister of Hungary. The speech is described as one in which Blair: “upholds the vision of a European Union which, while retaining its unique intergovernmental and supranational structure, will become a superpower comprising equal partner Nation States but not a super state. In his opinion, in place of a European Constitution, it would be preferable to draw up a statement of principles, serving as a charter of competences, which would be a political, not a legal document.” (adopted from www.ena.lu, Digital library on the history of Europe). The first example comes at the very beginning where Blair comments on Poland: 2) A few weeks ago, you celebrated the twentieth anniversary of the extraordinary revolution that gave birth to a movement called Solidarity. Poland grew to be the icebreaker for the end of communism in all of Europe and for the end of the Cold War. The second is concerned with the view and prospects of Europe: TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 3) 65 The trouble with the debate about Europe’s political future is that if we do not take care, we plunge into the thicket of institutional change, without first asking the basic question of what direction Europe should take The third example produces the generic space of JOURNEY. The mention of Poland and “the end of communism in all Europe and the Cold War” creates a DESTRUCTION space. We have the active party (Poland) and the affected party (communism). We are also introduced to another element that of the initial state (Cold War) and the resulting state (end of Cold War). We also introduced to the means by which the effect is reached (icebreaker). The second input space is created by the words “give birth” and “grew” which creates the PERSONIFICATION space. Here we have the entity that is evolving (Poland) , the initial state is of infant (Solidarity) and then adulthood (icebreaker). We also have the effected entities of this evolution; one positive (revolution) and negative (communism). We are also presented with the effect (end of Cold War). The blend incorporates participants from the DESTRUCTION space while taking some effected entities from the PERSONIFCATON space. The resulting blend could be than described as a combination of LIFE and CONFLICT spaces (Table 4). Generic space JOURNEY Input space 1 DESTRUCTION Input space 2 PERSONIFCATION Elements Active party Affected party Elements Poland communism Initial state (type) Cold War Final state/effect (means) End of Cold War Elements Poland Revolution, communism Infant (Solidarity movement) Adult (icebreaker of Cold War) Blend LIFE AND CONFLICT Elements Poland Revolution, communism Cold War (Solidarity movement) End of Cold War (icebreaker of Cold War) Table (4): Tony Blair’s LIFE and CONFLICT blend. Next we will look at the Polish translation. It was done by Ilona Duchowicz for the Polish daily newspaper “Gazeta Wyborcza”. The extract reads as follows: B) Kilka tygodni temu Polacy obchodzili 20. rocznicę niezwykłej rewolucji, która zrodziła "Solidarność". Polska zapoczątkowała erę upadku komunizmu w całej Europie oraz spowodowała zakończenie zimnej wojny. Back translation: A few weeks ago, the Polish people celebrated the twentieth anniversary of the extraordinary revolution that gives birth to Solidarity. Poland initiated the era of communist fall in all of Europe and caused the end of Cold War. As it can be seen there are a number of differences. At first the TL translator introduces the PERSONIFICATION space by using the structure “zrodziła” (gave birth). This however is abandoned in the second sentence as there is no mention of “grew” anymore. On the other hand, it creates the mental space of CONSTRUCTION and CREATION by mentioning “initiated”. In this case we have the entity that starts the construction (Poland), the effected entity (Cold War) and the process (fall of communism). As there is no direct relation between the two inputs spaces present the translation can be seen as an example of a simplified blend (see chapter 2). From the cognitive point of view then, the blend in the TL does not convey the same type of information as the one present in the SL. As a result the conceptual event scenario present in the TL does not evoke the same kind of rhetorical function and se- 66 RAFAL PACHOLCZYK mantic schema as the one present in the SL structure. Despite failing to successfully translate the blend from one language into another from a strictly linguistic point of view the message is left intact (table 5). Generic space JOURNEY Input space 1 PERSONIFICATION Elements Active party Affected party Elements Polish people Revolution, communism Initial state (type) Final state/effect (means) Input space 2 CONSTRUCTION and CREATION Elements Poland Revolution, communism Cold War (communism) End of Cold War (caused) Blend CONSTRUCTION and CREATION Elements Poland Revolution, communism Cold War (communism) End of Cold War (caused) Table (5): Polish translation of the Tony Blair speech blend. The fourth example first creates a mental space of JOURNEY. In this case we have the active participant (We- Europe), the type of journey (political), the destination (institutional change), and the effect (no direction). The second mental space is set up by using the word “thicket”. From Encyclopedia Britannica 2008 thicket is defined as: “a dense grove of small trees or shrubs that have grown from suckers or sprouts rather than from seed. A coppice usually results from human woodcutting activity and may be maintained by continually cutting new growth as it reaches usable size.” (Encyclopedia Britannica 2008) Therefore, it sets up the input of PLANTS. From this point of view, we have the active party i.e. plant (industrial change), the affected party (We), the type of journey (plunge), the end result (thicket), and the effect (no direction). Therefore, the blend in this case will be the combination of those two input spaces JOURNEY AS PLANT. The participants are Europe, the affected party (We), the type of journey (plunge), the end result (thicket of institutional change) and the effect (no direction).The generic space that has characteristics common to both input spaces, is MOVEMENT. Table 6 summarizes this blend. Generic space MOVEMENT Input space 1 JOURNEY Input space 2 PLANT Elements Active party Elements We- Europe Affected party Type of movement The end result (destination) The effect We political industrial change Elements Plant (institutional change) We plunge thicket no direction no direction Blend JOURNEY AS PLANT Elements We- Europe We plunge Thicket of institutional change no direction Table (6):Tony Blair’s Prospects of Europe blend. The translation was done by the same author as in the previous fragment. C) Problem związany z debatą o przyszłości politycznej Europy polega na tym, że jeżeli nie będziemy uważni, zagłębimy się w gąszczu zmian instytucjonalnych, nie zadając sobie najpierw podstawowego pytania o kierunek, jaki powinna obrać Europa. Back translation: The problem with Europe’s political future is that, if we are not TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 67 careful, we will be immersed in the thicket of institutional change, without asking the basic question of what direction should Europe take. In this case, the translator is a lot more successful in conveying the blend, than in the last example. Like in the SL extract, she sets up the same mental space of JOURNEY by using the word “future”. As such neither do the participants and the actions and effects change. The same is done with “gąszcz” as it once again sets up the PLANT input space. As such there are also no visible changes to its elements or effects. As a result neither the generic space nor the blend has to change its semantic schema or function in order to best sustain the rhetorical effect the author of the SL wished to convey. The last example shows how to successfully substitute one input space for another. The extract is taken from a speech by Joschka Fischer at the Humboldt University in Berlin, 12 may 2000 entitled “From Confederacy to Federation: Thoughts on the finality of European Integration.” The extract reads: 4) Following the collapse of the Soviet empire, the EU had to open up to the east, otherwise the very idea of European integration would have undermined itself and eventually selfdestructed. In this example, we once again have the input space of CREATION AND CONSTRUCTION set up by the word “collapse”. We have the affected party (EU), the initial state (the collapse of the Soviet empire), and the final state (open up to the east). The word ”otherwise” sets up a counterfactual space of DESTRUCTION with the effected party (idea of European integration), the initial state (undermined itself) and the final state (self-destruct). This creates the blend of CREATION and CONSTRUCTION JOURNEYS with the affected party (EU), the initial state (Soviet empire), the final state (open up to the east) and the counterfactual effect (undermine and self-destruct).The generic space in this case is PATH (table 7). Generic space PATH Input space 1 CREATION AND CONSTRUCTION Input space 2 DESTRUCTION Elements Affected party Elements EU The initial state the collapse of the Soviet empire open up to the east Elements idea of European integration undermined itself The final state Counterfactual effect self-destruct Blend CREATION and CONSTRUCTION JOURNEYS Elements EU Soviet empire open up to the east undermine and selfdestruct Table (7):Joschka Fischer’s CREATION and CONSTRUCTION JOURNEY blend The translation was done by Jacek Stanaszek for the daily newspaper “Gazeta Wyborcza”. The translation reads: D) Po rozpadzie imperium radzieckiego Unia musiała otworzyć się na wschód. w przeciwnym razie idea integracji wypaliłaby się i w końcu umarła Back translation: After the collapse of the Soviet Empire, the EU had to open up to the east, otherwise the idea of integration would have burned up and eventually died. 68 RAFAL PACHOLCZYK Here, once again, the word “collapse” sets up the input space of CREATION AND CONSTRUCTION. All the elements stay the same. In the second part of the extract the counterfactual input space of DESTRUCTION is substituted by FIRE by the use of “burned up”. In this instance, the affected party is the idea of integration, the initial state is (lit), the final state (burned up) with the effect (death). The blend stays mostly the same only changing the counterfactual effect to (burned up and died). This short analysis has proved that some of the blends could be translated or substituted successfully while retaining the semantic schema, conceptual event scenario, and the effect on the reader. Some of them, however, were done in an unsuccessful way which doesn’t necessarily have to imply their untranslatability. 6. Summary In this paper my main aim was to establish whether blends could be translated. If so how it could be achieved? And what kind of errors could be committed. For this purpose I have incorporated the theory presented by Mandelblit which suggested an analysis of blended spaces between the SL and the TL structures. The study was conducted using political speeches. One of the conclusions was that if the produced image schema is general enough (GEOGRRAPHICAL STAGE, Tony Blair’s speech) it can be reproduced successfully. However, if there are differing elements that don’t change the effect the translation can still remain successful (the “I have a dream” blend). Another type of translation is creating a simplified blend from a more complex one. This, however, can result in some loss of the SL message (as seen in the 3rd political example). The fifth discussed translation proved that a blend can be substituted for a blend of equal complexity (CREATION AND CONSTRUCTION JOURNEY blend). The analysis was conducted on a limited number of examples. A more thorough research is needed to draw further conclusions. All the same, the few remarks drawn from the research bear a similarity to those produced by Newmark for translating metaphors (outlined in section 1). As I have, discussed in the previous chapter’s metaphors are a more simplex form of blends. As a result some of the same rules can be applied to their translation. For this porpoise I have utilized Newmark’s conclusions for the purpose of my study: 1. 2. 3. 4. 5. 6. 7. 8. Reproducing the same blend in the TL. Replacing the blend in the SL with a standard TL blend which does not clash with the TL culture. Translating blend by simile, retaining the image schema. Translating blend by simile plus sense (or occasionally a blend plus Sense) Converting blend to a more simplex blend (only with standard 4-space model blends) Deletion, if the blend is redundant. Using the same blend combined with sense, in order to enforce the image schema. Complex blends need to retain their SL image schema As I have mentioned above, the analysis can be seen only as an initial one. In order to establish translation rules for conceptual integration process a more thorough examination is another, one which exceeds the scope of this paper. All the same , blends prove to be a valuable source for translators wishing to bather understand the nature of the SL texts and their translator’s implications. References Baker, M. (1992), In Other Words: A coursebook on trnalsation. London: Routledge. Baker, M. (2001), The pragmatics of Cross-Culutral Contact and Some False Dichotomies in Translation Studies. London: Routledge. Bateson, G. (1972), Steps to an Ecology of Mind. New York: Ballantine Books. TRANSLATION AND MEANING, PART 10, 2013, 53 - 70 69 Catford, J. (1965), A Linqustic Theory of Translation. London: Oxford University Press. Cicero. M. [46 BCE] (1960), De invention. De Optimo Genere oratouru. London: Heinemann. Closs, E. and M. L. Pratt (Eds.) (1980), Linguistics for students of literature. New York: Harcourt Brace Jovanovich. Coulson, S. (1996), ”The Menendez Brothers Virus: Analogical Mapping in Blended Spaces”. Conceptual Structure, Discourse, and Language. Palo Alto, CA: 67-81. Coulson, S. (1999), ”Conceptual Blending and Metaphor”. Metaphor in Cognitive Linguistics. Amsterdam & Philadelphia: John Benjamins. Grady, J., Oakley, T.12-18 Coulson, S. (2001), Semantic Leaps: Frame Shifting and Conceptual Blending in Meaning Construction. Cambridge University Press. Coulson, S. & Oakley, T. (2003), “Metonymy and Conceptual Blending”. Metonymy and Pragmatic Inferencing. Amsterdam/Philadelphia: John Benjamins Publishing Company.1ff. Dolet, E. [1540] (1997), How to translate well form one language into another. Manchester: St. Jerome:95-97. Dryden, J. [1680] (1992), Mataphrase, Paraphrase and Imitation. University of Chicago Press. Downing, P. (1977), “On the creation and use of English Compounds”. Language 53. Fauconnier, G. (1985), Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge, Massachusetts London, England: Bradford Books. Fauconnier, G. and M. Turner (1995), ”Conceptual Integration and Formal Expression”. Journal of Metaphor and Symbolic Activity, Vol. 10, number 3;1ff. Fauconnier, G. (2001), ”Conceptual Integration”. Department of Cognitive Science (http://arvo.ifi.uzh.ch/ailab/people/lunga/Conferences/EDEC2/invited/FauconnierGilles.pdf). (date of access: 04 03 2010). Fauconnier, G. and M. Turner (2001), ”Conceptual Integration Networks”. Cognitive Science, 22(2) 1998, 133-187. (date of access: 27 01 2010) (http://markturner.org/cin.web/cin.html). Fillmore, C. (1982), “Frame semantics”. In Linguistic society of Korea. Linguistics in the Morning Calm. Seoul: Hanshin, 11-137. Hatim, B. and I. Mason (1990), Discourse and the Translator. London: Longman. House, J. (1997), Translation Quality Assessment: A model revisited. Tubingen: Gunter Narr. Jakobson, R. (1959), On Linguistic aspects of translation. London: Routledge. Konwicki, T. (1989), Kompleks Polski. Zapis. Lakoff, G. and Johnson, M. (1980), Metaphors we live by. Chicago,IL: University of Chickago Press. Lakoff G. (1987), Women, fire and dangerous things: What cathegories reveal about the mind. Chicago, IL: University of Chickago Press. Lakoff G. (1988), “Cognitive semantics”. In U. Eco. Meaning and mental representations. Bloomington, IN: Indiana University Press. Langacker, R. W. (1991), Concept, image and symbol. The cognitive basis of grammar. Berlin: Mounton de Gruyter. Langacker, R. W. (1987), Foundations of cognitive grammar. Theoretical Prerequisities vol 1. Stanford. Stanford University Press Langacker, R. W. (1995), Viewing in cognition and grammar. Amsterdam/Philadelphia: John Benjamins. Legeżyńska, A. (1999), Tłumacz i jego kompetencje autorskie. Warszawa. PWN. Levi, J. (1978), The Syntax and Semantics of Complex Nominals. New York: Academic Press. Libura, A. (2007), Amalgamty kognitywne w sztuce. Kraków: Universitas. Linke, M. (2008), A cognitive approach to equivalence in literary translation. Grado: Toruń. Luther, M. (2004), An Open letter on Translating. Transl. G. Mann Inc. McElhalon, K. A. (2006), “From simple metaphors to conceptual blending: the mappings of analogical concepts and praxis of translation”. Journal of Translation vol 2. Mandelblit, N. (1997), Grammatical blending: creative and semantic aspects in sentence processing and translation. San Diego: University of California. Newmark, P. (1981), Approaches to translation. Oxford/New York: Pergamon. Newmark, P. (1988), A Textbook of translation. London: Prentice Hall. Newmark, P. (1998), More Paragraphs on Translation. Clevedon, England: Multilingual Matters. Nida, E. and C. Taber (1969), The theory and practice of translation. Leiden: Brill. Nida, E. (1964), Towards a science of translating. Leiden EJ: Brill. Nord, C. (1991), Text Analysis in translation: theory, methodology and didactic application of a model for translation-oriented analysis. Amsterdam: Rodopi. Nord, C. (1997), Transalating as a Purposeful Activity. Manchester: St: Jerome. 70 RAFAL PACHOLCZYK Rummerhalt, D. (1980), „On evaluating story grammars”. Cognitive science (4)3.313-316. Shank, R. (1982), Reminding and memory organization: an introduction to MOPS. M.H. Ridge NJ: Lawrence Erlbaum Associates. Schleiermacher, F. [1813] (1992), On the different methods of Translating. Chicago/London. University of Chicago Press. Shunmugam, K. (2004), The Translation of Metaphors in Malay Pantuns into English. Monash University. Snell-Hornby, M. [1988] (1995), Translation studies and integrated approach. Amsterdam Philadelphia: John Benjamins. Tabakowska, E. (1993), Jezykoznastwo kognitywne a poetyka przekładu. Kraków: Universitas. Talmy, L. (1988), “The relation of grammar to cognition”. W:B.Rudzka-Ostyn red.1988.165-206. Taylor, J. (1992), Linguistic categorisation: Prototypes in linguistic theory. Oxford: Oxford University Press. Toury, G. (1995), Descriptive translation studies and beyond. Amsterdam/Philadelphia: John Benjamins. Tytler, A. [1797] (1997), Essay on the Principles of Translation. Edinbourgh: Cadell and Davies. Venuti, L. (1995), The translator’s invisibility. A history of translation. London/New York: Routledge. Web pages (http://cogweb.ucla.edu/ 1999). (date of access 21.02.2010) (http://www.iclnet.org/pub/resources/text/wittenberg/luther/luther-translate.txt) (date of acess 12.11.2009) (http://www.caffeineheadache.org/) (date of access 12.03.2010) TRANSLATION AND MEANING, PART 10, 2013, 71 - 77 71 LES PARAMÈTRES SOCIOCULTURELS DE LA TRADUCTION DE TEXTES HUMORISTIQUES Magdalena Szeflińska-Baran University of Łódź, Poland Abstract: This paper intends to introduce some extralinguistic considerations to the interdisciplinary field of humour translation. Essential to this task is a brief introduction to the most relevant theory of humour translation: Attardo’s adaptation of the General Theory of Verbal Humour for translation. Attardo has added to this four extralinguistic considerations, which account for factors which lie outside both the source and target texts. L’étude de l’humour verbal dans la perspective traductologique que nous proposons ici s’inscrit dans un champ de réflexion bien plus vaste qu’une approche purement linguistique. L’analyse des problèmes liés à la traduction de l’humour verbal se prête parfaitement à englober dans ses principes méthodologiques les prémisses partant de l’idée que traduire ne saurait se réduire à une pure et simple opération linguistique, nous pensons qu’il convient de resituer la traduction dans le champ d’études interculturelles. Cordonnier (2001: 53) propose une notion d’altérité pour décrire le champ épistémologique de la traduction vue sous l’optique culturelle. La traduction en réalité est d’abord une activité culturelle, une recherche de rapports avec un autre (le destinataire), avec sa culture par l’intermédiaire de la langue. La notion de rapport opposée à celle de transport (transfert) (Meschonnic, 1999: 160-167), toutes les deux ayant pour but de définir le principe opératoire de la traduction, semble mieux caractériser l’approche que l’on doit appliquer à la traduction de l’humour verbal. Sans avoir établi de relations multiples et hétérogènes avec la culture cible, un énoncé ludique traduit uniquement au niveau linguistique ne saurait atteindre son effet humoristique. Le texte humoristique traduit doit jouer des mêmes ressorts que le texte original, c’est-à-dire s’imprégner d’abord de ce qui, en langue-culture cible, (Antoine, 2001:25) donnerait un cadre culturel univoque, équivalent à celui de la langue-culture source. La notion de langue-culture semble primordiale dans l’analyse de la spécificité de la traduction du comique verbal car elle rend aussi bien la dualité de l’ancrage du texte humoristique de départ qu’un double conditionnement de la réception de sa traduction aboutissant au même effet humoristique. Ce terme recouvre, entre autres, la notion de représentations partagées définie par Boyer (2001: 35) comme suit: « J’ai proposé [...] de parler de représentations partagées à propos des mobilisations, généralement implicites, en discours, d’un sens plus ou moins commun, représentations intra- et intercommunautaires à teneur plus ou moins clairement normative et présentant un degré de figement plus ou moins important, qui se manifestent au travers de proverbes, d’évaluations, de catégorisations, d’allusions traditionnelles à la mémoire nationale-identitaire. » Sous le nom de représentations partagées, Boyer distingue deux groupes de phénomènes s’interpénétrant: la première strate à dominante patrimoniale, mythologique, emblématique (les grandes dates, les personnages, les lieux de mémoire), l’autre qui relève pour une large part de la socioculture (les grandes images du vécu communautaire, stéréotypées). Le premier composant patrimonial recueille un consensus maximal qui soude l’identité de la communauté en particulier, face aux autres communautés. Il se caractérise par la stabilité dans le temps par l’opposition au caractère passager, instable et basé sur l’actualité de la strate socioculturelle. Dans celle-ci, on observe des clivages et des manifestations d’identités de groupes qui s’opposent plus ou moins clairement. Les représentations partagées dont on parle ci-dessus font partie d’une compétence culturelle (composante d’une compétence de communication), elles constituent sa composante essentielle qui permet, entre autres, de comprendre et de reconnaître un texte humoristique comme tel en confirmant sa réception par une réaction adéquate: le rire. Une compétence culturelle partagée est nécessaire au 72 MAGDALENA SZEFLIŃSKA-BARAN sein d’une communauté pour que ses membres puissent participer pleinement à un échange humoristique verbal. Lors du passage interculturel qui a lieu pendant un processus de traduction, le manque de cette expérience culturelle partagée constitue un obstacle majeur au transfert du discours humoristique de la langue-culture de départ vers la langue-culture cible. La compétence culturelle fonctionne en discours sur le mode de l’implicite codé de représentations partagées propres à une communauté. Cet imaginaire ethnosocioculturel qui constitue un essor important du comique s’avère le plus souvent intransmissible entre des groupes sociaux, des communautés nationales. Il se crée et se forme dans une relation très étroite et irreproductible dans d’autres conditions entre trois éléments: communauté – langue – culture. La part de représentations partagées qui se reflète dans un système langagier et la manière dont elle est reprise par la langue est unique et typique d’une communauté socioculturelle qui recouvre souvent une communauté linguistique. Dans ce triangle de relations réciproques, il y a lieu la production et la réception du discours comique, ce qui prouve un très fort ancrage socioculturel d’énoncés humoristiques. Sans pourtant nier un caractère universel du rire, en tant que capacité à rire étant une réaction à un stimulus comique, il faut souligner que la question concernant l’objet dont on rit reste beaucoup moins ouverte à des mécanismes généraux. Boyer (2001: 39) souligne que: « L’humour est bien de l’ordre de la mise en scène ludique [...] et ne peut fonctionner que si les destinataires partagent avec le destinateur un imaginaire ethnosocioculturel qui est celui de leur communauté d’appartenance et auquel sont étrangers tous ceux qui n’ont pas été exposés durant leur socialisation à ses manifestations [...]. On peut donc raisonnablement penser que l’humour est un fonctionnement socioculturel intra-communautaire et qu’il est aussi, par là même, un obstacle interculturel (et dont la transposition en langue-culture étrangère s’avère ainsi problématique) [...] » Dans l’optique des observations faites ci-dessus, l’humour est traité comme une pratique de communication profondément immergée dans une ethnosocioculture précise, et de même, il doit faire l’objet d’une approche spécifique en traduction. Laurian (1989: 13) résume les présupposés cognitifs des situations de traduction de l’humour, constituant une composante de première importance de ce que l’on a nommé ci-dessus une ethnosocioculture ou des représentations partagées. Les propositions des items de la liste se présentent comme suit: « les types de connaissances communes au locuteur et à l’auditeur nécessaires pour la compréhension des blagues, plaisanteries, histoires drôles, jeux de mots: − − − − − − − − références précises des mots (et en particulier pour les langues où les références extralinguistiques de l’une sont inexistantes pour l’autre); connotations précises des mots (et en particulier dans le cas où les connotations liées à une référence pour une langue n’ont rien de commun avec celles liées à la référence correspondante de l’autre langue); homonymies, ambiguïtés, doubles sémantismes de chaque langue; perception de ressemblances phoniques; mentalités, comportements, traits psychologiques propres ou donnés pour propres à un groupe linguistique; types de textes, types de styles, types de publications propres à un groupe linguistique; valeurs (morales, religieuses, scientifiques, etc.) qui imprègnent les locuteurs d’une langue; environnement social, politique, économique, d’un groupe linguistique (actualité et histoire). Dans la perspective sociolinguistique, au niveau des représentations collectives, l’humour semble fonctionner comme le langage et sa fonction essentielle est d’ordre identitaire. Gasquet-Cyrus (2002) souligne que l’approche sociolinguistique permet de cerner la complexité de l’humour et de son fonctionnement. Il constate qu’il y a des humours et des représentations humoristiques propres à des communautés humaines, aussi bien à l’échelle nationale que régionale, et que l’humour, dans sa dimension langagière, remplit d’évidentes fonctions identitaires. L’humour varie d’une région à l’autre du pays, d’une nation à l’autre, en fonction des variations historiques, culturelles et langa- TRANSLATION AND MEANING, PART 10, 2013, 71 - 77 73 gières: il ne répond pas partout aux mêmes normes ce qui rend son transfert interlingual avant tout l’affaire de la médiation culturelle et sociale que langagière. Selon Labov (1976: 338) une communauté linguistique est constituée par des locuteurs qui « ont en commun un ensemble d’attitudes sociales envers la langue ». On pourrait de la même manière définir une communauté d’humour qui, pour Gasquet-Cyrus (2002: 258) « serait un ensemble de personnes ayant en commun des normes et des attitudes sociales envers le risible. » Une étude sociolinguistique de l’humour se doit de décrire le contexte général de la communication humoristique: émetteurs et récepteurs, buts recherchés, stratégies discursives, variation sociale, cadre de l’échange verbal. A ce type d’analyse de l’humour, Gasquet-Cyrus (2002) propose d’appliquer le modèle de l’ethnographie de la communication de Hymes qui envisage la communication humoristique aux niveaux micro- (une interaction verbale, une blague) et macro- (un film comique). Dans ces structures de l’humour, ce qui se manifeste comme son trait intrinsèque dans le fonctionnement sociolinguistique ce sont les aspects collectifs de l’humour, ses fonctions identitaires. Dans cette perspective, on peut proposer le concept de rire stéréotypé de David Victoroff (1953) qui s’oppose au rire spontané. A la différence de ce dernier, qui naît selon des processus psychologiques et cognitifs, le rire stéréotypé repose sur les représentations collectives et les normes en circulation dans une société, une culture, une communauté. L’humour représente donc, quant à son fonctionnement au sein d’un groupe, à la fois le caractère exogène, sous forme de moquerie, et le caractère endogène, sous forme d’humour proprement dit. D’où l’ambivalence de l’humour au niveau de la communication sociale: les interlocuteurs se font simultanément cibles et producteurs d’humour. Pour Gasquet-Cyrus (2002), la sociolinguistique permet d’appréhender de façon intéressante les problématiques de l’humour verbal. D’une part, elle prend en compte les situations de contacts linguistiques (bilinguisme, plurilinguisme), elle s’intéresse au phénomène de variation, qu’il s’agisse des variations sociales dans l’utilisation du langage ou, au niveau de l’analyse de discours ou conversationnelle, des registres linguistiques ou stylistiques. D’autre part, la sociolinguistique lie étroitement langue et identité, en s’intéressant aux formes non standard des langues (argots, sociolectes). Tout cela permet d’appréhender l’humour en contexte et de constater que les normes de l’humour varient suivant les communautés géographiques, culturelles et sociales, donc suivant les communautés linguistiques. On peut citer Gasquet-Cyrus qui avance l’hypothèse « que l’humour n’existe pas ; il existe des pratiques comiques ou ludiques (jeux de mots, blagues, sketches, romans) et des représentations de ces pratiques, c’est-à-dire des catégorisations (ironie, humour noir, esprit français). » (2002: 252) L’humour d’un texte est donc autant affaire de contexte, de ton, de structure, de culture que de mots. Le problème lié à la traduction de l’humour paraît encore plus compliqué vue la nécessité d’exprimer celui-ci dans la langue-culture d’arrivée et de le faire fonctionner dans la communauté d’arrivée, ce qui exigerait du texte humoristique traduit de faire naître le même type de rapports sociaux et communicationnels que ceux qui fonctionnaient dans la communauté de départ. L’humour comme un phénomène linguistique et culturel n’est pas homogène du point de vue traductologique car il s’infiltre et existe dans différents types de textes en constituant l’un de leurs éléments constitutifs. Il paraît que la seule forme textuelle qui puisse être entièrement représentative du phénomène humoristique, ce soit l’histoire drôle. L’histoire drôle en tant que forme prélittéraire, incarne tous les traits définitoires de l’humour. Cette constatation semble être vraie sur tous les niveaux du fonctionnement des histoires drôles: sur leur niveau linguistique, culturel et social. Les paramètres traductologiques du transfert de textes humoristiques se situent aussi bien du côté de leur compréhension que de celui de leur reproduction dans la langue-culture cible et même au stade de choix de textes à traduire. Dans sa théorie générale de l’humour verbal, Attardo (2002) énumère quatre facteurs extralinguistiques qui déterminent la traduction de textes humoristiques. Parmi ces facteurs, il y a des limites temporelles, géographiques et sociales liées à l’apparition d’un texte humoristique. Les conditions du transfert de l’humour verbal se présentent comme un éventail de facteurs 74 MAGDALENA SZEFLIŃSKA-BARAN hétérogènes parmi lesquels semblent prévaloir ceux qui appartiennent au domaine socioculturel contre les facteurs purement linguistiques. Dans la phase du choix de textes à traduire, le rôle des paramètres extralinguistiques est primordial. Il paraît qu’uniquement l’humour à portée plus ou moins universelle se prête à la traduction ce qui exclut d’autres types d’humour: par exemple l’humour ethnique soumis aux limites géographiques ou bien l’humour politique déterminé du point de vue temporel par le principe d’actualité. Le manque du fond commun soit géographique soit socioculturel entre des communautés réduit la possibilité du transfert entre elles de textes humoristiques qui garderaient leur caractère et leur fonction. Les jugements parfois opposés, portés sur la traduisibilité de l’humour proviennent du caractère ressenti immédiatement comme doublement constitué de l’humour: caractère linguistique et caractère socioculturel. La combinaison dans des proportions variables de ces deux facteurs rend la traduction plus ou moins aisée. Cette ambivalence est déjà visible dans l’histoire de la notion d’humour. Escarpit (1960: 37) en parlant de l’humour actif et passif, souligne que la différence entre un homme d’humour, qui renvoie ici à une attitude active, et un humoriste, qui représente une posture passive chez Morris (1744: 13), cette distinction se grave dans la définition contemporaine de l’humour. C’est ainsi que l’humoriste, l’homme qui a le sens de l’humour et qui continue à vivre en Angleterre, représente un côté national de l’humour, constitue la partie intraduisible, inexportable du phénomène de l’humour, celle qui restera à jamais inaccessible aux étrangers. « Il est, à proprement parler, le produit du Réflexe anglais [...] » (Escarpit 1960: 37). L’homme d’humour qui renvoie à l’humour esthétique, au contraire, est susceptible d’un langage universel, traduisible et transmissible entre de différentes nations. La notion d’humour qui cristallise au XVIII siècle, en gardant surtout la partie intellectuelle, consciente et esthétique plus proche de l’esprit, perd la vitalité du sens de l’humour. Escarpit (op.cit.) souligne le fait qu’une vocation sociale de l’humour a été découverte par les poètes français, à la charnière du XIXème et XXème siècles, qui voient dans l’humour un moyen d’exorciser les angoisses de l’homme social moderne. Escarpit cite Max Jacob, qui définit l’humour ainsi: «Une étincelle qui voile les émotions, répond sans répondre, ne blesse pas et amuse. »1 Cette définition signifie la découverte du sens social de l’humour, non plus au niveau superficiel des automatismes, des réflexes, comme chez les Anglais, mais de manière créatrice comme une attitude face à la vie. Escarpit conclut en prétendant que l’humour moderne est sociologique ce qui semble trouver sa confirmation dans un rôle assez important joué par l’humour dans la communication d’aujourd’hui. Escarpit (op.cit.: 72) enferme le rôle social de l’humour dans la formulation suivante: « L’humour est l’unique remède qui dénoue les nerfs du monde sans l’endormir, lui donne sa liberté d’esprit sans le rendre fou et mette dans les mains des hommes, sans les écraser, le poids de leur propre destin. » Pour la traduction de l’humour aussi bien que pour d’autres types de traduction, on se pose en général la question sur la notion même de traduction et de ses méthodes et objectifs. Les théories classiques de la traduction nous apprennent ce qu’une traduction devrait faire et comment elle devrait le faire. Au centre de ces approches, il y a le plus souvent la notion de sens. L’une des définitions possibles de la traduction précise qu’un texte écrit en une langue B doit exprimer le même sens qu’un texte écrit en une langue d’origine A ; et pour exprimer le même sens, il faut passer des mots aux idées, de la syntaxe aux concepts, des formulations aux sémantismes pour le réexprimer en languecible. En ce qui concerne les textes humoristiques, tels que: blagues, plaisanteries de conversation, récits ou images sous-titrées, il faut se poser la question de savoir où est le sens et quelle est la signification de sens lorsqu’il s’agit d’humour. Dans la plupart des cas, une explication des mots et 1 Escarpit (1960: 69) cite la définition de l’humour de Max Jacob, prise de : Conseils à un jeune poète, p. 81. TRANSLATION AND MEANING, PART 10, 2013, 71 - 77 75 des référents ne couvre pas l’entière signification. Dans la pragmatique de la communication humoristique, les intentions de l’auteur et les données du texte se mêlent en général à un ensemble de connotations ou de références incluant l’histoire, la politique, les attitudes, les modes de vie, les traditions, la littérature, la science, et toutes sortes de domaines dans lesquels les locuteurs natifs sont immergés en permanence et qui varient d’un pays à l’autre, donc de langue à langue dans la mesure de la solidarité de la géoculture et de la linguistique. L’humour est souvent fondé sur une connivence entre l’auteur et le lecteur, le locuteur et l’auditeur. Ce sont les éléments de cette connivence, de ces connaissances communes et de ces intentions partagées qui constituent le font nécessaire pour une communication humoristique réussie. Il y a aussi des types d’humour où la culture partagée des participants de la communication n’est plus du domaine des données citées ci-dessus (données encyclopédiques ou données expérimentales), mais du domaine du sentiment stylistique. Anne-Marie Laurian (1989: 9) parle du style de l’énonciation scientifique appliqué à des phénomènes prosaïquement quotidiens. Elle en donne comme exemple, les Lois de Murphy: Law of the perversity of nature: You cannot successfully determine beforehand which side of the bread to butter. (Loi de la perversité de la nature: il est impossible de déterminer à l’avance quel côté du pain il faut beurrer.) (Laurian 1989: 9) Ce type d’humour ne pose pas de grands problèmes en traduction à condition de trouver le même type de décalage entre le contenu et le style. Dans cet énoncé, l’humour naît de la généralisation au niveau du contenu et du style de l’énonciation qui caractérise son expression. Lorsque l’humour naît d’un double jeu sur la langue et sur la réalité à laquelle se réfèrent les mots, alors la traduction présente des difficultés parfois difficiles à surmonter. En effet si le présupposé est une connaissance de faits de civilisation, il faut que ces connaissances soient acquises au préalable. La traduction de l’humour exclut une adjonction des explications, des notes de bas de page qui tuent l’effet de surprise et modifient la structure interne d’un texte humoristique et d’un échange comique qui possèdent leur propre dynamique. Cette dernière remarque s’avère particulièrement vraie pour le traitement en traduction des blagues politiques supposant des connaissances préalables. Pour que leur traduction soit fidèle, tout ce fond de connaissances doit être connue du locuteur de la langue-cible. Ainsi, la traduction de l’humour oscille-t-elle entre le transfert appauvri du contenu original, rendant uniquement compte des jeux de langue humoristiques de surface, et la traduction qui fournit aussi des allusions historiques, sociales, culturelles propres à une société donnée. Le même problème concerne les blagues ethniques nationales qui jouent sur des préjugés nationaux, supposés connus de la communauté. C’est ainsi que le schémas stéréotypé humoristique se répète sous forme de versions nationales: les histoires belges françaises sont traduites par des histoires polonaises aux États-Unis et peuvent apparaître comme des histoires irlandaises en Grande-Bretagne ou des blagues russes en Pologne. La théorie interprétative de la traduction (la théorie du sens) souligne également l’importance du savoir partagé pour la compréhension du sens du texte, ce qui constitue aussi une condition sine qua non de la compréhension et de la reproduction du discours humoristique. Dans le processus de construction du sens, le traducteur a connaissance de la situation de communication, qui inclut l’émetteur, le destinataire, le lieu, l’époque et la raison d’être du message. Les éléments non linguistiques qui contribue à la compréhension ne s’arrêtent pas là et, pour que la communication s’établisse vraiment, il faut que les interlocuteurs aient un certain savoir partagé, qui englobe une certaine connaissance commune du sujet du discours ou du texte. La saisie du sens par le lecteur/ traducteur repose donc non seulement sur les significations linguistiques du message, mais aussi sur le recours à son bagage cognitif, antérieur à la réception du texte, à ses souvenirs ou aux associations d’idées suscitées en lui. Cela constitue une parenté claire avec la compréhension de la communication humoristique qui, n’ayant pas de but strictement informationnel et étant une forme indirecte de communi- 76 MAGDALENA SZEFLIŃSKA-BARAN cation, construit son effet humoristique sur les éléments extralinguistiques communs pour une communauté linguistique. Le traducteur ajoute à ses connaissances de la situation et du sujet les éléments qui lui sont apportés par le contexte verbal et cognitif, c’est-à-dire par les composantes formelles et narratives de l’original. Dans la théorie interprétative de la traduction, tous les connaissances préalables et concomitantes à la réception du texte qui entrent en jeu dans la compréhension du sens constituent un ensemble de savoir appelé les compléments cognitifs. Il serait bon de préciser, surtout dans le cas du discours humoristique, que la saisie du sens ne se situe pas uniquement sur le plan notionnel. En effet, la traduction ne vise pas seulement à rendre le contenu du texte ou discours donné, mais aussi son effet émotionnel, indissociable du premier. En effet, le traducteur ne perçoit pas consécutivement d’abord la dénotation, puis les nuances connotatives ou stylistiques. Le sens inclut donc le vouloir dire de l’auteur et, même dans un texte à visée pragmatique, son « comment dire ». Dans la communication humoristique, c’est justement ce « comment dire » qui prévaut sur « ce qui est dit », en imposant une interprétation du contenu dans l’optique humoristique qui le détermine. Seule la restitution combinée de ces deux niveaux peut garantir une bonne traduction. La traduction est une opération mentale dont l’objectif n’est pas de produire des correspondances linguistiques, mais des équivalences textuelles. Cela signifie qu’elle se situe non dans le domaine de l’identité en traduction, mais de l’analogie fonctionnelle et pragmatique. C’est dans cette optique de l’équivalence ou de l’analogie, du « comme », que se situe la traduction de textes humoristiques, ce qui passe inévitablement par l’interprétation et les choix personnels du traducteur. L’humour verbal peut être défini comme une attitude spécifique face à la réalité, exprimée à l’aide de moyens linguistiques et paralinguistiques propres à l’expression du comique au sein d’une communauté, dont la fonction primordiale et celle de régulateur social. Il s’ensuit que l’approche traductologique de l’humour doit obligatoirement tenir compte de son côté pragmatique et fonctionnel sans pourtant nier son aspect linguistique. Bibliographie Antoine, F. (2001), “L’humoriste et le traducteur ou quand la traduction s’en mêle”, in: Laurian A.-M., Szende T. (Eds.), Les mots du rire: comment les traduire? Peter Lang, pp. 19-34. Attardo, S. (1994), “Linguistic Theories of Humor”, in: Victor Raskin, Mahadev Apte (Eds.) Humor Research 1, Mouton de Gruyter, Berlin-New York. Attardo, S, (1997), “The semantic foundations of cognitive theories of humor”, in: Humor, 10: 4, p. 395-420. Attardo, S. & Raskin, V. (1991), “Script theory revis(it)ed : Joke similarity and Joke representation model”, in: Humor, 4, 3/4, pp. 293-347. Attardo, S. (2000), Humorous Texts, Berlin, Mouton De Gruyter. Attardo, S. (2002), “L’analyse des textes humoristiques”, in: Mongi Madini 2000 ans de Rire. Permanence et Modernité. Coll. Linguistique et sémiotique 42, Presses Universitaires Franc-Comtoises, pp. 273-280. Attardo, S. (2002), “Translation and Humour. An Approach Based on the General Theory of Verbal Humour (GTVH)”, in: The Translator 8 (2), p. 171-192. Boyer, H. (2001), “L’humour comme connivence intraculturelle et comme obstacle interculturel”, in: Laurian A.-M., Szende T. (Eds.) Les mots du rire: comment les traduire?, Peter Lang, pp. 35-44. Cordonnier, J.-L. (1995), Traduction et culture, Credif-Hatier/Didier, Paris. Cordonnier J.-L. (2001), “Traduire le discours: Colon et les femmes dans « La harpe et l’ombre » de Alejo Carpentier”, in: Laurian A.-M., Szende T. (Eds.), Les mots du rire: comment les traduire?, Peter Lang, pp. 53-72. Cuellar Irala, J. & Garcia-Falces Fernandez, A. (2004), “Cultura y humor: traductores al borde de un ataque de nervios”, in: Linguax. Revista de Lenguas Aplicadas, Madrid, Universidad Alfonso X el Sabio. Dupreel, E. (1928), “Le problème sociologique du rire”, in: Revue Philosophique de la France et de l’Etranger, CVI, pp. 212-260, in : « Essais pluralistes », Paris, PUF, 1949, pp. 26-29. Escarpit, R. 1960 (1967,1994), L’Humour, Coll. Que sais-je?, no 877, P.U.F., 4e édition. Gasquet-Cyrus, M. (1998), Pour une étude sociolinguistique du comique à Marseille, Mémoire de DEA de Linguistique, sous la direction de L.-J. Calvet, Université de Provence. TRANSLATION AND MEANING, PART 10, 2013, 71 - 77 77 Gasquet-Cyrus, M. (2002), “Pour une étude sociolinguistique de l’humour: l’humour marseillais”, in: Mongi Madini 2000 ans de Rire. Permanence et Modernité, Coll. « Linguistique et sémiotique » 42, PU FrancComtoises, Besançon, pp. 251-259. Labov W. (1976), Sociolinguistique, Paris, Minuit ; trad. Française de Sociolinguistics Patterns. Laurian, A.-M. & Szende, T. (Eds.) 2001, Les mots du rire: comment les traduire ? Essais de lexicologie contrastive, Peter Lang, Berne. Laurian A.-M. (1989), “Humour et traduction au contact des cultures”, in: Meta: journal des traducteurs, vol. 34, n 1, pp. 5-14. Meschonnic H, (1999), Poétique du traduire, Lagrasse, Ed. Verdier. Morin, V. (1966), “L’histoire drôle”, in: Communications, 8, pp. 102-119. Morris, C. (1744), An Essay towards fixing the True Standards of ... Humour, Londres, p. 13, (cité par Escarpit 1960 : 36). Raskin, V. (1985), Semantic Mechanisms of Humor, D. Reidel Publishing Compagny, Dordrecht, Holland. Victoroff, D. (1953), Le rire et le risible. Introduction à la psycho-sociologie du rire, Paris, PUF. 78 TRANSLATION AND MEANING, PART 10, 2013, 79 - 90 79 DOLORES DE LA TRADUCCION, OR LOLITA IN TRANSLATION Janusz Wróblewski University of Łódź, Poland Abstract: In 1711 Joseph Addison called puns “false wit” and suggested that in order to test a piece of wit, one should simply translate it into another language: “If it bears the Test, you may pronounce it true; but if it vanishes in the Experiment, you may conclude it to have been a Punn” (The Spectator, No. 61, Thursday, May 10, 1711). In other words, according to Addison, (1) puns are to be frowned upon, and (2) puns are untranslatable. The first point is a matter of taste and, as such, cannot be disputed. The second point, however, is a matter of facts, and the facts – at least today – are different. It is true that wordplay can be difficult to translate, and that some puns may actually prove to be impossible to translate into some languages in a specific manner, but it is also true that the same puns can turn out to be perfectly translatable into some other languages or in a different manner. The present paper will try to show how various types of wordplay can be translated on the basis of the novel Lolita by Vladimir Nabokov. 1. Introduction Lolita is one of the most interesting books for analyzing wordplay in translation because, originally written – and first published – in English, it then got translated by the author himself into Russian, which enables us to learn how to translate puns from the author. Moreover, this novel exists in two separate translations into Polish, namely by Robert Stiller (based on both the original English version and Nabokov’s Russian Lolita) and by Michał Kłobukowski (based on the English text only) – both of these translators being well-known and distinguished – so that we can compare the procedures for rendering wordplay used by Nabokov with those used by the two Polish translators. Following Lewandowska-Tomaszczyk, translation is viewed here as “re-conceptualization of a message in the totality of its context and situation” (2010: 108), and while in the translation of ordinary texts, this re-conceptualization is governed partly by the target language conventions – the linguistic (morpho-syntactic and phonological) structure of the TL (as opposed to the source language conventions), partly by the context of the TL text/message – new time, new place, new audience (as opposed to the context of the SL text/message), and partly by the translator's individual preferences and random choices, re-conceptualization in the translation of wordplay will involve one additional factor – an attempt to recreate the various instances of wordplay. Wordplay can be defined very generally as “A figure of speech depending upon a similarity of sound and a disparity of meaning” (Fogle, 1974:681). A preciser definition is offered by Delabastita: “wordplay is the general name indicating the various textual phenomena (i.e. on the level of performance or parole) in which certain features inherent in the structure of the language used (level of competence or langue) are exploited in such a way as to establish a communicatively significant, (near)simultaneous confrontation of at least two linguistic structures with more or less dissimilar meanings (signifieds) and more or less similar forms (signifiers)”. (1993: 57, but cf. also 2001: 48) This confrontation can assume many forms: either one word (or structure) is placed in such a context that two (or more) separate, individual meanings are activated at once or almost at once (e.g., when Pratchett in Night Watch talks about “the Bridge of Size” [2003:40], he is obviously playing on the better-known Bridge of Sighs, or Ponte dei Sospiri in Venice, and confronting Size with its homophone Sighs), or the given word is used twice, each time in a different sense (“If we don't hang together, we'll hang separately” – Benjamin Franklin), or two different words or structures (occasionally more) which look or sound similar are placed next to each other (as in the famous Italian 80 JANUSZ WRÓBLEWSKI maxim, traduttore traditore). Wordplay includes also playing with parts of words (for example, mauvemail is offered as a milder form of blackmail – Nabokov 1980: 70–71, while therapist can become the rapist – Nabokov 1980: 147–148), and many other ways of playing with language in general. In terms of classification, the neatest one has been proposed again by Delabastita. He divides puns into four major groups, namely those based on homonymy, homophony, paronymy, and homography; moreover, each of these types can be vertical and horizontal, depending on whether the two (or more) meanings are activated simultaneously or successively, which gives us eight categories altogether (1993: 78-81). There are of course certain problems with Delabastita’s classification: for example, it is not at all clear whether homophones include near-homophones or not (specifically, should the following pun: “The dairymaid is a girl who ought to know butter”, echoing the wellknown expression to know better, meaning 'to be wise enough not to do something' [Maltzev,1980:15], be classified as a vertical homophonic pun or perhaps as a paronymic one?); moreover, some puns will not fit any of Delabastita’s categories easily (e.g., the above-quoted mauvemail), but other than that, this classification provides quite a useful framework for a discussion of the procedures for translating wordplay. Let us therefore look now at some of the cleverer instances of wordplay from Lolita,1 following the classification of puns proposed by Delabastita, in the original and in the three translations: one Russian and two into Polish. Reference will be made here also to Delabastita’s nine procedures, methods or techniques2 (some of them with numerous subdivisions) for dealing with wordplay; they are: (1) (2) (3) (4) (5) (6) (7) (8) (9) PUN > PUN; PUN > NON-PUN; PUN > PUNOID; PUN > ZERO; Direct copy: PUN S.T. = PUN T.T.; Transference: PUN S.T. = PUN T.T.; Addition: NON-PUN > PUN; Addition (New textual material): ZERO > PUN; Editorial Techniques (Footnotes, etc.) (Delabastita 1993: vii and 192-227). (Interestingly enough, the term transference, which roughly involves imposing the SL meaning onto the TL word, does not reappear on the lists in Delabastita 1996: 134 and 2004: 604; moreover, in Delabastita 2004: 604, the term PUNOID gets replaced with RELATED RHETORICAL DEVICE). 2. Vertical homonymic puns This is the ideal, prototypical play on words proper, where one word is put in a context in which both its meanings are activated at once, as in Mercutio’s last and pain-full pun “Ask for me tomorrow and you shall find me a grave man” from Romeo and Juliet (Act III, Scene i; Shakespeare 1958: 492) – grave meaning 'serious' but also evoking the image of a grave, or as in the following extended pun from Lolita, when Humbert agrees to let Lolita appear in the school play: “She can take part in that play. Provided male parts are taken by female parts” (Nabokov 1980:194). The ambiguity is both in the word parts ('roles' and 'private parts, i.e., sexual organs') and in the verb to take, which covers both playing a role and physical taking. Although homonymic pairs in one language seldom correspond to similar pairs in another language, this pun is actually not too difficult to translate as a homonymic pun, that is, the parts part will not work in Russian or Polish, but the general idea will. The Russian version (“Она может участвовать в пьесе. Но если часть ролей, мужская, то ставлю условие: мужская часть поручается 1 A comprehensive study of all the puns form this novel would be beyond the scope of a single paper. 2 Delabastita (1993) uses those terms interchangeably. TRANSLATION AND MEANING, PART 10, 2013, 79 - 90 81 девочкам”3) says more or less that if part of the roles is for men, then that male part will be entrusted to or left in the care of the girls. Stiller reproduces the ambiguity of take, adding the equally ambiguous prepositional phrase onto themselves, and conjuring up images of the girls working under the boys: “Pozwalam jej zagrać w tej sztuce. Pod warunkiem, że w jej obsadzie wykonawców męskich wezmą na siebie dziewczęta” (Nabokov 1991: 219). Kłobukowski’s version is rather odd. It offers a kind of homonymic pun based on the verb rolować, but it forces the reader to assign to this verb a meaning which it does not have. Rolować is a slangy word meaning 'to cheat', but in this context it has to mean also 'to play a role', which suggests that Humbert has an imperfect command of the language: “Pozwolę jej wystąpić w tej sztuce. Pod warunkiem, że postaci męskie będą rolować dziewczynki” (Nabokov 1997: 238). One of the meanings of this construction would then be 'male roles will be played by (the) girls' and the other one 'men will cheat the girls' (unless we are meant to interpret the verb rolować as having a sexual meaning, perhaps as a calque of to roll). One more possible vertical homonymic pun – the crushed kids pun – will be discussed in the next section, and the novel contains also several borderline cases. For example, the following pun could be classified as homonymic (it is definitely vertical), but it could be argued that it is homophonic because of the capital letter required by the second meaning: “The noncommittal mauve mountains half encircling the town seemed to swarm with panting, scrambling, laughing, panting Lolitas who dissolved in their haze.” (Nabokov 1980: 222; emphasis mine – JW). The primary sense of the word haze is the state “when the air is not very clear because of something such as heat or smoke, making it difficult to see well” (CALD 2003), but, since the main character is called Dolores Haze, the word was probably meant to activate the associations with the surname as well (the book contains a few related horizontal puns – see below). Since the Russian and Polish equivalents of haze sound completely different from the English word and since the surname Haze is semantically empty in the two languages, recreating this pun is virtually impossible; consequently, neither Nabokov’s Russian version nor the two Polish translations make any attempt at reproducing any pun here and simply render the basic meaning of haze (see Nabokov 1991: 249 and 1997: 271). One more vertical homonymic pun (based on the word lo) will be discussed in the next section instead of here, because it forms a kind of series with related horizontal puns, which appear earlier in the book. 3. Horizontal homonymic puns Homonyms are repeated, occasionally with a slight grammatical modification (e.g., singular vs. plural, etc.). The first pun of this type appears already in the first paragraph of the Foreword: “Mr Clark’s decision may have been influenced by the fact that the editor of his choice had just been awarded the Poling Prize for a modest work (‘Do the Senses make Sense?’)” (Nabokov 1980:5). Horizontal puns are generally much easier to translate than vertical ones, the idea being to build a different pun (not necessarily homonymic) on one of the meanings (the more important one). The same strategy was followed here by Nabokov and the two Polish translators; all the three versions have horizontal paronymic puns based on the word zmysły ('senses'): “Можно ли сочувствовать чувствам?” [literally: 'Can one sympathize with the senses?'] „Co myśleć o zmysłach?” ['What to think of the senses?'] (Nabokov 1991: 5) „Czy zmysły są zmyślne?” ['Are the senses clever?'] (Nabokov 1997: 5) Another horizontal homonymic pun, however, apparently proved more difficult to render into Russian: “pumps of crushed kid for crushed kids” (Nabokov 1980: 107). The first kid is obviously 3 This and all the remaining quotations from Nabokov's Russian Lolita come from the text downloaded from http://lib.ru/NABOKOW/lolita.txt [accessed 9 July 2009] and hence no page numbers are available. 82 JANUSZ WRÓBLEWSKI “very soft leather made from the skin of a young goat” (CALD 2003) while crushed means 'pressed' or 'treated in such a way that it has a permanently crinkled, wrinkled or rumpled appearance'4; the second kid refers to a child or a young person and crushed seems to be related to the expression to have a crush on someone and means having “a strong but temporary attraction for someone” (CALD 2003), although it could be an additional vertical homonymic pun based on 'pressed'. The Russian version “туфельки из мятой лайки для мятых девочек” repeats the adjective мятый, which means 'creased' or 'crumpled' but which has no connection with having a crush on someone, and the two kids are rendered independently as leather, goatskin, or chevrette, and girls, respectively. Stiller renders crushed as “soft” in both cases, but manages to reproduce a fairly effective semi-homonymic pun based on the fact that in Polish the words koza ('a female goat') and kózka (diminutive of the above, 'a young female goat') are often used to refer to young girls: “czółenka z mięciutkiego koźlęcia dla mojej mięciutkiej kózki” (Nabokov 1991: 120), whereas Kłobukowski renders the basic meaning of the first occurrence of crushed kid and produces a horizontal paronymic pun on the second: “pantofelki z marszczonego zamszu dla zmuszanych burczymuszek” (Nabokov 1997: 130), phonetically not too far away from the word zamsz ('suede' or 'chamois leather'). Nabokov was far more successful in rendering his next horizontal homonymic pun: “Dr Hummer, do you realize that for the modern pre-adolescent child, medieval dates are of less vital value than week-end ones (twinkle)? – to repeat a pun that I heard the Beardsley psycho-analyst permit hereslf the other day.” (Nabokov 1980: 176; emphasis in this one and in all the remaining quotations mine – JW) Instead of the word dates, he uses the word поход, which means both 'a (military) campaign' and 'going out', and produces a perfect homonymic pun which retains one of the senses of the original: “Доктор Гуммер, отдаете ли вы себе отчет в том, что для современного подростка какой-нибудь средневековый поход представляет меньше жизненной ценности, чем поход (она чуть не подмигнула) в кафетерию с молодым человеком?” Stiller renders the source text wordplay by means of a semi-homonymic pun which plays on the word średni in the senses of 'the Middle (Ages)' and 'middling' or 'mediocre': “Czy uświadamia pan sobie, doktorze Hummer, że dla współczesnego dziecka w okresie dojrzewania całe średniowiecze jest mniej istotne niż (mrugnęła znacząco) średnia randka?” (Nabokov 1991: 197). Kłobukowski manages to offer another homonymic pun, based on the double meaning of period: “Doktorze Hummer, może nie zdaje pan sobie sprawy, ale dla wspólczesnej dziewczynki u progu pokwitania znajomość okresów historycznych jest mniej istotna niż jej własny okres [perskie oko]?” (Nabokov 1997: 214). In the previous section we looked at a vertical pun based on the word haze. Let us now analyze a related horizontal pun: “I lost myself in the pungent but healthy heat which like summer haze hung about little Haze” (Nabokov 1980: 59). Since it is impossible in Russian and Polish to build a homonymic or homophonic pun based on the name Haze, Nabokov decided to render this instance of wordplay by means of a neologism, a kind of vertical paronymic conflation of the name Долли with the diminutive of the noun долина ('a valley'): “Я терялся в едком, но здоровом зное, который как летнее марево обвивал Доллиньку Гейз”. Paszek and Skwarek (1992: 80) suggest that this could be copied directly into Polish: “Zatracałem się w ostrym, ale zdrowym upale, który jak letnia mgła 4 Cf. http://www.thefreedictionary.com/crushed [accessed 23 August 2010]. TRANSLATION AND MEANING, PART 10, 2013, 79 - 90 83 otulał Dollinkę Haze”, but Stiller came up with a very clever pun based on the semantic contrast between the adjective letni ('summer' as an adjective) and its negation nieletni, which means 'a minor, a person under age': “Zatracałem się w rozognionej, ale zdrowej ostrości unoszącej się jak letni, drgający skwar wokół nieletniej Lo Haze” (Nabokov 1991: 66). Kłobukowski is much less sucessful here: he replaces “little Haze” with “little misty [or hazy] Haze”, which lets him produce a pun, but this pun sounds rather forced: “Zatraciłem się w cierpkim lecz zdrowym skwarze, który letnią mgłą spowijał małą mglistą Haze” (Nabokov 1997: 71). On the other hand, it must be remembered that he could not borrow from Stiller, that he had to invent something new, something different. (In one article, Stiller actually accused Kłobukowski of plagiarizing from his translation; he said that Kłobukowski had not translated Lolita from English but only rewrote his – Stiller’s – translation, using various synonyms; see Stiller 19975). Let us consider the following homonymic pun based on two meanings of the word Lo: (1) the shortened form of the name Lolita, and (2) an archaic form of the verb look. When Lolita wants to accompany Humbert and her mother driving shopping, she jumps into the car and says: “Move your bottom, you,” said Lo. “Lo!” cried Haze (sideglancing at me, hoping I would throw rude Lo out). “And behold,” said Lo (not for the first time, as she jerked back, as the car leapt forward. (Nabokov 1980: 50). This will obviously not work outside English, and hence Nabokov decided to render it as a punoid – in this case a rhyming couplet; in Russian, instead of the phrase “And behold”, Lolita says simply “Ло-барахло”, which means roughly 'Lo good-for-nothing'. Stiller uses the same method (as well as quoting and explaining the original wordplay in an endnote): “ – Ty, rusz swój tyłek! – rzekła do mnie Lo. – Lo! – krzyknęła Haze (spojrzawszy na mnie z ukosa, w nadziei, że ja wyrzucę tę niegrzecznie odzywającą się Lo). – I pstro! – odpowiedziała jej (nie po raz pierwszy) Lo …” (Nabokov 1991: 56 + 358). Kłobukowski modifies the word dola (someone's 'lot or fate'), conflating it with the name Dolly, and offers a kind of vertical paronymic pun: „Rusz tyłek”, powiedziała do mnie. „Dolly!”, krzyknęła Haze (zerakając na mnie z ukosa, w nadziei, że wyrzucę pyskatą Lo). „O, dollo nieszczęsna!”, odparła Lo (nie po raz pierwszy) …” (Nabokov 1997: 60) That was a horizontal pun, as the two meanings are activated successively, but several pages later (at the beginning of Chapter 14) we encounter a possible modified vertical version of the same pun (like the haze pun, perhaps this one too should be classified as a homophonic pun, because of the capital letter required by the second meaning): “I felt proud of myself. I had stolen the honey of a spasm without impairing the morals of a minor. Absolutely no harm done. The conjurer had poured milk, molasses, foaming champagne into a young lady’s new white purse; and lo, the purse was intact.” (Nabokov 1980: 61) 5 Personally, I do not believe Stiller's claim to be fully justified – his translation contains a few obvious mistakes, where Kłobukowski's version is correct; vide, for example, the case of the nicknames of the classrooms at Beardsley, discussed towards the end of this paper. 84 JANUSZ WRÓBLEWSKI On the one hand, this is not wordplay at all because technically the word lo has only one meaning in this context (and Nabokov’s Russian text has a non-pun rendering: “[...] раз, два, три и сумка осталась неповрежденной” – literally: “one, two, three, and the purse was intact”6), but on the other, we cannot avoid associations with Lolita (especially in view of the previous occurrence of this, which Kłobukowski marks by repeating his coinage dollo: “Iluzjonista wlał mleko, melasę i spieniony szampan do nowej białej torebki młodej damy, a torebka – o, dollo szczęsna! – pozostała nietknięta.” (Nabokov 1997: 74) Interestingly enough, this wordplay reappears with slight modifications two more times. The first: “I had only to turn away for a moment – to walk, say, a few steps in order to see if our cabin was at last ready after the morning change of linen – and Lo and Behold, upon returning, I would find the former, les yeux perdus, dipping and kicking her long-toed feet in the water on the stone edge of which she lolled …” (Nabokov 1980: 159) openly relating the archaic expression to Lolita’s name (“the former”), gets a non-pun rendering in all the three versions, but the second one is very nicely reproduced by Kłobukowski. At one point Humbert observes: “I looked up from the letter and was about to – There was no Lo to behold.” (Nabokov 1980: 221–222). Neither Nabokov’s Russian version nor Stiller’s text has any pun here (“Podniosłem oczy znad listu i już miałem – co to? Nie ma Lolity.” to quote from the latter – Nabokov 1991: 249), but Kłobukowski managed to find an interesting phonetic parallelism for Lo: “Oderwałem się od listu i już miałem … lecz nie zdołałem zlokalizować Lo.” (Nabokov 1997: 270) 4. Vertical homophonic puns When spoken, vertical homophonic puns are no different from homonymic puns: one word or phrase activates more than one meaning simultaneously. For example, we hear a phrase [mai kju:], and in our mind’s eye we see 'my cue', 'my queue', and perhaps also 'my Q'. Of course, when we see this phrase written, one of the meanings is foregrounded, so the spelling spoils the effect to a certain extent (although it can also help to make the punning effect possible, as in “the Bridge of Size”, mentioned above), but in the proper circumstances the multiple activation is still there. Therefore, when we read in Lolita: “‘Vivian Darkbloom’ has written a biography, ‘My Cue’, to be published shortly, and critics who have perused the manuscript call it her best book” (Nabokov 1980:6), we know that apart from suggesting the basic “visible” meaning (“a word or action in a play or film, which is used as a signal by a performer to begin saying or doing something” – CALD 2003), the word Cue, or rather the sound [kju:], refers also to the nickname and the initial of the surname of the character named Clare Quilty (it is his biography). Should anybody have any doubts that it is a pun, the Russian version leaves no room for doubt: “Г-жа Вивиан Дамор-Блок (Дамор – по сцене. Блок – по одному из первых мужей) написала биографию бывшего товарища под каламбурным заглавием “Кумир мой”, которая скоро должна выйти в свет; критики, уже ознакомившиеся с манускриптом, говорят, что это лучшая ее вещь.” 6 So has Stiller, who replaces “and lo” with “voilà!” (Nabokov 1991: 69). TRANSLATION AND MEANING, PART 10, 2013, 79 - 90 85 (The names Vivian Darkbloom and Vivian Damor-Blok are anagrams of Vladimir Nabokov.) The highlighted words mean 'under the punning title'. “Кумир мой”, as it is spelled, means 'my idol', but it can also be interpreted as “Ку – мир мой” ('Cue / Q – my world'). Robert Stiller has a non-pun rendering: "Vivian Darkbloom" napisała swą autobiografię [sic!], Mój klucz, która ma się wkrótce ukazać, a krytycy, którzy zapoznali się z tym rękopisem, uważają go za jej najlepszą książkę. (Nabokov 1991: 6) and Kłobukowski’s version is not much better here: „Vivian Darkbloom” napisała biografię zatytułowaną „Mój Kuku” (Nabokov 1997: 6). Inspired by the Russian version, Paszek and Skwarek offer a few interesting punning solutions in their article on “Puns in Lolita”: e.g., Kupalnik mój, Kuracja moja, Pochwały Kupały (1992: 78), of which the last one is probably the best (The Praises of / for Kupala – an old Slavonic deity of sex, fertility and midsummer – as well as for Cue’s penis, on the basis of a homophonic reinterpretation), except that Kupała is so strongly rooted in Slavonic tradition that some readers might find it odd to see that name in an English-based novel. My own suggestion for the title of the biography written by Vivian Darkbloom would be purely homonymic “Pyta Q”, which can be interpreted to mean both 'Q asks' and 'Q’s penis'. Sometimes a pun will be based on near-homophones; Lolita being officially called Dolores Haze, at one point the narrator plays on the sound of her name in this manner: “[…] I can imagine so well the rest of the colourful classroom around my dolorous and hazy darling” (Nabokov 1980: 52). Obviously, translating the meaning is bound to deprive the words of any phonetic similarity to the name Dolores Haze, so Nabokov produces a calque of the adjective dolorous, replaces hazy with hazilyrosy, and thus manages to retain a pun: “А кроме того, я могу так ясно представить себе остальную часть этого красочного класса вокруг моей дымчато-розовой, долорозовой голубки.” Stiller’s version is interesting because it is based on what might be termed interlingual similes, which taste of horizontal wordplay if one knows what Dolores means in Spanish and Haze in English: “Czy dlatego, że tak łatwo mi wyobrazić sobie resztę tej barwnej klasy wokół mej dotkliwej jak Dolores, mglistej jak Haze, mojej najdroższej” (Nabokov 1991: 59). Kłobukowski on the other hand seems to have chosen a non-pun rendering: “A może dlatego, że tak dokładnie staje mi przed oczami reszta barwnej klasy wokół mej boleściwej mglistej ukochanej” (Nabokov 1997: 62), but of course readers familiar with the semantics of the name Dolores Haze will understand the significance of the two adjectives, so that the two words do form some kind of double vertical pun. One of the best vertical puns in Lolita is also based on near-homophones; Quilty begs Humbert not to kill him and offers him some money: “[…] I have not much at the bank right now but I propose to borrow – you know, as the Bard said, with that cold in his head, to borrow and to borrow and to borrow.” (Nabokov 1980: 300) This is Macbeth’s “Tomorrow, and tomorrow, and tomorrow” (Act V, Scene v; Shakespeare 1958: 1216) said with his nose blocked by a cold, and while this wordplay is perfectly translatable into German, where the opposition morgen vs. borgen mirrors exactly that of English, there is no way this can be preserved in Russian or Polish. Therefore translating his Lolita into Russian, Nabokov used a different rhetorical device and replaced the original joke with an allusion to Eugene [Yevgeny] Onegin: 86 JANUSZ WRÓBLEWSKI “Обещаю вам, Брюстер, что вы заживете здесь счастливо, пользуясь великолепным погребом и всем доходом с моей следующей пьесы, - у меня сейчас маловато в банке, но ничего, буду жить долгами, как жил его отец, по словам поэта.” (Nabokov 1989). The highlighted words refer to the beginning of Stanza III of the First Chapter: Служив отлично-благородно, Долгами жил его отец, ['Having served his country faithfully His father lived a life of debt.']7 Stiller’s strategy was different here: he translated the text literally: “ – w tej chwili akurat nie mam dużo na koncie, ale będę pożyczał – pan wie, jak to było u Wieszcza, kiedy miał katar: pożyczać i pożyczać i pożyczać.” (Nabokov 1991: 338) and resorted to editorial techniques, explaining the original English wordplay in an endnote, as well as mentioning briefly, but not explaining, the allusion to Pushkin (see Nabokov 1991:402). Michał Kłobukowski translated the pun, but the result – based on the disputable phonetic similarity between jutro and futro – is rather dubious (on the other hand, it will be observed that he enhanced the effect by using a related rhetorical device – rhyme): “[…] akurat w tej chwili niewiele mam w banku, ale gotów jestem dać w zastaw ostatnie futro, jak Makbet, gdy wśród cieni znienacka zaseplenił: futro, a po futrze znów futro i futro.” (Nabokov 1997: 365) My own suggestion would be to retain the wordplay in the original as a travestied quotation from Macbeth, explain it in a footnote, and perhaps add – by way of compensation – the following horizontal homonymic pun, based on the homonymy of long and debts (new textual material): “Swoją drogą, zawsze wiedziałem, że kiedyś mój długi wpędzi mnie w długi.” 5. Horizontal homophonic puns Horizontal homophonic puns do not occur in Lolita (unless I missed them), except for the borderline Haze/haze case, which was discussed in the section on homonymic puns. 6. Vertical paronymic puns Paronyms are words which look similar, but which have different meanings. As an example of a vertical paronymic pun Delabastita (1993: 81) quotes the word intrinsicate from Antony and Cleopatra (this word is a conflation of intricate and intrinsic). To the best of my knowledge, Lolita does not contain any vertical paronymic puns, but it does have numerous horizontal paronymic puns. 7 This comes from the following website: http://www.pushkins-poems.com/Yev002.htm [accessed 9 July 2009]. TRANSLATION AND MEANING, PART 10, 2013, 79 - 90 87 7. Horizontal paronymic puns Making a horizontal paronymic pun consists in putting two similar-looking words close to each other. It is probably the easiest type of pun to make, and therefore many people consider it an inferior form of wordplay, but still, horizontal paronymic puns are widely used in literature, including Lolita. For example, to most people, women are women, but Humbert in Lolita distinguishes two female sexes, and states: “But to me, through the prism of my senses, ‘they were as different as mist and mast’.” (Nabokov 1980:18). As could be expected, this is relatively easy to translate; the Russian text has a horizontal paronymic pun here which, moreover, retains one of the original meanings: “они были столь же различны между собой, как мечта и мачта” (literally, “as a dream and mast”). The general principle of translating paronymic puns seems to be then to translate both key words semantically, and then see which one is easier to produce a nice paronym on. In actual fact, neither Stiller nor Kłobukowski retains the “mist and mast”, but both manage to remain within the sphere of nautical terminology with one word and juxtapose it with a semantically significant other word. Stiller’s version is: “Ale dla mnie, przez pryzmat moich zmysłów, „różniły się jak raj od rej” (Nabokov 1991: 20). Back-translated, this means “as Paradise and yards”, the word yard being used here in the sense of a spar, i.e. “a strong pole, especially one used as a mast to hold the sail on a ship” (CALD 2003), while Kłobukowski has: “Lecz dla mnie, widziane przez pryzmat zmysłów, „różniły się niczym żagiel i żagiew”. (Nabokov 1997: 20). Żagiel means 'a sail', and żagiew (literally, 'a piece of burning wood') is the word which Kłobukowski used in the first sentence of Chapter 1 to render – perhaps unfortunately – the first word of the phrase “fire of my loins”). One paronymic pun from Lolita is actually self-translatable: “She had a Nansen, or better say Nonsense, passport …” (Nabokov 1980: 27). Since Nansen is a proper name and nonsense is an “international” word, both Stiller and Kłobukowski render this pun literally without any problems: “Posługiwała się paszportem nansenowskim (słuszniej można by o nim powiedzieć nonsensownym) …” (Nabokov 1991: 29); “Paszport nansenowski, a raczej nonsensowny …” (Nabokov 1997:31). Oddly enough, the Russian Lolita does not reproduce this wordplay, although, as Anna Ginter states, Russian does have the word nonsense (2003: 223). 8. Vertical homographic puns By definition, homographic puns involve the use of words which have the same spelling but different pronunciations. Interestingly enough, as an example of a vertical homographic pun, Delabastita quotes the word The-rapist (1993: 81), which Nabokov uses “horizontally”, so to speak (to my knowledge, Lolita does not have any vertical homographic puns). 9. Horizontal homographic puns In Chapter 1 of Part Two of Lolita Humbert observes: “The rapist was Charlie Holmes; I am the therapist – a matter of nice spacing in the way of distinction.” (Nabokov 1980: 147–8). As such, this looks completely untranslatable, because it is impossible to preserve the meaning and the wordplay. Consequently, in the Russian version Nabokov substitutes a horizontal paronymic pun for the original wordplay: “Растлением занимался Чарли Хольмс; я же занимаюсь растением, детским растением …” (roughly, the contrast is between the seduction of a minor and looking after a child's growth). The same procedure is applied by Stiller, who, moreover, provides an endnote which quotes this wordplay in the original; his version reads: “Twoim gwałcicielem był Charlie Holmes; a ja jestem terapeutą. Jeden znieprawia, a drugi naprawia: zwróć uwagę na to subtelne rozróżnienie.” [“One demoralizes, the other one repairs …”] (Nabokov 1991: 165 + 371). 88 JANUSZ WRÓBLEWSKI Kłobukowski simply repeats Stiller’s solution (see Nabokov 1997: 179). 10. Miscellaneous Leaving Delabastita’s neat classification aside and moving into less classifiable areas, we might mention first of all the following punning coinage: “Then, figuratively speaking, I shattered the glass, and boldly imagined […] how eventually I might blackmail – no, that is too strong a word – mauvemail big Haze into letting me consort with little Haze …” (Nabokov 1980: 70-71). Playing on the etymology of the word blackmail, Nabokov replaces black with mauve, thus giving this word a somewhat “gentler” character. Since the equivalents of the word blackmail in Russian and Polish do not contain the element of colour, it is obviously impossible to reproduce the wordplay based on colours, but, fortunately, a similar effect of “softening” the word meaning 'blackmail' can be achieved in the two languages by using diminutives. The Russian version is not terribly inventive here, offering simply the plain diminutive шантажик: “Затем, говоря метафорически, я разбил бокал вдребезги и смело представил себе [...], как постепенно я перейду на шантаж – о, совсем легкий, дымчатый шантажик – и заставлю большую Гейзиху позволить мне общаться с маленькой, ...” but Stiller succeeded in producing a creative diminutive – a neologism: “Po czym, obrazowo mówiąc, szkło prysło! i zuchwale wyobraziłem sobie (…), jak to mógłbym się puścić w końcu na szantaż – nie, to za mocne słowo – na jakieś liliowe szantażątko wobec dużej Haze, aby pozwoliła mi współżyć z małą Haze …” (Nabokov 1991: 79) Retranslated into English, the highlighted phrase means more or less “a mauve blackmail-cub” (the colour is retained because of its significance for the whole book). Kłobukowski is less successful here, offering a rather silly and meaningless rhyming triplet instead of a neologism, but again, it must be remembered that he was following in Stiller’s footsteps and could not repeat any of the previous solutions: “… i śmiało sobie wyobraziłem […], że mógłbym w końcu szantażem … nie, to za mocne określenie: raczej masażem, sondażem i persyflażem skłonić dużą Haze, aby pozwoliła mi zbliżyć się z małą, …” (Nabokov 1997: 85) Another unclassifiable coinage – this time of the portmanteau type – is the following one, combining libido and dream: “I am sure Dr Blanche Schwarzmann would have paid me a sack of schillings for adding such a libidream to her files.” (Nabokov 1980: 53) The Russian version translates this word literally, but also adds another, similar – and better – portmanteau coinage, this time blending libido and dossier: “Не сомневаюсь, что доктор Биянка Шварцман вознаградила бы меня целым мешком австрийских шиллингов, ежели бы я прибавил этот либидосон к ее либидосье.” Since the word либидосье is based on Latin and French roots, Stiller quite naturally copies it into Polish, giving us libidossier: TRANSLATION AND MEANING, PART 10, 2013, 79 - 90 89 “Dr Blanche Schwarzmann z pewnością nie pożałowałaby mi worka szylingów austriackich za uzupełnienie takim snem jej libidossier.” (Nabokov 1991: 60) Kłobukowski renders the less clever original English word and offers us the blend libidrzemka: “Nie wątpię, że doktor Blanche Schwarzmann dałaby mi wór szylingów za to, że wzbogaciłem jej kartotekę taką libidrzemką.” (Nabokov 1997: 64) Among the cleverer instances of miscellaneous wordplay we must mention the nicknames of the classrooms at Beardsley: “Mushroom, Room-In 8, B-room, Room-BA and so on. ” (Nabokov 1980: 195) The first word is obvious; the remaining ones play on ruminate, broom, and rumba, respectively. This seems extremely easy to translate into Polish – all we need is a selection of words with the element sala in them, but, for some strange reason, Stiller seems not to have noticed the punning nature of those names and to have translated their surface meanings only: “Grzyb, In Octavo, B-Klasa, Izba BA i tym podobne.” (Nabokov 1991: 220) and it is only in Kłobukowski’s translation that we see the puns translated as puns: “Sala Mandra, Sala Mina, Sala Da, Mar Sala i tak dalej.” (Nabokov 1997: 239) Incidentally, the Russian version is less impressive here – instead of producing a comparable play on words, based on the word класс (e.g. Класс-Ика, Класс-Ификация, Класс-Ицизм, Класс-Овость ['Class-Ics, Class-Ification, Class-Icism, Class Character']), Nabokov chose to use a series of silly rhymes (including one artificial rhyme) for the names of the classrooms: “Класс-Квас […], КлассРаз, Класс-Два-с, Класс-Алмаз и прочее.” 11. Compensation The translation of puns involves inevitable losses – either a pun gets omitted altogether or it does get translated but the target language pun is not as effective as the original wordplay. Consequently, sometimes a translator will introduce wordplay in places where the source text does not have it, but where the target language version offers such possibilities. This is known as compensation, and was used by Stiller in a number of places; for instance, for Humbert’s plain statement: “I became sickeningly conscious that Trapp had changed his tactics and was still with us, in this or that rented car.” (Nabokov 1980: 225)8 Stiller offers a clever horizontal paronymic pun: “… aż skręciło mnie od uświadomienia, że Trapp zmienił taktykę trapienia i dalej nam towarzyszy: w takim lub inym wynajętym samochodzie.” [Literally: “ … that Trapp had changed his tactics of pestering us …”] (Nabokov 1991: 253) Nabokov himself expanded “the cowman and the sheepman” from his English Lolita (1980: 298) into two paronymic puns in the Russian version (“мы оба пыхтели, как королю коров и барону баранов [i.e.: “the king of cows and the baron of rams” in the dative case] никогда не случается The Russian Lolita (“… а затем мне стало отвратительно ясно, что Трапп переменил тактику и продолжает ехать за нами, но уже в других, наемных машинах.”) has no wordplay here either. 8 90 JANUSZ WRÓBLEWSKI пыхтеть после схватки”), which Stiller naturally reproduced in Polish as: “obaj sapaliśmy tak okrutnie, jak to królowi krów i baronowi baranów po ich walce nigdy się nie zdarzało” (Nabokov 1991: 335), which – incidentally – shows once again how easy it is to translate some puns between related languages. (Since Kłobukowski worked solely with the English text, he has only “krowiarz i owczarz” – Nabokov 1997: 362). Naturally, there are more examples of compensation in the three translations under analysis (just as there are many more puns in the original novel), but for reasons of space they cannot all be listed here. In conclusion, it can be stated that although at first glance many puns may look untranslatable, they are untranslatable only in terms of semantic equivalence. If we redefine translation to include cases of formal substitution, then most instances of wordplay will turn out to be perfectly translatable, even if initially they do pose a challenge for the translator. And, as has been shown, some puns can actually be rendered semantically, especially when the source and target language are genetically closely related (English and German, Polish and Russian), or when the puns are based on “international” words, i.e. words of Latin or French origin: e.g. Nabokov’s “sexcapades” (1980: 297) end up as “sekskapady” in both Polish translations (Nabokov 1991: 334 and 1997: 361), to quote one final example from among the numerous alluring puns of Lolita. References Cambridge Advanced Learner’s Dictionary (2003). Cambridge: Cambridge University Press. Delabastita, D. (1993), There's a Double Tongue: an investigation into the translation of Shakespeare's wordplay, with special reference to Hamlet. Amsterdam / Atlanta, GA: Editions Rodopi B.V. (Approaches to Translation Studies, Vol. 11). Delabastita, D. (1996), “Introduction”. In: Delabastita, D. (ed.), Wordplay & Translation. Special Issue: The Translator. Studies in Intercultural Communication. 2 (2): 127-139. Delabastita, D. (2001), “Aspects of interlingual ambiguity: polyglot punning”. In: Quitte ou double sense. Articles sur l'ambiguïté offerts à Ronald Landheer. Textes réunis par Paul Bogaards, Johan Rooryck et Paul J. Smith. Avec la collaboration de Véronique van Gelderen. Amsterdam / Atlanta, GA: Rodopi: 45-64. Delabastita D. (2004), “Wordplay as a translation problem: A linguistic perspective”. In: Kittel, Harald, Juliane House, Brigitte Schultze, et al. (eds), Übersetzung: ein internationales Handbuch zur Übersetzungsforschung. Berlin: Walter de Gruyter GmbH & Co. 600-606. Fogle, S.F. (1974), “Pun”. In: Preminger A. (ed.), Princeton Encyclopedia of Poetry and Poetics. Enlarged Edition. London: The Macmillan Press. 681-682. Ginter, A. (2003), Świat za słowami Vladimira Nabokova. Zabawy słowne i ich przekład. Łódź: Wyd. UŁ. Lewandowska-Tomaszczyk, B. (2010), “Re-conceptualization and the emergence of discourse meaning as a theory of translation”. In: Lewandowska-Tomaszczyk, B. and M. Thelen (eds), Meaning in Translation. (Łódź Studies in Language, Vol. 19). Frankurt am Main: Peter Lang. 105-147. Maltzev, V.A. (1980), An Introduction to Linguistic Poetics. Minsk: Higher School. Nabokov, V. (1980), Lolita. Harmondsworth: Penguin Books. Nabokov, V. (1991), Lolita. Przełożył z angielskiego i rosyjskiego Robert Stiller. Warszawa: PIW. Nabokov, V. (1997), Lolita. Translated by Michał Kłobukowski. Warszawa: Wydawnictwo Da Capo. Nabokov, V. (1998), Лолита. Перевод с английского: Владимир Набоков. Downloaded from http://lib.ru/NABOKOW/lolita.txt [accessed 9 July 2009]. Paszek, J. and Skwarek, I. (1992), “Kalambury w Lolicie”. In: Fast, P. (ed.), Przekład artystyczny. Vol. 3: Tłumaczenia literatury polskiej na języki obce. Katowice: Uniwersytet Śląski: 77-91. Pratchett, T. (2003), Night Watch. London: Corgi Books. Shakespeare, W. (1958), The Complete Works of William Shakespeare. Ed. by G. B. Harrison. New York: Harcourt, Brace & World, Inc. Stiller, R. (1997), “Ty żagwio! albo Nabokov Kłobukowskiego”. Wi 91 Section II: Translation & Languages in Contrast 92 TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 93 INTERROGATION, HYPOTHESE, ARGUMENTATION. LE FRANÇAIS ET L’ANGLAIS EN CONTRASTE Agnès Celle Université Paris-Diderot, Paris, France Abstract: This paper examines the rhetorical function of interrogative and hypothetical clauses in academic discourse from a contrastive point of view. The study is based on a sample of both English texts translated into French and French texts translated into English. Interrogative clauses and if-clauses, which are highly frequent in French academic discourse, are shown to serve a similar topicalising function by simulating speaker-addressee interaction. In English, however, expository questions imply that speaker and addressee do not stand on an equal footing, which is not appropriate for social science texts, as opposed to textbooks for instance. As for if-clauses, they require a hypothetical feature in English, whereas si-clauses may function as contrastive topics in French, irrespective of whether the idea is realized or not. The ability of interrogatives and si-clauses to play a purely information structural role in French has no equivalent in English, and may partly explain why French academic discourse sounds more abstract than English academic discourse. 1. Introduction Le point de départ de cette étude est la lecture de plusieurs ouvrages de sciences humaines en français, en anglais et dans leur version traduite dans chaque langue. La comparaison interlangues fait apparaître des différences significatives dans la structuration de l’argumentation, qui repose sur un recours constant à l’interrogation et sur l’emploi fréquent de si à des fins rhétoriques en français, alors qu’en anglais il y a très peu d’hypothèses en if en dehors du champ de la condition, et très peu d’interrogations. L’analyse présentée ici se fonde sur un relevé systématique des interrogatives et des hypothèses en if et en si en position antéposée dans les textes de départ et leur traduction1. Dans la continuité des travaux de Guillemin-Flescher (2003, 2006), l’objectif ici n’est pas de proposer une méthode pour bien traduire, mais plutôt, à partir du produit fini que constitue la traduction, de dégager les récurrences propres à chaque système linguistique. La traduction en tant que donnée observable permet au linguiste d’accéder à des « schémas intériorisés » (Guillemin-Flescher 2006, 249) dans l’organisation du discours au sein des deux langues mises en contraste. Ces observations soulèvent un certain nombre de questions: quel rôle joue l’interrogation dans l’argumentation en français ? Pourquoi est-elle si fréquente, et pourquoi a contrario l’emploi de l’interrogation en anglais est-il si limité ? Pourquoi la conjonction si est-elle si fréquente en français au-delà de la simple condition, c’est-à-dire dans des emplois rhétoriques qui expriment la concession, le contraste ou encore l’explication, alors qu’en anglais on trouve des marqueurs de concession comme although ou de causalité comme since ou because ? La traduction révèle que if n’est pas employé dans les mêmes conditions que si : pourquoi, alors même qu’à en croire l’abondante littérature sur la question, if et si ont en commun le même caractère polysémique ? Depuis Jespersen, nombreux sont les linguistes qui ont rapproché l’interrogation et l’hypothèse, en tout cas d’un point de vue formel. La relative rareté de if rhétorique et de l’interrogation dans l’argumentation en anglais contrastée à la haute fréquence de l’interrogation et de si dans l’argumentation en français est-elle l’indice d’un fonctionnement discursif comparable ? Autrement dit, peut-on rapprocher l’interrogation et l’hypothèse au titre d’une fonction argumentative commune dont la traduc1 Ponctuellement, nous avons également eu recours au corpus ICE-GB pour être en mesure de comparer l’emploi de if rhétorique tel qu’il apparaît dans un corpus de sciences humaines à son emploi dans d’autres types de textes. 94 AGNÈS CELLE tion serait le révélateur ? Quirk et al. (1985, 15.37) font explicitement ce rapprochement à propos des questions rhétoriques et des « rhetorical conditional clauses » : « Rhetorical conditional clauses give the appearance of expressing an open condition, but (like rhetorical questions […]) they actually make a strong assertion. » Nous poserons que l’hypothèse et l’interrogation peuvent avoir une fonction discursive qui s’écarte de leur forme en français, alors que cet écart entre forme et fonction discursive est soumis à plus de contraintes en anglais. C’est la traduction qui nous donnera accès à cette différence entre les deux langues. 2. L’interrogation dans le discours théorique 2.1 Interrogation et interaction2 Le corpus examiné est composé de textes théoriques dans lesquels l’argumentation est essentielle. Plusieurs études, concernant tant l’anglais (Halliday (1994, 325), Hyland (2005, 37)), que le français (Fløttum et al., 2006, 30-31), ont mis en évidence la dimension tout à la fois interactionnelle et rhétorique de ce type de texte. A partir de corpus comparables en anglais, en français et en norvégien, Fløttum & al. (2006, 213) montrent en outre que l’anglais est, des trois langues examinées, celle qui établit la relation la plus étroite par rapport au lecteur à travers le métatexte. A l’inverse, le français est la langue qui est la plus orientée vers l’auteur. Fløttum & al. (2006, 213) ne considèrent pas le cas de l’interrogation. Toutefois, leurs résultats concernant le discours académique dans son ensemble et l’observation de notre relevé d’énoncés interrogatifs en français, en anglais et dans des textes traduits nous amènent à un constat a priori paradoxal : l’interrogation, qui semble être par excellence le signe d’une énonciation dialogique et interactionnelle, est fortement représentée en français et non en anglais, alors que le métatexte privilégie une relation interactionnelle dans les textes théoriques anglais et non français. Nous avons proposé une explication qui permet de résoudre cet apparent paradoxe dans Celle (2009), dont nous allons ici reprendre les grandes lignes, avant d’esquisser la comparaison avec l’hypothèse. Dans un corpus de textes théoriques, l’absence de référent situationnel a une conséquence sur le statut de celui à qui est destiné le texte: l’allocutaire ne peut avoir le statut de co-locuteur puisqu’il n’y a aucune interlocution. Par conséquent, l’allocutaire ne peut répondre aux questions qui sont posées. Quelle est alors la fonction de l’interrogation? A qui s’adresse-t-elle, et correspond-elle véritablement à une question si elle ne manifeste pas un réel recours à autrui? La structure interrogative ne correspond pas nécessairement à un acte de question, et vice versa. Kerbrat-Orecchioni (1991, 14) définit en ces termes l’acte de question: « un énoncé qui se présente comme ayant pour finalité principale d’obtenir de L2 un apport d’information. » Les interrogatives de notre corpus étant coupées de toute interlocution, elles n’ont jamais cette finalité. Huddleston (1994, 415), tout en prenant soin de distinguer l’interrogation de la question - la première est la réalisation syntaxique d’un type de proposition alors que la seconde renvoie au sens et non à la forme - observe que si la fonction prototypique de la question est de demander une information à l’allocutaire, elle peut également être employée dans bien d’autres cas. Dans des textes théoriques, elle vise selon lui à attirer l’attention de l’allocutaire sur une réponse que l’énonciateur donne lui-même à la question qu’il pose. De fait, dans les questions non rhétoriques de notre corpus, il arrive fréquemment que la suite textuelle apporte la réponse à la question posée : (1) What, then, is the test that determines whether a generalized Phrase-marker is the deep structure of some sentence? The answer is very simple. (Aspects p. 138) 2 Nous remercions la Revue Française de Linguistique Appliquée de nous autoriser à reprendre dans cette partie le début de notre article intitulé « Question, mise en question : la traduction de l’interrogation dans le discours théorique » et paru dans le numéro thématique Linguistique et Traduction, XIV, 2009-1, p. 39-52. Cet article est consultable en ligne sur le portail CAIRN à l’adresse suivante : http://www.cairn.info/revue-francaise-de-linguistique-appliquee/htm TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 95 Quel est alors le test déterminant si un Indicateur généralisé est la structure profonde d’une phrase? La réponse est très simple. (p. 190) (2) What might be the explanation for this phenomenon? As Ross observes, it can be explained in terms of the particular grammar of English if we assume, in addition, that certain transformations apply in a cycle… (Language p. 39) Quelle pourrait être l’explication de ce phénomène? Comme l’observe Ross il peut être expliqué en termes de la grammaire particulière de l’anglais si nous admettons en outre que certaines transformations s’appliquent dans un cycle… (p. 71) Les questions posées dans ces deux exemples ne sont pas des demandes d’information. Elles visent plutôt à anticiper les questions que tout allocutaire sera amené à se poser dans toute situation de lecture (cf. (1)), ou bien à souligner un problème qu’à la fois l’énonciateur et l’allocutaire peuvent rencontrer et qui va faire l’objet d’un traitement argumenté (cf. (2)). De cette façon, l’enchaînement « P ? Q » correspond à une suite topique / commentaire, et l’interrogation a une fonction topicalisante, comme cela a souvent été relevé (voir en particulier Grésillon 1981, 70). « P ? » introduit une problématique, ou ce qui peut poser problème à l’allocutaire. Cette fonction topicalisante nous paraît liée au recours à autrui qu’introduit l’interrogation, même si ce recours est fictif. La question suppose ici la prise en compte d’un co-énonciateur fictif, non pas en tant qu’instance de validation puisque la réponse ne peut venir que de l’énonciateur, mais dans l’anticipation de la question que l’allocutaire pourrait (se) poser. De plus, on attire ainsi l’attention de l’allocutaire en l’orientant vers la réponse que va produire la suite du texte (cf. Huddleston 1994, 414). Néanmoins, pour reprendre une expression forgée par J. Milner et reprise par Grésillon, il n’y a pas « égalité des droits » entre les deux partenaires, puisque l’allocutaire peut seulement faire entendre sa voix par le biais d’une question que lui prête l’énonciateur, sur laquelle il n’a pas pouvoir de validation. Cette absence d’« égalité des droits » est ce qui explique la rareté de l’interrogation en anglais dans le genre discursif étudié ici. Nous n’avons relevé qu’une occurrence d’interrogative non imbriquée dans Aspects of the Theory of Syntax et deux dans Language and Mind, alors que l’interrogative imbriquée y est récurrente. En l’absence de verbe recteur, l’interrogation est attribuée par défaut à l’énonciateur. Or précisément, l’origine de l’interrogation est rarement le seul énonciateur en anglais. Dans la traduction anglaise de textes originaux français, l’interrogative imbriquée est privilégiée, ce qui permet d’attribuer l’interrogation à une instance subjective qui éventuellement inclut l’allocutaire, comme dans l’exemple suivant : (3) Pour qu’il y ait effet de figure, il faut que les locuteurs soient bien conscients que tel effet de sens est déviant et en apprécient l’intention, poétique ou ludique. C’est alors que se fait la recherche d’un sens. Comment opère la règle de transfert sémantique? Il semble que toute anomalie soit décodée en deux temps. (Alice p. 167) An expressive value is only obtained when we realize that the meaning of an utterance is deviant and that a poetic effect or a joke is intended: only then do we search for an interpretation via the semantic transfer rule. We should perhaps now explore how the semantic transfer rule actually operates in practice. It seems to be the case that each anomaly of meaning is decoded in two separate stages. (p. 131) L’interrogative imbriquée est préférée à la question How does the semantic rule actually operate in practice? Les raisons de cette préférence sont discursives. Avec le choix de la 1ère personne du pluriel we et du verbe explore, c’est un processus métadiscursif commun à l’énonciateur et à l’allocutaire qui est privilégié en anglais. Dans ce contexte, il est plus difficile de poser une question, car cela supposerait qu’énonciateur et allocutaire soient séparés. Nous rejoignons ici les conclusions de RowleyJolivet (2007, 176-177), qui constate la rareté des questions dans les articles de recherche en anglais, 96 AGNÈS CELLE et l’attribue au fait que les questions exercent un contrôle sur le discours et une forme d’autorité qui seraient perçus comme condescendants par le lecteur. La stratégie discursive adoptée est donc différente dans les deux langues : en français, la question suppose qu’énonciateur et co-énonciateur soient distingués sans être, comme nous l’avons vu, sur un pied d’égalité; en anglais, il est plus difficile de maintenir cette différence dans le discours théorique. La différence entre les partenaires énonciatifs est réduite, ce qui permet de les envisager dans un domaine commun, mais fictif. La distance est ainsi modale mais pas fondamentalement intersubjective. 2.2 De l’interrogation à l’hypothèse Le lien entre protase à forme interrogative et question a été noté par Jespersen (1940, 374). La parenté entre question et hypothèse3 apparaît clairement dans les exemples suivants, où la réponse implicite et positive à la question totale peut parfaitement s’interpréter comme une protase qui déclenche l’apodose : (4) Faut-il en conclure que les réseaux d’intermariage sont à peu près constants, en taille absolue, dans toutes les sociétés humaines? Dans l’affirmative, la nature complexe d’une société résulterait moins d’une dilatation de l’isolat primitive, que de l’intégration d’isolats relativement stables dans des ensembles de plus en plus vastes… (Anthropologie structurale p. 350) (5) De plus, il existe une relation certaine entre le mode de fonctionnement et la durabilité d’une structure sociale, et l’effectif de la population. N’y aurait-il pas des propriétés formelles des groupes qui seraient directement et immédiatement fonction du chiffre absolu de la population, indépendamment de toute autre considération? Dans l’affirmative, il faudrait commencer par déterminer ces propriétés et par leur faire une place, avant de chercher d’autres interprétations. (Anthropologie structurale p. 349) Dans l’affirmative joue le rôle de protase en apportant la réponse à la question totale qui précède : oui en (4) et si en (5). On peut toutefois préciser une différence entre les deux questions. En (4), la question au présent est une vraie question, à laquelle dans l’affirmative apporte en guise de réponse la valeur p en écartant la valeur complémentaire. On pourrait ici paraphraser dans l’affirmative par si oui. En (5), l’interronégative au conditionnel de non-prise en charge4 est par contre une question rhétorique orientée vers une confirmation positive. L’énonciateur est conduit à admettre et à faire admettre à l’allocutaire une conclusion positive. On pourrait remplacer dans l’affirmative par alors. En introduisant ou en reprenant la valeur positive, dans l’affirmative pose un cadre hypothétique qui déclenche l’emploi du conditionnel d’hypothèse dans la suite que l’on peut interpréter comme une apodose. Dans la traduction anglaise, on constate que les questions sont supprimées. On retient du français l’émergence de la valeur p au détriment de la valeur complémentaire, ce qui justifie l’assertion. Alors que l’argumentation procède par étapes en français, dans une succession de questionnement fictif et de mise à distance fictive avec le conditionnel, l’assertion est privilégiée en anglais si la valeur p a un caractère évident. Le jugement modal apparaît alors sous la forme d’un nom (inference) ou d’un adjectif (evident) et porte sur la complétive qui suit ou la construction extraposée : (4) The inference is that, while the absolute size of the intermarrying group remains approximately on the same scale in all human societies (the ratio of the French types in relation to the average primitive types being about 10 to 1), a complex society becomes such not so 3 On pourra consulter sur cette question Chuquet (1994) pour l’anglais et Béguelin & Corminboeuf (2005) pour le français. 4 Sur ce choix terminologique, voir Celle (2007). TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 97 much because of an expansion of the isolate itself as on account of an expansion of other types of social links… (p. 294) (5) There is an obvious relation between the functioning and even the durability of the social structure and the actual size of the population. It is thus becoming increasingly evident that formal properties exist which are immediately related to the absolute size of the population, whatever the group under consideration. These should be the first to be assessed and taken into account in an interpretation of other properties. (p. 293) Interrogation et hypothèse disparaissent dans le processus de traduction au profit d’une reconstruction de l’argumentation dans le plan du certain où il s’agit aussi de poser, d’évaluer et de hiérarchiser les arguments, mais d’une autre façon. On s’aperçoit que l’on a le même type de structure en anglais dans la traduction de l’hypothèse que dans la traduction de l’interrogation (cf. (3)) : une structure subordonnée5. 3. Hypothèse et topicalisation: entre prédiction et évaluation épistémique Le statut de topique des propositions hypothétiques a été abondamment décrit en linguistique générale. Haiman (1978) affirme dans un article devenu célèbre « conditionals are topics ». Haiman (1986, 215) s’appuie explicitement sur la parenté entre les propositions hypothétiques et interrogatives: « A protasis is established as a given, or topic, by means of a question. The formal identity of topics and questions is thus pragmatically rather than semantically motivated. » Pour reprendre les exemples précédents, « dans l’affirmative » en (4) et (5) reviendrait à prendre pour topique, sous forme déclarative, la question qui précède. Une fois la question posée et la réponse obtenue ou suggérée, il y aurait un consensus, un savoir partagé concernant une proposition qui pourrait jouer un rôle de topique dans le discours. L’équivalence que Haiman établit entre topic et given est problématique selon nous. Nous rejoignons Akatsuka (1986, 343) dans la critique qu’elle formule concernant l’assimilation par Haiman du topic au given. Akatsuka opère une distinction entre contrastive topic et thematic topic. Les propositions hypothétiques en position antéposée seraient des contrastive topics, mais pas des thematic topics. Si le rapprochement entre questions et protases est justifié, c’est précisément parce que dans les deux cas il y a pour l’énonciateur une part d’incertitude et d’incontrôlabilité (Akatsuka, 1986: 343). La question marque le recours au co-énonciateur pour lever cette incertitude, tandis que dans l’hypothèse, l’énonciateur sélectionne une valeur qui est mise à distance en étant fictivement validée. Comme le résume Chuquet (1994, 220): « la question est une suspension d’assertion faute de pouvoir localiser la relation, alors que l’hypothèse repose sur une assertion, même si elle est d’un type particulier, et relève donc avant tout d’une forme de validation. » Les hypothèses en position antéposée rencontrées dans notre corpus assurent essentiellement une fonction prédictive ou épistémique. 3.1 Prédiction Au même titre que l’interrogation, l’hypothèse joue un rôle particulier dans un corpus de textes théoriques. A si marqueur de condition peut correspondre if marqueur de condition, comme dans l’exemple suivant : (6) If we establish the proper “psychic distance” from such elementary and commonplace phenomena as these, we will see that they really pose some non-trivial problems for human psychology. (Language p. 52) 5 Pour une analyse détaillée des différences entre interrogative imbriquée et non imbriquée, voir Celle (2009). 98 AGNÈS CELLE Si nous prenons le “recul psychique” nécessaire devant des phénomènes aussi élémentaires et aussi communs que ceux-là, nous verrons qu’ils posent réellement des problèmes importants pour la psychologie humaine. (p. 91) Dans chaque langue, la protase a ici une valeur prédictive : elle introduit la condition de l’actualisation de l’apodose. Toutefois, une différence tend en général à apparaître dans la traduction. On peut observer que l’anglais distingue plus nettement que le français ce qui relève du certain de ce qui relève de l’hypothèse, ce qui se manifeste par une fréquence supérieure de if dans des contextes non assertifs, et inversement par une absence de if dans des emplois méta-énonciatifs, sur lesquels nous allons revenir. Examinons tout d’abord le cas où if apparaît dans la traduction d’énoncés assertifs : (7) Les choses sont claires: l’inconscient se reflète dans la parole sans passer par la langue. La créativité du discours, consciente ou inconsciente, ne doit rien à la langue. (Alice, p. 141) If we accept Benveniste’s arguments, we are led to conclude that the unconscious is reflected in parole without being filtered by langue. (p. 109) Dans cet exemple, la prédication « l’inconscient se reflète dans la parole sans passer par la langue » est exprimée sous forme assertive en français, alors qu’il s’agit d’une déduction à partir de la position de Benveniste. En anglais, la dépendance par rapport au point de vue de Benveniste passe par l’expression de l’hypothèse : c’est dans ce cadre-là, fictif et non réel, que la conclusion est permise. A l’inverse, if, contrairement à si, n’est pas employé dans des énoncés méta-énonciatifs qui visent à caractériser les étapes d’un dire qui est en train de s’énoncer : (8) Returning to the form right, we see that the final consonant would be correctly determined as [č] rather than [š] if in the underlying representation there were a continuant preceding it… (Language p. 35) Si nous revenons à la forme right, nous voyons que la consonne finale serait déterminée de façon correcte comme [č] plutôt que [š] si elle était précédée d’une continue dans la représentation sous-jacente… (p. 65) (9) Putting the same requirements in somewhat different terms, we must require of such a linguistic theory that it provide for… (Aspects p. 31) Si nous formulons les mêmes exigences en des termes un peu différents, nous devons requérir d’une telle théorie linguistique qu’elle prévoie… (p. 50) (10) Si l’on accepte de s’abstraire de cette dernière, le problème se limite à une opposition très simple entre: (Alice p. 176) Leaving that aspect to one side, however, the problem can be expressed as a very simple opposition between: (p. 139) La proposition participiale en position détachée et antéposée a la même valeur topicalisante en anglais que la protase en français. Toutefois, elle ne simule pas une hypothèse, à la différence de si en français, qui construit sur un mode fictif l’énonciation des étapes de l’argumentation. Avec des verbes qui contribuent lexicalement à la définition du métatexte, la forme en –ing dans une participiale antéposée actualise la relation de la participiale, quel que soit le type de modalité qui apparaît ensuite dans la principale. La participiale en –ing n’identifie pas le sujet, qui peut s’identifier a posteriori par coréférence avec le sujet de la principale comme en (8) et (9), ou bien par défaut avec l’énonciateur comme en (10), où il n’y a pas coréférence avec le sujet de la principale. Cette souplesse syntaxique facilite la mise en place d’un cadre pour toute principale, quel que soit son sujet et quelle que soit la modalité dont elle est porteuse. TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 99 La participiale fournit ici un cadre à la principale tout en la qualifiant, et seule la principale dispose de ses propres coordonnées modales et aspectuo-temporelles. En français avec si, une situation fictive est mise en place, même si en réalité le dire est en train de s’énoncer. On retrouve donc en anglais la même réticence à mettre en scène un processus métalinguistique qui est en cours, que ce soit au moyen de l’interrogation (voir (3), (4’) et (5’)) ou au moyen de l’hypothèse, alors que les deux procédés sont utilisés en français. 3.2 Les emplois épistémiques de si et if Dans les emplois épistémiques, la protase n’a pas de valeur prédictive en vue de l’actualisation de l’apodose. Il s’agit plutôt d’évaluer dans l’apodose ce que la protase implique ou ce qu’elle veut dire. Les valeurs explicatives, concessives, adversatives, dialectiques, métalinguistiques6 que recouvre l’emploi épistémique de if et de si existent en français et en anglais et sont décrites de façon comparable dans la littérature7. Pourtant, à l’examen de notre corpus, elles ne se recoupent pas, et cela transparaît dans la traduction. Si apparaît beaucoup plus que if dans des emplois concessifs, adversatifs et explicatifs. Essayons de comprendre pourquoi si accède à cette fonction rhétorique plus facilement que if. Dans l’exemple suivant, on peut paraphraser la protase par s’il est vrai que ou comme il l’explique : (11) Il est frappant de constater que, pour Radcliffe-Brown, le problème n’existe pas. Si toute organisation sociale se réduit à un conglomérat de relations de personne à personne, le système est extensible indéfiniment: pour tout individu masculin il y a, au moins théoriquement, une femme qui sera avec lui dans la relation de fille du frère de la mère…. (Anthropologie structurale p. 363) La dépendance par rapport au point de Radcliffe-Brown n’est pas explicitement marquée en français, ce qui laisse possible l’interprétation prédictive. Or la proposition « toute organisation sociale se réduit à un conglomérat de relations de personne à personne » est bien le point de vue de RadcliffeBrown, et ce point de vue va être mis à l’épreuve, évalué dans l’apodose. La protase en si permet de le prendre comme topique8, comme cadre fictif. Il y a bien une reprise de ce qui précède pour en évaluer l’implication dans la suite du texte. En anglais, un autre choix est fait : (11’) For Radcliffe-Brown, however, there is no problem involved, since he considers any kind of social organization as a mere conglomerate of simple person-to-person relations and since, in any society, there is always somebody who may be regarded as one’s mother’s brother’s daughter… (p. 305) On identifie l’argument avancé au point de vue de Radcliffe-Brown, ici au moyen de he considers, pour privilégier une relation causale (since) à une relation hypothétique. Dans la traduction, on réduit la relation hypothétique du français à une spécification de la proposition principale (there is no problem involved) qui, avec since, présuppose (voir Dancygier & Sweetser 2005, 182-183) l’existence de la cause. La cause est alors nécessairement acquise et incontestable, ce qui ne serait pas le cas avec if: if laisserait une ambiguïté, la protase étant susceptible de jouer un rôle prédictif en conditionnant l’actualisation de l’apodose. 6 Nous reprenons ici les valeurs que distingue de Vogüé (1999, 94). Pour l’anglais, voir la description particulièrement exhaustive de Declerck & Reed (2001), ainsi que Dancygier & Sweetser (2005). Pour le français, voir notamment de Vogüé (1999), Adam (2005) et Achard-Bayle (2006). 8 Le statut de topique de si, comparé à d’autres adverbiaux antéposés, est analysé par Charolles (2003), qui met en évidence les mécanismes de reprise. Nous verrons que sur ce point, la traduction révèle une différence entre l’anglais et le français. 7 100 AGNÈS CELLE On tend en français à présenter une proposition sous forme fictive avec un si épistémique dans le seul but de l’évaluer, même si elle n’a aucun caractère hypothétique. Cet écart entre le caractère non hypothétique d’une relation et l’emploi d’une forme hypothétique explique en partie la divergence que l’on peut observer dans les deux langues concernant la traduction de si épistémique. Nous allons voir que cet écart repose sur une organisation différente de la structure informationnelle. 3.2.1 La valeur explicative A la fois if et si peuvent être employés dans une protase que l’apodose vise ensuite à expliquer. L’évaluation de la protase par l’apodose implique que l’apodose « donne la mesure » de la protase (de Vogüé 1999, 114), qui se trouve, dans le cas du si explicatif, « banalisée ». L’exemple qu’elle donne est « s’il est riche, c’est parce qu’il a gagné au loto ». La protase se vérifie, mais, comme l’explique de Vogüé (1999, 94), l’apodose vise à banaliser, à minimiser le fait qu’il soit riche. If a un emploi tout à fait comparable: (12) If Beckett makes few value judgements in his text, it is because his whole position is one of assertive though ungrounded evaluation. <ICE-GB:W2A-004 #030:1> Dans cet exemple, il s’agit d’expliquer la raison d’être de la rareté des jugements de valeur dans l’écriture de Beckett. Il y a matière à polémique dans la protase, et l’apodose relativise l’importance de cette particularité d’écriture en l’expliquant à l’aune de l’attitude générale de l’écrivain. Le mécanisme est le même dans l’exemple suivant, où deux protases se succèdent et où l’apodose apporte une explication en remontant à la cause de la manifestation de la prédication exprimée dans chaque protase. Finalement, l’enchaînement protase apodose rend normale la prédication au sein de chaque protase en l’expliquant : (13) Le mouvement de la passion qui se poursuit jusqu’à se rompre et à se retourner contre luimême, le surgissement de l’image, et les agitations du corps qui en étaient les concomitances visibles – tout cela, au moment même où nous essayions de le restituer, était animé secrètement déjà par ce langage. Si le déterminisme de la passion s’est dépassé et dénoué dans la fantaisie de l’image, si l’image, en retour, a entraîné tout le monde des croyances et des désirs, c’est que le langage délirant était déjà présent… (Histoire de la folie p. 256) The movement of passion which persists until it breaks and turns against itself, the sudden appearance of the image, and the agitations of the body which were its visible concomitants – all this, even as we were trying to reconstruct it, was already secretly animated by this language. If the determinism of passion is transcended and released in the hallucination of the image, if the image, in return, has swept away the whole world of beliefs and desires, it is because the delirious language was already present… (p. 100) Il semble toutefois qu’il y ait une contrainte essentielle à cet emploi en anglais qui n’existe pas en français: la protase doit correspondre à une reprise et non à un nouvel argument. Il y a bien consensus entre énonciateur et co-énonciateur sur le fait que l’écriture de Beckett contient peu de jugements de valeur en (12), ce qui autorise la reprise sous la forme de la protase. En (13), les deux protases reprennent explicitement le passage qui précède, qui correspond à un savoir partagé. Tant que la protase peut s’appuyer sur un savoir partagé, l’emploi de if explicatif est possible en traduction. La différence entre les deux langues tient au fait que si explicatif ne nécessite pas ce savoir partagé. La prédication qui apparaît dans la protase en français peut correspondre à un nouvel argument : (14) Les métaphores militaires envahissent, à notre insu, tant elles sont usées, toute la rhétorique politique et syndicale. Si les métaphores nous frappent dans les langues étrangères et en particulier dans les langues dites “exotiques” […], c’est qu’elles ne nous sont pas familières. (Alice p. 168) TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 101 Metaphors in foreign languages strike us more than they do in our native language because they are unfamiliar to us. (p. 133) Dans cet exemple, l’énoncé qui précède la protase signale que dans sa propre langue, un locuteur ne remarque pas les métaphores militaires. Dans la protase, le topique se construit par contraste, en opposant les métaphores de la langue maternelle aux métaphores des langues étrangères. Ce topique ne repose pas sur un savoir partagé : il n’est pas établi au préalable que les métaphores nous frappent dans les langues étrangères. En l’absence de reprise possible, on maintient en anglais la relation causale (because), mais il n’y a pas lieu de poser en plus un enchaînement protase apodose qui viserait à banaliser la prédication de la protase, puisque cette prédication n’a pas encore été posée. On peut même avoir en français un enchaînement de topiques successifs : (15) Quant aux calembours, s’ils sont d’autant meilleurs qu’ils sont plus approximatifs, c’est que, justement, l’effet de distorsion volontaire en est souligné. C’est l’équivoque du sens, non l’ambiguïté, qui produit l’humour. (Alice p. 178) A play on words is thus exploiting only a pseudo-ambiguity: it is an equivocal, not an ambiguous meaning, which produces humour. (p. 141) Quant aux calembours introduit un premier topique par contraste, puis la protase, qui contient les deux termes d’une relation de comparaison, en introduit un second, suivi ensuite d’un commentaire explicatif. En anglais, on prédique d’abord la propriété qu’il s’agit d’expliquer. En français, l’énoncé condense deux opérations simultanément : il met en place un topique contrastif, qui est aussitôt évalué, en l’occurrence banalisé. Si explicatif peut aussi apparaître en français dans la méta-énonciation, c’est-à-dire pour caractériser le dire. Il est frappant de constater que l’articulation à ce qui précède ne repose pas sur les mêmes bases avec si et avec if : (16) All of this has been said before. I repeat it at such length because it has been so grossly misunderstood. (Aspects p. 38) Tout cela a déjà été dit; si je le répète aussi longuement, c’est à cause des erreurs grossières qui sont apparues à ce propos. (p. 60) Dans l’énoncé anglais, il faudrait imaginer un contexte polémique pour qu’une mise à distance du dire soit justifiée. Si une critique avait été formulée contre la répétition en question, celle-ci pourrait être reprise dans une proposition en if pour donner lieu à une explication. Or ici, il s’agit simplement de caractériser le dire qui est en train de se faire. Dans la traduction française, si s’articule sur l’énoncé précédent (tout cela a déjà été dit) en mettant en cause l’implication de celui-ci : « tout a déjà été dit » implique qu’il n’est normalement pas nécessaire de répéter. L’articulation à l’énoncé précédent repose donc sur la mise en question d’une implication mais pas nécessairement sur une reprise avec si. Lorsque le contenu de la proposition explicative annonce des raisons qui vont être développées dans la suite du texte, la structure correspondante n’est pas employée en anglais: (17) Pour lui, le rôle de l’ethnologue est d’établir des corrélations entre divers types de religions et divers types d’organisations sociales (1945). Si sa sociologie religieuse se solde finalement par un échec, c’est, semble-t-il, pour deux raisons. (Anthropologie structurale p. 375) The anthropologist’s task is to discover correlations between different types of religions 102 AGNÈS CELLE and different types of social organization. Radcliffe-Brown failed to achieve significant results, however, for two reasons. (p. 313) (18) Another reason for the failure of traditional grammars, particular or universal, to attempt a precise statement of regular processes of sentence formation and sentence interpretation lay in the widely held belief that there is a “natural order of thoughts” that is mirrored by the order of words. (Aspects p. 6- 7) Si les grammaires traditionnelles, particulières ou universelles, n’ont pas tenté d’énoncer avec précision les processus réguliers de la formation et de l’interprétation des phrases, c’est aussi pour une autre raison: la croyance largement répandue en un “ordre naturel des pensées” qui serait reflété par l’ordre des mots. (p. 17) En (17), l’échec de Radcliffe-Brown est une donnée informationnelle totalement nouvelle. L’emploi de si explicatif repose sur la proposition explicative en c’est […] pour deux raisons, qui annonce les deux raisons énoncées dans la suite textuelle. Le lien à l’amont du texte n’est pas de l’ordre de la reprise: il s’agit plutôt ici de contraster le point de vue de Radcliffe-Brown énoncé préalablement (Pour lui) à celui de l’énonciateur, qui va tout à la fois mettre en évidence l’échec de cette position et l’expliquer par la suite. En l’absence de reprise par rapport à l’amont du texte, et vu l’annonce de l’explication dans la suite du texte, la structure if p, it is because n’est pas adaptée en anglais. L’énonciation de l’explication nécessite d’abord la prédication indépendante de l’événement qu’il s’agit d’expliquer. Il est intéressant d’observer en (18) que dans le sens anglais français, on assiste au phénomène symétrique. En anglais, l’énonciation de l’explication fait l’objet d’une prédication indépendante. La traduction au moyen de si explicatif est dictée par des raisons de topicalisation en français, et non par une quelconque reprise. Il suffit du contraste que fournit un élément nouveau, ici une autre raison, pour que le recours à si explicatif soit possible. D’un point de vue discursif, il serait très maladroit en français de commencer la phrase par un indéfini en prédiquant l’existence d’une autre raison comme en anglais (une autre raison […] résidait dans…). La longueur du segment en question joue également un rôle, et il est préférable de le placer en position de topique pour lui apporter ensuite une explication. 3.2.2 La valeur concessive Les contraintes qui apparaissent dans l’emploi de if concessif sont également liées à la structure informationnelle de l’énoncé et au statut d’hypothèse de la protase. Certes, l’emploi concessif de if est attesté : (19) If the new ideas do provide empirically successful studies it still remains to be asked why such brain damage (to any one or more of several components) leads to such a specific effect (Baddeley, 1990). <ICE-GB:W1A-004 #101:1> La concession déconstruit une relation inférentielle entre deux propositions (Ranger, 1998, 24). Normalement, p entraîne non q. Mais ce processus inférentiel est bloqué, ce qui crée une contradiction entre p et q. Dans cet énoncé, on s’attend à disposer de nouveaux résultats. If dans la protase, soutenu par still dans l’apodose, indique que contrairement à cette attente, il reste des questions sans réponse. Tout en admettant p, l’énonciateur rejette les conséquences logiques de p en sélectionnant q. Selon les termes de de Vogüé, (1999, 94), l’apodose sert à minimiser la protase. D’un point de vue discursif, la protase en if permet à l’énonciateur de reprendre objectivement un point de vue exprimé par TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 103 d’autres locuteurs9, pour ensuite l’évaluer personnellement dans l’apodose. La fonction échoïque est ici particulièrement nette avec l’auxiliaire do. Il y a reprise d’un contenu propositionnel sur lequel il y a consensus, ce qui fournit une transition pour exprimer le point de vue de l’énonciateur qui va dévaloriser le précédent. Toutefois, plusieurs études contrastives ont montré que if concessif est limité en anglais. Paillard (1993, 222) souligne qu’il est difficile d’interpréter if comme concessif dans la référence à une occurrence passée. Lewis (2005, 40, 42) note pour sa part que cet emploi de if est cantonné à des « contextes subjectifs ». Selon elle, il doit s’agir d’un jugement et non d’un fait, afin de préserver le statut incertain de la protase en if : “The English if-construction […] strongly implies, and in certain varieties probably requires, that the status of the idea in the if-clause is hypothetical or unknown, making it very difficult to use in contexts where the idea is realized. » A l’inverse, Lewis constate que si concessif joue un rôle important dans la structure informationnelle de l’énoncé. En effet, on peut en français avoir recours à une assertion fictive, même si le fait exprimé est avéré, dans l’unique but de mettre à distance un argument au profit d’un autre argument. Ce choix discursif apparaît dans la traduction de l’exemple suivant : (20) In some ways, the Chinese state has made it much easier to discuss the “duty of memory” than at least some western societies. The state has made it very clear what the “correct” interpretation of the Massacre’s history is, and in doing so, have freighted the terrible events of winter 1937-38 with a clear moral burden, a patriotic injunction that China must never again let itself be vulnerable to such a traumatic event. This is an imperative that is not always at odds with historical analysis, but which nonetheless encourages interpretation along paths that are already defined, rather than letting it roam freely. (Ritter) D’une certaine manière, l’État chinois a rendu le débat sur « le devoir de mémoire » plus facile que dans certains pays occidentaux ; il a très clairement désigné l’interprétation « correcte » de l’histoire du massacre, et ce faisant, a lié aux terribles événements de l’hiver 1937-1938 une dette morale, une injonction patriotique à ne jamais laisser la Chine être de nouveau la proie d’un tel traumatisme. Si cet impératif ne va pas nécessairement contre l’analyse historique, il oriente néanmoins l’interprétation dans un sens prédéfini, plutôt que de la laisser se développer librement. En anglais, le marqueur adversatif but suivi de nonetheless exprime la concession. Celle-ci est donc déjà présente sémantiquement, même si elle n’est pas exprimée par la subordination. Dans la traduction française, on remplace la structure coordonnée par un enchaînement protase apodose, c’est-àdire que l’on accroît le poids de l’argument exprimé dans l’apodose en « simulant une hypothèse » (Paillard, 1993, 220). Ce choix discursif est caractéristique de l’écrit en français. On évalue une proposition en la prenant pour hypothèse, qu’elle réfère à un événement avéré ou pas, ce qui permet de hiérarchiser les arguments. Dans le sens français anglais, on assiste de façon symétrique au remplacement de la subordination par la coordination en but : (21) Si la Révolution n’a pas modifié les normes de cette écriture, parce que le personnel pensant restait somme toute le même et passait seulement du pouvoir intellectuel au pou9 Pour reprendre les termes de Lyons (1977, 805), on pourrait dire que la protase introduit ici une modalité épistémique de nature objective, et non subjective. 104 AGNÈS CELLE voir politique, les conditions exceptionnelles de la lutte ont pourtant produit, au sein même de la grande Forme classique, une écriture proprement révolutionnaire, non par sa structure, plus académique que jamais, mais par sa clôture et son double, l’exercice du langage étant alors lié, comme jamais encore dans l’Histoire, au Sang répandu. (Degré zéro p. 19) The Revolution did not modify the norms of this writing, since its force of thinkers remained, all things considered, the same, having merely passed from intellectual to political power; but the exceptional conditions of the struggle nevertheless brought about, within the great Form of classicism, a revolutionary mode of writing proper, defined not by its structure (which was more conventional than ever) but by its closed character and by its counterpart, since the use of language was then linked, as never before in history, to the Blood which had been shed. (p. 21) La circonstancielle de cause en incise (parce que le personnel pensant…) rend particulièrement problématique la traduction par if. En effet, cette circonstancielle de cause à l’imparfait présuppose une réalité passée qui entrerait en contradiction avec la valeur hypothétique de if. On préfère donc en anglais traduire d’abord la relation causale au prétérit, en renvoyant ainsi à un fait passé et nullement hypothétique, puis marquer au moyen de la conjonction de coordination but la relation adversative. Alternativement, la même organisation thématique qu’en français peut être adoptée dans la traduction, mais avec la conjonction although qui introduit sans ambiguïté la valeur de concession, et qui peut être suivie du modal may, lequel introduit la modalité du possible : (22) Or, s’il est légitime, et en un sens inévitable, d’avoir recours à l’interprétation naturaliste pour essayer de comprendre l’émergence de la pensée symbolique, celle-ci une fois donnée, l’explication doit changer aussi radicalement de nature que le phénomène nouvellement apparu diffère de ceux qui l’ont précédé et préparé. (Anthropologie structurale p. 68-69) And although it may be legitimate or even inevitable to fall back upon a naturalistic interpretation in order to understand the emergence of symbolic thinking, once the latter is given, the nature of the explanation must change as radically as the newly appeared phenomenon differs from those which have preceded and prepared it. (p. 51) 3.2.3 Valeur adversative A la différence de si concessif, si adversatif n’indique pas une contradiction entre protase et apodose. Il y a un contraste entre les deux propositions, qui signale que la prédication de la protase n’est pas exclusive: elle coexiste avec la prédication exprimée dans l’apodose, laquelle prend plus de poids dans l’argumentation. En anglais, if adversatif existe, tout comme if explicatif et if concessif : (23) If Sisley was prepared to suggest renovation and class mixture, Pissarro selected a viewpoint that narrowed the street... <ICE-GB:W2B-002 #091:2> If prend une valeur clairement adversative dans cette protase au prétérit, dont le sujet s’oppose à celui de l’apodose. En revanche, dans la traduction des exemples au présent de notre corpus, on aura plutôt while, ou bien on the other hand. Dans ces exemples, la valeur adversative n’apparaîtrait pas de prime abord avec if, qui, associé au présent, pourrait être interprété comme une condition en l’absence de reprise permettant d’asseoir la protase sur un savoir partagé: (24) Si la polysémie peut être génératrice d’ambiguïtés, volontaires ou involontaires, elle n’est pas aussi propice au calembour que l’homophonie et la paronymie. (Alice p. 175) Polysemy is a source of ambiguities, whether deliberate or not. Homophony and paronymy, on the other hand, lend themselves better to punning. (p. 138) TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 105 (25) Il en découle que s’il faut être sociologue pour faire de la sociologie et mathématicien pour faire des mathématiques, il n’est nul besoin d’être linguiste pour faire de la linguistique, car le langage est à tout le monde. (Alice p. 12-13) From this it follows that while you have to be a sociologist to do sociology, and a mathematician to do mathematics, there is a form of intuitive, pre-theoretical, spontaneous linguistics which is practiced by all speakers merely because they are speakers. Language belongs to everyone. (p. 2) Le choix de while et de on the other hand va dans le même sens : il permet d’envisager sous forme assertive la proposition qui traduit la protase du français tout en la posant explicitement comme le premier terme de la relation de contraste entre les deux propositions. La traduction révèle donc deux points importants: 1) Bien que if et si disposent des mêmes emplois rhétoriques, il y a des limites à l’emploi épistémique de if. Les différentes valeurs rhétoriques de if s’articulent sur la reprise d’un savoir partagé. Cette reprise permet d’installer la protase dans une valeur épistémique objective. En l’absence de savoir partagé récupérable, la protase en if est susceptible de prendre un rôle prédictif en conditionnant l’actualisation de l’apodose. En français, si apparaît comme « l’accessoire obligé de la thématisation », selon la formule de de Cornulier (1986, 67). Cela ne signifie pas que si perde pour autant sa valeur d’hypothèse. L’emploi rhétorique de si consiste précisément à prendre pour hypothèse une proposition à une seule fin argumentative, en faisant abstraction du caractère avéré ou pas de la proposition en question. Il semble beaucoup plus difficile d’abstraire la proposition en if d’une situation à caractère hypothétique, alors que si permet de mettre à distance toute prédication par une simulation d’hypothèse, dès lors qu’un lien sémantique peut être construit entre protase et apodose par la concession, l’explication ou le contraste. Dans le discours théorique, si est abondamment employé pour poser un topique qui tisse un lien avec l’avant du texte en simulant un consensus avec le co-énonciateur, et il marque une transition vers un argument que l’énonciateur souhaite valoriser. En anglais, il semble que l’emploi de if rhétorique repose nécessairement sur un consensus réel – matérialisé par la reprise - et non simulé avec le co-énonciateur. 2) Il y a de nettes divergences discursives dans l’emploi rhétorique de if et de si selon le type de discours. Dans un contexte oral et polémique, il est possible en anglais d’intégrer sous forme d’hypothèse le dire d’autrui pour l’évaluer avec if. Ford & Thompson (1986, 354), de même que Rowley-Jolivet & Carter-Thomas (2008) ont noté que if est beaucoup plus fréquent à l’oral qu’à l’écrit. De son côté, Carter-Thomas (2007) a également noté que si en français est plus fréquent dans l’argumentation que if en anglais. Nos résultats fondés sur un corpus écrit de textes de sciences humaines sont concordants, et soulignent qu’à l’intérieur du genre des sciences humaines, il conviendrait d’affiner le contraste entre l’oral et l’écrit, puisque les fréquences d’emploi sont dissymétriques : if est fréquent dans des textes académiques à l’oral et non à l’écrit, contrairement à si dont l’emploi rhétorique est plus fréquent à l’écrit qu’à l’oral. 4. Conclusion Interrogation et hypothèse sont des outils rhétoriques indispensables à la construction de l’argumentation en français. Dans le genre discursif que nous avons étudié, elles ont en commun l’aptitude à construire un topique, et elles procèdent toutes les deux d’une simulation : le recours à autrui est fictif dans l’interrogation, l’assertion est fictive dans l’hypothèse. La question est plus difficile à envisager sans véritable acte de questionnement en anglais. La traduction nous a permis d’établir qu’en dépit de leur caractère polysémique commun, if et si ne sont pas employés dans les mêmes conditions. If nécessite une véritable hypothèse sous-jacente, alors que si peut construire une assertion fictive en faisant abstraction de la réalité des faits dans l’unique but de valoriser l’argument 106 AGNÈS CELLE exprimé dans l’apodose. C’est peut-être ce qui explique en partie que les textes argumentatifs français passent pour abstraits, alors que les mêmes textes en anglais sont perçus comme plus directs. Corpus ICE-GB, The International Corpus of English. University College London 2006. Barthes, R. (1953), Le degré zéro de l’écriture. Paris, Editions du Seuil. Traduction : Lavers, A. & C. Smith (1967): Writing Degree Zero. New York, The Noonday Press. Chomsky, N. (1965), Aspects of the Theory of Syntax. Cambridge, The Massachussets Institute of Technology. Traduction : Milner, J.-C. (1971) : Aspects de la théorie syntaxique. Paris, Editions du Seuil. Chomsky, N. (1968), Language and Mind. New York, Harcourt, Brace, Jovanovich. Traduction : Calvet, L.-J. (1969) : Le langage et la pensée. Paris, Payot. Foucault, M. (1961), Histoire de la folie à l’âge classique. Paris, Plon. Traduction : Howard, R. (1965) : Madness and Civilization, A History of Insanity in the Age of Reason. New York, Random House. Levi-Strauss, C. (1958), Anthropologie structurale. Paris, Plon. Traduction : Jacobson, C. & Grundfest Schoepf, B., (1963), Structural Anthropology. New York, Basic Books. Ritter, R. The Duty of Memory: The Nanjing Massacre, Memory and Forgetting in China and Japan. Traduction: Poncharal, B. Devoir de mémoire : le massacre de Nankin, mémoire et oubli en Chine et au Japon. Vingtième siècle : Revue d'histoire, n°94, 2007/2, Presses de Sciences Po. Yaguello, M. (1981), Alice au pays du langage, pour comprendre la linguistique. Paris, Seuil. Traduction : Harris, T. & Yaguello, M. (1998) : Language Through the Looking Glass, Exploring Language and Lin guistics. Oxford, Oxford University Press. Références bibliographiques Achard-Bayle, G. (2006), “Si Polysémique et si Polyphonique”. In L. Perrin (éd.), Le sens et ses voix, Recherches linguistiques 28. Metz, Université de Metz, 407-434. Adam, J.-M. (2005), “Variété des usages de si dans l’argumentation publicitaire“. In M. Burger & G. Martel (éds), Argumentations et communications dans les médias. Québec, Nota Bene, 81-109. Akatsuka, N. (1986), “Conditionals are discourse-bound”. In E. Closs Traugott, A. ter Meulen, J. Snitzer Reilly & Ch. A. Ferguson (éds), On Conditionals. Cambridge, Cambridge University Press, 333-351. Béguelin, M.-J. & G. Corminboeuf (2005), “De la question à l'hypothèse: étude d'un phénomène de coalescence”. In C. Rossari et al. (éds), Les États de la question, actes du Colloque de Fribourg, mai 2003. Québec, Editions Nota Bene, 67-89. Carter-Thomas, S. (2007),”The ‘iffiness’ of medical research articles, a comparison of English if and French si”. In K. Fløttum (éd), Language and Discipline Perspectives on Academic Discourse. Newcastle, Cambridge Scholars Publishing, 150-176. Celle, A. (2007), “Analyse unifiée du conditionnel de non prise en charge en français et comparaison avec l’anglais”. In L. de Saussure, J. Moeschler et G. Puskas (éds), Cahiers Chronos n°19, Etudes sémantiques et pragmatiques sur le temps, l'aspect et la modalité. Amsterdam, Rodopi, 43-61. Celle, A. (2009), “Question, mise en question : la traduction de l’interrogation dans le discours théorique”. Revue française de linguistique appliquée, Linguistique et Traduction, XIV-1, 39-52. Charolles, M. (2003), “De la topicalité des adverbiaux détachés en tête de phrase”. In M. Charolles & S. Prévost (éds), Adverbiaux et topiques. Louvain la Neuve, Travaux de Linguistique, 47, 11-51. Chuquet, J. (1994), “Forme interrogative et hypothèse en anglais contemporain”. In : P. Boucher & J.-M. Fournier (éds), Interrogation, des marques aux actes, Travaux du CERLICO 7-8. Presses universitaires de Rennes, 213-236. Dancygier, B. & E. Sweetser (2005), Mental Spaces in Grammar, Conditional Constructions. Cambridge, Cambridge University Press. Declerck R. & S. Reed (2001), Conditionals, A Comprehensive Empirical Analysis. Berlin, Mouton de Gruyter. De Cornulier, B. (1986), Effets de sens. Paris, Editions de Minuit. De Vogüé, S. (1999), “Le champ des subordonnées dites conditionnelles du français: conditions, éventualités, suppositions et hypothèses”. L’Hypothétique. LINX n°41, 93-118. Fløttum, K., T. Dahl, et T. Kinn (2006), Academic Voices Across Languages and Disciplines. Collection Pragmatics and Beyond New Series, Amsterdam / Philadelphia, John Benjamins. TRANSLATION AND MEANING, PART 10, 2013, 93 - 107 107 Ford, C. & S. Thompson (1986), “Conditionals in discourse, a text-based study from English”. In E. Closs Traugott, A. ter Meulen, J. Snitzer Reilly & Ch. A. Ferguson (éds), On Conditionals. Cambridge, Cambridge University Press, 353-372. Haiman, J. (1978), “Conditionals are topics”. Language 54: 564-89. Haiman, J. (1986), “Constraints on the form and meaning of the protasis”. In: E. Closs Traugott, A. ter Meulen, J. Snitzer Reilly & Ch. A. Ferguson (éds), On Conditionals. Cambridge, Cambridge University Press, 215-227. Halliday, M.A.K. (1994), An Introduction to Functional Grammar. London, Edward Arnold. Hyland, K. (2005) : Metadiscourse, Exploring Interaction in Writing. London, New York, Continuum. Grésillon, A. (1981), “Interrogation et interlocution”. DRLAV n°25, Dans le champ pragmatico énonciatif, 6175. Guillemin-Flescher, J. (2003), “Théoriser la traduction”. Revue Française de Linguistique Appliquée, VIII-2, 718. Guillemin-Flescher, J. (2006), “Traduction et fonctionnement du langage“. In D. Ducard & C. Normand (éds), Antoine Culioli, un homme dans le langage. Paris, Ophrys, 249-266. Huddleston, R. (1994), “The Contrast between Interrogatives and Questions”. Journal of Linguistics, Vol. 30, No. 2 (Sep., 1994), 411-439. Jespersen, O. (1940), A modern English Grammar on historical principles, vol. 5: Syntax. London, George Allen and Unwin. Kerbrat-Orecchioni, C. (1991), la Question, Presses Universitaires de Lyon. Lewis, D. (2005), “Mapping adversative coherence relations in English and French” Languages in Contrast, vol.5 Issue 1, 2004/2005. 33-48. Lyons, J. (1977), Semantics, vol.II. Cambridge, Cambridge University Press. Paillard, M. (1993), “Les chemins de la concession: quelques contrastes entre anglais et français”. In : C. Muller & D. Roulland, Subordinations, Subordination, Travaux linguistiques du CERLICO n°6. Rennes, Presses universitaires de Rennes, 207-226. Quirk, R., S. Greenbaum, G. Leech & J. Svartvik (1985), A Comprehensive Grammar of the English Language. Londres, Longman. Ranger, G. (1998), Les constructions concessives en anglais: une approche énonciative. Paris: Ophrys. Rowley-Jolivet, E. (2007), “A Genre Study of If in Medical Discourse”. In : K. Fløttum (éd.), Language and Discipline Perspectives on Academic Discourse. Newcastle, Cambridge Scholars Publishing, 176-201. Rowley-Jolivet, E. & S. Carter-Thomas (2008), “When practice belies ‘theory’: Form, function and frequency of if-conditionals in specialised discourse”. ASp [En ligne], 53-54, mis en ligne le 01/11/2011, consulté le 16/10/2010. URL : http://asp.revues.org/343. 108 TRANSLATION AND MEANING, PART 10, 2013, 109 - 119 109 TRADUIRE DES JEUX DE MOTS DU FRANÇAIS VERS LE POLONAIS. EXEMPLE DE RENE GOSCINNY Annick Cataldi Université de Varsovie, Warszawa, Poland Abstract: Any act of translation is intended to allow the reader access to a text written in another language than his vernacular. To do so, the translator will try to clearly and accurately transmit the whole message contained in the original text. This accurate transmission requires not only excellent linguistic skills in both languages but also very good knowledge of the specific cultural input of both countries. However, in certain kinds of literary text, where figures of speech, such as word play, play a key role in the overall understanding of the text, the translator will often encounter in the target language the great difficulty to render these fixed formulas from the source language, while being faithful to the message conveyed by the author. Our analysis will focus on the study of the translation, from French to Polish, of word play and idioms existing in several albums of two series of comics by René Goscinny, namely the series of Asterix and Little Nicholas. It is obvious that these so typically French albums – where word play is abundant and is the backbone of all the text, and where humour is frequently linguistic – will give the translator a real linguistic challenge. After analyzing a number of examples of word play and its translation into Polish, we will try to identify the means used by the translator to build his text as he wanted. Then we can ask ourselves if in this case we are dealing with translation or adaptation. It will lead us to question about the limits of freedom of the translator and the (im)possibility to convey, in a non-French text, the whole socio-cultural content of the language of Goscinny’s albums. 1. Introduction Les figures de style sont souvent considérées comme des problèmes délicats de traduction. Si certaines d’entre elles sont relativement faciles à transposer, les jeux de mots, eux, demeurent une pierre d’achoppement pour tout traducteur. Nous nous trouvons face à la question: ces jeux de langue, choisis intentionnellement par l’auteur pour véhiculer son message, doivent-ils obligatoirement apparaître en langue cible? Nous entrons là dans la bouillante polémique de la traduisibilité ou de l’intraduisibilité des jeux de mots. Nous appuyerons notre étude sur la traduction des jeux de mots du français vers le polonais dans les textes de René Goscinny en suivant deux de ses héros favoris: le petit Nicolas et Astérix, à travers Le petit Nicolas1, Les récrés du petit Nicolas2, Le tour de Gaule d’Astérix3, Astérix en Corse4 et Astérix et les Normands5. Les aventures de ces personnages sont mondialement connues et il nous paraît intéressant de nous arrêter sur ces textes où les jeux de mots abondent et constituent comme la charpente de toute la construction de l’auteur. 1 Sempé-Goscinny, (1960), Le petit Nicolas, Paris, Editions Denoël. Sempé-Goscinny, (1961), Les récrés du petit Nicolas, Paris, Editions Denoël. 3 Goscinny, R. (1965), Le tour de Gaule d’Astérix, Paris, Dargaud Editeur. 4 Goscinny, R. (1973), Astérix en Corse, Paris, Goscinny-Uderzo. 5 Goscinny, R. (1967), Astérix et les Normands, Paris, Goscinny-Uderzo. 2 110 ANNICK CATALDI Nous pourrons alors nous interroger sur la traduction des textes de Goscinny et nous demander si, dans ces cas précis, nous sommes en présence de traduction ou d’adaptation ; il nous restera à nous poser la question de la limite de la liberté du traducteur. Pour ce faire, nous allons d’abord répertorier les différentes formes de jeux de mots qui apparaissent chez Goscinny puis, définir leurs fonctions dans le texte source, pour parvenir à l’analyse des traductions de ces jeux de mots telles qu’elles sont proposées dans le texte en polonais. 2. Description des jeux de mots Avant de décrire et de classer les jeux de mots, il convient de préciser ce que recouvre cette figure de style. Pour que l’on puisse parler de jeu, au sens général du terme, il faut qu’il y ait une activité purement gratuite qui procure du plaisir. Or, qu’est-ce qu’un jeu de mot sinon ce plaisir tout à fait gratuit que l’auteur s’offre - et nous offre. Pour partager ce plaisir gratuit que nous propose Goscinny, nous allons adopter la classification établie par Jakobson6 dans sa distinction entre l’axe syntagmatique et l’axe paradigmatique, à savoir, la substitution et l’enchaînement. A ces deux types de classement nous ajouterons celui que propose Pierre Guiraud7 : l’inclusion. 2.1 La substitution La substitution ou l’équivoque représente la forme de jeux de mots la plus fréquente en français. Il s’agit de remplacer un son ou un mot par un autre qui lui ressemble étrangement afin d’obtenir une allusion plaisante8. Nous étudierons deux types de substitution : le calembour et l’à-peu-près. 2.1.1 Le calembour La notion de calembour est liée à celle de plurivocité, l’auteur joue intentionnellement sur les sens pluriels du mot. Cette plurivocité peut être soit phonique, soit sémique, soit, plus fréquemment dans les albums d’Astérix, polysémique avec, en particulier, de nombreux calembours complexes ou de multiples jeux lexicaux portant sur les noms propres de personnes ou de lieux. 2.1.1.1 L’homonymie Ce jeu peut porter sur un mot, tel le mot « brut » qui, par sa nature grammaticale peut être soit adjectif soit substantif. Durant leur Tour de Gaule (p.18) Astérix et Obélix parviennent à Reims et là ils découvrent que le Champagne peut être brut, sec ou doux. Astérix, voulant frapper un Romain demande à son ami de lui passer une amphore. « -Brut ? demande Obélix, - Brut. » , répondra Astérix empressé d’attaquer son ennemi, et le Romain, frappé par le coup répliquera aux Gaulois «- Brutes!» 2.1.1.2 L’homophonie Sous cette catégorie nous présenterons un exemple difficile à classer puisqu’il s’agit du nom « un grognard » pris dans le sens de « soldat de la garde napoléonienne » et de son homophone, substantif inventé par Goscinny pour parler des chefs Corses qui communiquent d’un village à l’autre en poussant des grognements semblables à ceux des sangliers. Bien que l’orthographe des deux mots soit identique, nous considérons cette figure comme une homophonie - et non homonymie - dans la mesure où l’un des substantifs est inventé par l’auteur. Dans Astérix en Corse, le petit chef Corse, sous 6 Jakobson, R. (1970), Essais de linguistique générale, Paris, Editions du Seuil. Guiraud, P. (1979), Les jeux de mots, Presses Universitaires de France. 8 Suhamy, H. (2006), Les figures de style, Paris, Presses Universitaires de France. 7 TRANSLATION AND MEANING, PART 10, 2013, 109 - 119 111 lequel le lecteur a vite fait de reconnaître Napoléon, voyant ses troupes arriver en poussant leurs grognements se félicitera en disant : « ils sont tous là mes Grognards... » (p. 38). Magnifique jeu de mots. 2.1.1.3 La paronymie Ce type, plutôt simpliste, de jeu de mots, qui concerne deux mots phonétiquement très proches, apparaît assez rarement chez Goscinny. Dans Astérix et les Normands (p.45) nous remarquons une paronymie si discrète qu’elle est presque imperceptible et, en ce trait réside son charme. Le chef Normand propose à Astérix de lui offrir un festin pour le remercier, proposition à laquelle Astérix répliquera : « Votre départ nous suffit. Partir, c’est nourrir un peu ». Paronymie de la célèbre phrase d’Allais : partir c’est mourir un peu. 2.1.1.4 La synonymie Nous présenterons–là deux cas de synonymies, le premier, simple, consiste à remplacer un mot par un autre de sens voisin, le second concernera des phrases de type proverbial. Dans Les récrés du petit Nicolas (p.98) le surveillant annonce aux élèves qu’ils vont déposer une gerbe devant un monument. « -C’est quoi, une gerbe ? demande Rufus. - C’est un bouquet, répond Agnan » puis une dispute entre deux élèves s’engage, le surveillant intervient et donne une retenue à chacun des belligérants «- Ça, c’est le bouquet, riposte le premier élève. –C’est la gerbe » ajoutera un enfant. Nous avons là, tout d’abord, une synonymie simple « gerbe/bouquet » à laquelle fait suite une expression familière « c’est le bouquet » reprise par un synonyme insolite. Le second cas concerne la synonymie dans des phrases de type proverbial telles : « C’est la goutte qui fait déborder l’amphore » (Astérix et les Normands, p.41) ou encore une double synonymie : « Si..si.. Avec des si on mettrait Lutèce en amphore. » Le tour de Gaule (p.28). On note ici que le mot « amphore » remplace successivement « vase » et « bouteille ». 2.1.1.5 La polysémie La polysémie pourra, selon les cas, être explicite ou implicite, parfois évidente d’autres fois plus subtile. Nous sommes à nouveau en Corse, auprès des vieillards commentant la violente bataille que les enfants du pays livrent aux Romains (Astérix en Corse, p.43). « -Tiens, Salamix est de retour » dit le premier vieillard, «- ça a l’air d’étonner ses copains romains » ajoute le second, « -ça les frappe » réplique le troisième, alors que sur la même vignette la violence de la bataille est signalée par des « Paf ! Bong ! Tchong ! ». Polysémie classique entre les verbes «étonner » et « frapper » . Mais, la polysémie la plus subtile chez Goscinny concerne les noms propres. D’une façon générale, l’auteur laisse le lecteur découvrir lui-même l’astuce qui se cache derrière la plupart des noms propres, toutefois, dans Le petit Nicolas, vu que le livre s’adresse à de jeunes lecteurs, Goscinny va décoder le nom du surveillant (Le petit Nicolas, p.22 et 148). « Le Bouillon, on l’appelle comme ça parce qu’il dit toujours: « Regardez-moi dans les yeux », et comme dans le bouillon il y a des yeux, on l’appelle le Bouillon. » Décryptage explicite du nom d’un des principaux personnages du monde du petit Nicolas. Les albums d’Astérix s’adressant en priorité à des adultes, le lecteur doit lui-même chercher l’astuce derrière des noms de Romains comme Faismoiducouscous ou Fleurdelotus ou, derrière des noms de Gaulois comme Quatredeusix, Labeldecadix, Plaintcontrix ou Squinotix. 112 ANNICK CATALDI Les noms propres de lieux ne sont pas, non plus, laissés au hasard, nous ne signalerons que quelques noms de camps romains bordant le littoral corse : Mercurocrum, Ouelcum, Chouingum ou Jolimum. 2.1.1.6 Les calembours complexes Nous distinguerons ici deux catégories : les calembours complexes correspondant à un double jeu de mots, et ceux plus subtils, qui nécessitent obligatoirement l’intervention de l’intelligence et de la culture du lecteur afin de saisir tout l’humour voulu par l’auteur. Ces calembours complexes recouvrent très précisément les notions de plaisir et de gratuité de l’écriture dont nous avons parlé plus haut. Nous ne donnerons que quelques exemples, allant du plus simple au plus complexe. Tout d’abord, un nom propre de personne : Ocatarinetabellatchitchix, le Corse (Astérix en Corse, p.9 et 14). Son nom nous rappelle un autre Corse, Tino Rossi, célèbre pour l’une de ses chansons populaire durant l’entre-deux guerres, Tchi-tchi, dont les paroles débutaient par ces mots : « O Catarinetta bella, tchi-tchi ». Certains calembours portent, eux, sur des expressions, ainsi, le chef Normand, annonce à ses hommes: « Nous allons établir notre camp. Creuser des trous pour planter des piquets. De beaux trous normands. » (Astérix et les Normands, p.14). L’humour ineffable de ce jeu de mot culturel peut passer inaperçu à celui qui ignore que « le trou normand » est le nec plus ultra de la gastronomie française. 2.1.2 L’à-peu-près L’à-peu-près se distingue du calembour par le fait que l’équivoque, pas toujours très cohérente, a parfois des traits un peu approximatifs. Nous distinguerons ici trois cas d’à-peu-près: l’à-peu-près phonétique, le pataquès et les mots étrangers. 2.1.2.1 L’à-peu-près phonétique Nous citerons un cas d’homophonie partielle et approximative qui correspond tout à fait à la définition de l’à-peu-près. Dans Astérix et les Normands (p.18), le Chef Gaulois s’adresse au jeune Lutécien par ces mots: « - ton Père m’avait demandé de te faire acquérir des qualités d’homme, de t’aguerrir ! » et le jeune Lutécien réplique : « acquérir... aguerrir... équarrir, oui ! » faisant ainsi allusion aux Normands qui pourraient le découper en quartiers. L’à-peu-près phonétique est, pourrait-on dire, présent au sens propre lorsque Goscinny s’amuse à faire parler chacun avec son accent régional ou national. Fabuleuses pages d’Astérix chez les Corses (p. 9, 11, 13) où de nombreux personnages, rencontrés dans d’autres albums, se retrouvent pour une Fête. Nous nous arrêterons seulement aux Arvernes : « -Chalut ! Je chuis chinchèrement ravi de te revoir ! » dira l’Auvergnat et, son épouse, discutant chiffon avec son amie Lyonnaise, s’exclamera : «- Cha ch’est un beau tichu, cha ! –Oh, vous savez c’est de la soie de chez nous, de Lugdunum –De la choie ! vous en avez de la chance ! » Comme on le voit, l’à-peu-près recouvre à la fois le burlesque et le farfelu et c’est là sa fonction. 2.1.2.2 Le pataquès ou lapsus volontaire Les meilleurs pataquès rencontrés chez Goscinny sont des paroles de chansons légèrement modifiées par le barde. Nous en citerons deux, tirés de l’album Astérix et les Normands (p.8, p.43). Nous sommes en Armorique, personne donc ne s’étonnera des paroles chantées par le barde « le folklore armoricain », or cet album d’Astérix a été écrit au moment où une chanson de Sheila faisait furie en France: « le folklore américain ». Plus loin, lors d’un autre concert, le barde entonnera : « Ma mère m’a dit: Assurancetourix, fais-toi tresser les cheveux, oh oui ! ». Magnifique pataquès de la chanson d’Antoine, célèbre alors: « Ma mère m’a dit : Antoine fais-toi couper les cheveux, oh yé ! » TRANSLATION AND MEANING, PART 10, 2013, 109 - 119 113 2.1.2.3 Les mots étrangers L’équivoque au niveau des mots étrangers peut être lexicale ou syntaxique. Un bon exemple d’équivoque syntaxique nous est donné par les habitants de Grande Bretagne, qui ont des difficultés à manier les structures grammaticales françaises et calquent leurs propres structures de langue. Nous lisons, par exemple : « Je dis ! N’est-il pas merveilleux de nous rencontrer.(...) Je dis, cette magique potion est si surprenante, n’est-il pas ? » ( Astérix en Corse, p. 9, 13). 2.2 L’enchaînement Certains mots ont un rapport mutuel voulu par l’auteur dans un but ludique. Comme la substitution, l’enchaînement peut être aussi bien phonique que sémique. Nous en distinguerons trois formes: l’enchaînement par homophonie, par écho et par automatisme. 2.2.1 L’enchaînement par homophonie Goscinny va combiner entre eux des termes à similarité phonétique dans un but purement ludique. Dans Astérix et les Normands, cette peuplade dont tous les noms, chez Goscinny, se terminent en « af », un guerrier explique comment il a assommé le Lutécien: «- un coup de massue et paf ! », on entend, en écho, une voix qui répond: « -Présent ! », suit alors la réplique : « -Mais non, Epaf ! Personne ne t’a appelé. » (p.21). Pur enchaînement homophonique, gratuit, vu qu’il est difficile de trouver un lien sémantique entre une onomatopée et un nom propre de personne. Dans Le tour de Gaule, suite à l’annonce de l’attaque qu’ils vont aller mener contre le fameux village gaulois, nous voyons les soldats romains faire la queue devant l’infirmerie de leur camp, ce qui déclenche la furie du Centurion : « -Vous refusez de porter pilum ?! –Ben... on préfère se faire porter pâles... », enchaînement phonétique appuyé par l’allitération du « p ». (p.5). 2.2.2 L’enchaînement par écho Cette forme d’enchaînement est la plus utilisée par Goscinny. A plusieurs reprises nous retrouvons des interjections adressées aux Dieux de chaque peuplade et se terminant en écho. Ainsi, nous entendons, côté romain « Par Jupiter ! Par Toutatis ! Par pitié ! » (Le tour de Gaule, p.7) ou encore, côté normand « Par Thor ! Par Odin ! Partons ! » (Astérix et les Normands, p.43). 2.2.3 L’enchaînement par automatisme Ici, les exemples ne seront pas des enchaînements par automatisme mais plutôt des enchaînements par anti-automatisme. Nous voyons les soldats Romains quitter la bataille en disant : «- On rentre. –On prend congé. –On file à la bretonne. » (Astérix et les Normands, p.29) alors que l’on attendait « filer à l’anglaise ». Ou encore, une affiche placardée à Bordeaux nous informe que « Le Général Motus prendra la parole » or chacun connaît l’expression « Motus et bouche cousue ». Il semble que Goscinny se plaise à rompre la chaîne des automatismes en créant ainsi un effet de surprise beaucoup plus ludique. 2.3 L’inclusion Le troisième type de jeux de mots rencontré est l’inclusion, ce nouvel arrangement phonique ou graphique à l’intérieur d’un mot ou d’une phrase. Chez Goscinny la forme d’inclusion la plus fréquente est la permutation, en particulier l’antimétabole, qui permute ou substitue un terme inattendu dans une expression figée. Nous donnerons deux exemples qui illustrent très bien ce qu’est l’antimétabole : « J’ai plus de mal que de peur » dira un Normand (Astérix et les Normands, p.9), ou encore, « La garde se rend et ne meurt pas » (Astérix en Corse, p.43), permutation de la célèbre phrase qui aurait été dite par Cambronne « La garde meurt mais ne se rend pas ». 114 ANNICK CATALDI La description de ces divers jeux de mots peut nous amener à une première conclusion : Goscinny maîtrise tout à fait son art et, s’il utilise davantage telle forme plutôt que telle autre, c’est pour servir son but qui est de divertir, voire d’amuser le lecteur. Nous allons à présent étudier les différentes fonctions de ces jeux de mots et considérer dans quelles mesures elles recoupent cette première conclusion. 3. Fonctions du jeu de mots Si nous regardons l’étude extrêmement riche menée par Jacqueline Henry9 sur la traduction des jeux de mots, nous constatons que pour elle, il est capital de cerner correctement les fonctions des jeux de mots afin de mieux saisir le projet de l’auteur. Jacqueline Henry souligne que ce travail de réflexion doit précéder la traduction et que la méconnaissance de ces fonctions peut « entraîner un a priori d’intraduisibilité des jeux de mots » (2003, p.31). Avant de se mettre à la traduction, il s’agit donc de distinguer le sens de la forme, de voir le pourquoi de l’emploi du jeu de mots et le vouloir-dire de l’auteur à travers cette figure. Nous n’allons pas analyser toutes les fonctions du langage décrites par Jakobson, nous nous contenterons d’en souligner trois: la fonction métalinguistique, la fonction littéraire ou poétique et la fonction ludique. Nous voulons par là essayer d’établir le rôle que le jeu de mot joue dans l’ensemble du texte de Goscinny et en déduire l’effet recherché par l’auteur en utilisant largement cette forme stylistique. 3.1 Fonction métalinguistique Cette fonction nous paraît capitale pour notre thème. Dans son travail intitulé Traduction des jeux de mots, Barbara Walkiewicz10 considère ce référent métalinguistique comme primordial, selon l’auteur, il serait « l’unité linguistique soumise au parasitage modificateur » (2005, p.69). Il semble, en effet, que Goscinny utilise le langage pour s’interroger sur le sens du langage lui-même. Nous citerons un exemple portant sur un mot. Le Capitaine des Pirates annonce sa tactique à ses mousses : « Une fois au large, on les passe par-dessus bord ! Nouvelle technique : ce n’est plus l’abordage mais le débordage ! » (Astérix en Corse, p. 18). Nous nous trouvons là devant un cas d’utilisation phonique du métalangage avec enchaînement par écho, le mot « abordage » ayant appelé phonétiquement le mot « débordage » créé par Goscinny. Pierre Guiraud (1979, p.112) parlera du jeu de mots comme d’une « défonctionalisation de l’activité linguistique normale» et Jacqueline Henry ira encore plus loin en décrivant le jeu verbal comme « une révolte contre le stéréotype » (2003, p.41). 3.2 Fonction littéraire Comme on le dit souvent, la littérature est « un art du langage ». Avec les jeux de mots, nous sommes dans un domaine littéraire particulier qui se rapprocherait davantage de la poésie. En effet, ici les mots expriment l’idée avec un maximum d’efficacité, nous avons parfois l’impression que Goscinny cherche des idées afin d’utiliser les jeux de mots auxquels il a déjà pensé et non l’inverse. Nous avons donc, comme en poésie, une inversion de la fonction linguistique, les mots précèdent les idées, pourrait-on dire. Nous illustrerons cette pensée par l’un des thèmes d’Astérix et les Normands : la peur. On dit communément que « la peur donne des ailes » et il semble qu’autour de ces mots « peur » et « ailes » Goscinny ait bâti nombre d’anecdotes de l’album et non l’inverse. 9 Henry, J. (2003), La traduction des jeux de mots, Paris, Presses Sorbonne Nouvelle. Walkiewicz, B. (2005), Traduction des jeux de mots. Aspects sémantiques, Łask, Wydawnictwo Leksem. 10 TRANSLATION AND MEANING, PART 10, 2013, 109 - 119 115 La fonction poétique du jeu de mots est de permettre de communiquer toutes sortes d’ émotions. La meilleure illustration en est l’ensemble de l’épisode concernant la Provence dans Le tour de Gaule. Nous y retrouvons, en sous-entendu, la Trilogie de Marcel Pagnol11 avec la partie de cartes et ses personnages célèbres, même le Lyonnais qui a du mal à s’intégrer est là, grâce à la poissonnière résonne à nos oreilles le parler marseillais avec son exagération reconnue et son accent chaleureux : «Voyez mes oursins, qu’ils sont tellement bons que c‘est même un plaisir de s’asseoir dessusse ! » et enfin nous participons à l’intense émotion qui entoure toute partie de pétanque : « Mais tirez, pécaïre ! Que je m’use la salive à vous le dire ! » (p.31, 32). Comment, après cette lecture, oserait-on penser qu’il n’y a pas un art du langage chez Goscinny ? 3.3 Fonction ludique Il est clair, à travers tous les exemples vus précédemment, que la fonction ludique est primordiale dans les textes de Goscinny. Régulièrement nous notons cette gratuité du jeu de mots et ce plaisir qu’il apporte au lecteur. Mais, si nous regardons ces jeux de mots de plus près, nous constatons une grande différence entre les livres du petit Nicolas et les albums d’Astérix. Dans le premier, la fonction ludique tient au fait que nous voyons le monde des adultes à travers le regard et les mots d’un enfant. La difficulté du traducteur va donc être essentiellement d’ordre lexical. Dans les albums d’Astérix il en est tout autrement. Goscinny s’adresse là à un public très large, l’enfant s’amusera de certains détails facilement repérables mais, l’adulte y trouvera un plaisir de type intellectuel, il se délectera en découvrant la somme culturelle cachée sous les mots, en faisant, pourrait-on dire, une lecture au second degré. Ce type de textes exige de la part du traducteur non seulement de grandes qualités linguistiques dans les deux langues, mais encore des connaissances intellectuelles et culturelles fort larges. Une des difficultés est liée au fait que, souvent, dans les albums d’Astérix, le jeu ne porte pas sur le signifiant (le mot) mais sur le signifié (l’idée). Pour illustrer cette pensée nous pouvons citer l’exemple vu plus haut, lorsque le Normand dit à ses subalternes : « Creuser des trous, de beaux trous normands ». A notre tour nous devons creuser le texte pour essayer d’en saisir toutes les allusions proposées par l’auteur afin de les transmettre au mieux en respectant la même tonalité d’ensemble. Nous touchons là au point clé de notre étude, en effet, il convient de se demander si, avec Goscinny et en particulier dans le texte de ses albums d’Astérix, nous sommes en présence de jeux de mots, ou plutôt, de jeux d’esprit. L’auteur ne joue-t-il pas davantage sur les idées que sur les mots ? Jeu d’idées, jeu de mots, la différence peut paraître subtile aux yeux du lecteur mais, elle est capitale pour le traducteur. Et cela nous amène à notre dernière partie qui concerne directement la traduction des jeux de mots dans les textes de Goscinny. 4. Traduction des jeux de mots Après avoir décrit les jeux de mots et avoir souligné quelles étaient leurs fonctions principales nous voyons plus clairement ce qui constitue la trame des textes à traduire et la tonalité recherchée par Goscinny. Nous avons aussi constaté que les livres du petit Nicolas présentaient une difficulté de traduction d’ordre essentiellement lexical et syntaxique alors que les albums d’Astérix, qui offrent un texte construit autour de jeux de mots et de mots d’esprit, exigeaient de nombreuses compétences du traducteur. De ce fait, nous considérerons, ici, seulement la traduction des albums d’Astérix. Notre travail se limitera aux textes du français vers le polonais. Nous utiliserons les traductions de Jolanta Sztuczynska pour les trois albums étudiés: Astérix et les Normands12, Astérix en Corse13 et Le 11 Marius, Fanny et César. Goscinny, R. (1992), Asteriks i Normanowie, Przekład Jolanta Sztuczyńska, Warszawa, Egmont Poland. 13 Goscinny, R. (1994), Asteriks na Korsyce, Przekład Jolanta Sztuczyńska, Warszawa, Egmont Polska. 12 116 ANNICK CATALDI tour de Gaule d’Astérix14, pour ce dernier album nous nous pencherons aussi sur la traduction faite par Jaroslaw Kilian15. Nous allons répertorier les diverses stratégies de traduction en prenant la typologie proposée par Jacqueline Henry, à savoir les traductions isomorphe, homomorphe, hétéromorphe et la traduction libre. 4.1 Traduction isomorphe La traduction isomorphe consistant à employer un jeu de mots identique dans les deux langues apparaîtra rarement dans les textes qui nous intéressent vu que les deux langues ont une origine plutôt éloignée. Toutefois, nous en relèverons quelques cas, essentiellement dans les noms propres ou dans les mots empruntés. L’album Astérix en Corse débute par une carte de l’île mentionnant toutes les villes ou camps romains qui bordent la côte. Les noms de camps se terminent tous par la désinence latine –um. On constate que chaque fois que le mot latin utilisé dans la version française fonctionne dans la langue polonaise, il est conservé, quitte à apporter une correction orthographique si nécessaire, ainsi des noms comme Geranium, Podium, Sternum, ou Ultimatum sont absolument identiques alors que d’autres vont être polonisés au niveau de la graphie tels Saeculasaeculorum/Sekulasekulorum, Desideratum/Dezyderatum ou Maximum/Maksimum. Certains noms propres de personnes restent inchangés, en particulier celui du héros corse Ocatarinetabellatchitchix/Okatarinetabellaczikcziks, mais nous pouvons nous demander si, pour le lecteur polonais, ce nom a un sens. Notre dernier exemple de traduction isomorphe sera un emprunt avec cet enchaînement par écho vu plus haut : « Ce n’est plus l’abordage mais le débordage ». Etant donné que le mot « abordaż » existe en polonais, Sztuczynska propose ce jeu de mots à l’ identique : « To już nie abordaż, a debordaż ! » (Asteriks na Korsyce, p.18). Là encore, reste à savoir si ce jeu est clair pour le lecteur polonais qui ne connaît pas le mot « déborder ». 4.2 Traduction homomorphe Utilisant, en langue cible, un jeu de mots qui repose sur le même procédé que celui utilisé en langue source, d’une façon générale, cette forme de traduction est assez souvent employée par les traducteurs. De fait, notre étude confirme cette règle. La traduction homomorphe, va servir à transmettre nombre d’informations contenues dans les noms propres de lieux et de personnes, de fait le texte de Goscinny devient, en langue cible, plus riche et plus ludique que dans le cas de la traduction isomorphe. Ainsi, la ville corse de Mercurocrum se transforme en Remedium et celle de Chouingum en Gumoleum, de même, le surveillant Le Bouillon est substitué par Rosół16 (Rekreacje Mikołajka, p.24) et, Faismoiducouscous par Zróbmikuskus (Wyprawa Asteriksa dookoła Galii, Przekład Sztuczyńska, p. 12). Plus loin, à Bordeaux, un placard affiché dans la ville nous informe que « le Général Motus prendra la parole », il deviendra « Mowus » /qui aime parler17/ chez Sztuczyńska (p.41) et « Zamotus » /qui a un parler ambigu18/ chez Kilian 14 Goscinny, R. (1994), Wyprawa Asteriksa dookoła Galii, Przekład Jolanta Sztuczyńska, Warszawa, Egmont Polska. 15 Goscinny, R. (1998), Wyprawa dookoła Galii, Przekład Jarosław Kilian, Warszawa, Egmont. 16 Sempé-Goscinny, (1991), Rekreacje Mikołajka, Przełożyły Tola Markuszewicz i Elżbieta Staniszkis, Warszawa, Nasza Księgarnia. 17 Notre traduction. TRANSLATION AND MEANING, PART 10, 2013, 109 - 119 117 (p.49). Grâce à cette forme de traduction homomorphe, le lecteur polonais entre plus facilement dans la tonalité ludique du texte et va pouvoir, lui aussi, s’amuser avec les divers noms propres. Des expressions figées subiront le même processus de transformation, ainsi le « filer à la bretonne » pour « filer à l’anglaise » va devenir chez Sztuczynska « wynosimy się po brytyjsku » (Asteriks i Normanowie, p.29), ce qui est aussi plaisant. La même forme de traduction va être appliquée aux paroles de la chanson « Ma mère m’a dit Assurancetourix va te faire tresser les cheveux » qui sont remplacées par les paroles du tube polonais de l’époque qui, de plus, fait allusion aux cheveux d’une jeune fille: « Była Blondynka, ten kolor włosów tak zwą sprawiła...» (Asteriks i Normanowie, p.43). Excellente adaptation de type homomorphe. 4.3 Traduction hétéromorphe Cette forme de traduction, où le jeu de mot utilisé repose sur un procédé autre que celui de l’auteur, laisse au traducteur une assez grande liberté. Nous citerons trois cas, pris chacun dans les deux traductions du Tour de Gaule d’Astérix, ce qui va nous permettre de comparer les propositions des deux traducteurs polonais. Nous avons vu plus haut le dialogue: «- Vous refusez de porter pilum ? –On préfère se faire porter pâles. » (p.5). Sztuczynska propose: « -Odmawiacie noszenia oszczepów ? –Tak... wolimy obnosić swoje choroby.. » (p.5), ce qui signifie « -Vous refusez de porter pilum ? -Oui, nous préférons faire étalage de notre maladie »19. Comme dans le texte en langue source, le jeu de mots est lié à la forme verbale. La proposition de Kilian est différente : «-Odmawiacie walki oszczepem ? –O szczepieniach ochronnych pamiętać musi każdy legionista» (p.13), «Vous refusez la lutte au pilum ? -Se faire vacciner doit être le souci de chaque légionnaire »20. Ici, le jeu de mots se fonde sur la forme nominale pilum/oszczep qui, phonétiquement, rappelle un des cas de la déclinaison du mot vaccin « o szczepieniach ». Nous sommes un peu loin du texte source mais l’effet ludique est réel, ne faut-il pas qualifier cette adaptation de bonne? Notre second exemple reprend le jeu de mots cité plus haut à propos du Champagne. Astérix, voulant frapper un Romain demande à son ami de lui passer une amphore. « -Brut ? demande Obélix, Brut. » , répondra Astérix empressé d’attaquer son ennemi, et le Romain, frappé par le coup répliquera aux Gaulois « - Brutes ! » (p.18). Sztuczynska propose : « -Ciężkiego ? -Ciężkiego –Ciężkie bydlaki ! », ce qui correspond à «–Lourd ? –Lourd –Lourd bétail ! » 21. Kilian, à nouveau, offre une autre version : « -wytrawnego ? –wytrawnego –wytrawny strzał ! » (p. 26) «–Sec? -Sec –Coup Sec ! »22. Le dernier exemple hétéromorphe est la phrase de type proverbial : « Si, Si... Avec des si, on mettrait Lutèce en amphore. » (p.28). Sztuczynska écrit : « Gdyby, gdyby, z tyloma gdyby możnaby Lutecie zmieścić w amforze ! » ce qui est, à peu près, la traduction mot à mot du dicton français. Kilian s’écarte davantage de l’original pour proposer : « Gdyby, gdyby.. gdyby Babcia miała wąsy, to 18 Notre traduction. Notre traduction. 20 Notre traduction. 21 Notre traduction. 22 Notre traduction. 19 118 ANNICK CATALDI bybyła dziadkiem ! » (p.36). Là encore, nous sommes loin du texte source puisque cela correspond à « Si, si...si ma grand-mère avait des moustaches, ce serait mon grand-père »23. Toutefois, nous qualifierons cette adaptation de réussie car Kilian tient compte ici à la fois de la tonalité du texte et du destinataire polonais qui, après une succession de « si » pense spontanément à cette phrase. Nous constatons donc que la traduction hétéromorphe des jeux de mots non seulement est bien présente dans Astérix en polonais mais elle est bien vivante et particulièrement réussie dans le texte de Kilian et qu’elle garde le caractère ludique souhaité par Goscinny. 4.4 Traduction libre Par traduction libre nous entendons aussi bien les créations en polonais de jeux de mots absents en français que l’inverse. Nombre de jeux de mots n’apparaîssent pas dans le texte polonais. On peut s’étonner, par exemple, que le camp de Ouelcum se réduise à Odium ou que celui de Jolimum soit devenu Laura, ou que l’assonance rencontrée dans « acquérir, aguerrir et équarrir » sonne artificiellement en polonais: « wyrobić, zahartować, obrobić » (Asteriks i Normanowie p.18). On peut regretter aussi que l’expression « des trous normands » ait complètement disparu ou que le magnifique jeu de mots concernant « les grognards » soit passé inaperçu. Mais cette déperdition quantitative semble inévitable dans un tel texte où jeux de mots, jeux d’esprit et humour constituent la trame de l’ensemble. Il convient, par contre, de reconnaître que quelques éléments de traduction libre sont particulièrement réussis. Ainsi, Sztuczynska dans Astérix en Corse, met magnifiquement à profit les sonorités polonaises et propose une transcription de l’accent auvergnat presque plus arverne que l’original : « Wszpaniałe ! Jaka to piękna tkanina ! -To jedwab od nas. –Jedwab od wasz ! Ty to masz szczęście ! » (p.9), le second exemple d’adaptation réussie est celui concernant les animaux locaux : l’exclamation « -des sangliers domestiques ! –non, ce sont des cochons sauvages » devient chez Sztucynska: « -Dziki domowe ! –nie, to dzikie świnie » (p.25). Dans ces deux cas, le texte polonais, par sa note émotionnelle et ludique ne détonne en rien du texte français et en soutient l’humour. Le dernier exemple, nouveau par sa forme, est la traduction libre de l’antimétabole proverbiale citée plus haut: « la garde se rend mais ne meurt pas » (Astérix en Corse). Sztuczynska traduit cette antimétabole mot à mot mais, elle la fait suivre d’un renvoi en bas de page où nous lisons : «Wiele lat póżniej, pod Waterloo, Generał Napoleoński Cambronne przytoczy te słowa w odwrotnej kolejności: Gwardia umiera, lecz nie poddaje się. » (Asteriks na Korsyce p.43) « De nombreuses années plus tard, un Général des armées napoléoniennes, Cambronne, a cité ces mots dans un ordre différent »24. Subtilité de l’humour du texte adapté. Walkiewicz parle de « l’insondable potentiel de chaque langue » et n’est-ce pas justement cette possibilité de traduction libre ou d’adaptation qui va permettre de transmettre au mieux le message voulu par l’auteur de l’original? 5. Conclusion A travers cette étude, nous avons vu d’une part, combien le jeu de mots est une activité linguistique à la fois gratuite et ludique et, d’autre part, combien Goscinny utilise sans vergogne cette figure de style. Devrions-nous en conclure qu’un texte ludique, dont la trame est tissée de jeux d’esprit et de jeux de mots est intraduisible ? Certes non. Il nous faudra, bien sûr, accepter que le traducteur recrée l’original, qu’il produise un texte équivalent ou analogue plutôt qu’un texte trop fidèle qui donnerait 23 24 Notre traduction. Notre traduction. TRANSLATION AND MEANING, PART 10, 2013, 109 - 119 119 des résultats forcés. Il nous faudra accepter que le texte soit parfois adapté pour mieux servir le but fixé, ou encore que certains mots d’esprit disparaissent. Il nous faudra, enfin, accepter l’originalité ou l’inventivité du traducteur qui vise son propre destinataire comme l’auteur lui-même a visé le sien. Il semble que le courant contemporain qui donne davantage de liberté au traducteur littéraire libère le texte et, la vraie liberté ne conduit-elle pas à choisir le mieux? 120 TRANSLATION AND MEANING, PART 10, 2013, 121 - 128 121 “TRANSLATION LANGUAGE” IN POLISH VERSIONS OF THE GOSPEL OF MARK – A CASE STUDY Aleksander Gomola Jagiellonian University, Kraków, Poland Abstract: The Western world, despite declaring its post-Christian and secular identity, has not lost its interest in the Bible. More a cultural heritage rather than the Sacred Scripture, the Bible is today an object of study of scholars representing various approaches in modern humanities. Aichele et. al (1997) and Jobling et. al (2001) present a wide array of its possible readings, definitely not complete. In recent years we have also had quite a few interesting Bible translations, among them into English and German, that reflect new trends and ideas in biblical studies and break with traditional rendering of the original imposed on translators by most Christian churches (Today’s New International Version (TNIV), Berger and Nord, 1999; Bail and Crüsemann, 2006). Some of these translations were already analyzed becoming an object of heated debate (e.g. TNIV, see: Carson, 1998; Strauss, 1998). In this paper we wish to present much less known, yet very interesting translation of the Gospel of Mark into Polish by Tomasz Węcławski and to demonstrate what this Gospel may look like if translated without theological presuppositions that are almost always present in the mind of a translator, although they are not always an inherent part of the source text. 1. Biblical translation and the “translation language” Biblical translation has been regarded for centuries as a unique form of translation due to the status of the Bible seen by many as God’s word and its fundamental importance in the Western societies till the Enlightenment. Whenever problems of biblical translations are discussed, we tend to think of Nida and his distinction between two types of equivalence, possible to be seen in broader terms as foreignization vs. domestication dichotomy (Rohrbaugh, 2008: 11-25). Yet in this paper we would like to turn our attention not so much to the issue of the cultural distance between the ancient authors of the Bible and its modern readers, but to the more basic, and hence not always visible approach adopted in the biblical translation. While in translation of literary texts we expect a translator to provide us with a new reading of the source text and its new interpretation, in the biblical translation, on the contrary, we want a translator to reassure us in our beliefs and convictions. Novelty is not commended, tradition is recommended. This approach results in what Nord calls a “translation language”, accepted most often without criticism and regarded as being characteristic of a specific text type (Nord, 2005: 73). In the case of the Bible the “translation language” manifests itself, among other things, through a “biblical style”, the impact of which is hard to underestimate if one keeps in mind that vernacular versions of the Bible, such as KJV or Luther’s Bible, or Bibles in other national languages were the most widely read printed texts in the West for many generations. Yet even more important part of the “translation language” of the Bible is its theological importance since biblical terms, especially the ones we find in the New Testament are the basis of the complex philosophical, moral, and theological system of Christianity, construed, expanded, modified and refined for hundreds of years. Terms such as “baptism”, “grace”, “redemption” or “repentance” regarded as elements of religious language today, were taken from everyday, koine Greek spoken in the Eastern part of the Roman Empire and took on their new, religious meaning in the first decades of Christianity when oral preaching (kerygma) turned into a written gospel. In time this new meaning became the dominant one, while the previous one was pushed into the background. Although such metamorphosis is nothing new in a natural language, happening all the time with regard to many areas of lexis, it may sometimes lead to interesting effects in translation. Let us take into account the beginning of Yeats’ poem Lapis lazuli: 122 ALEKSANDER GOMOLA “I have heard that hysterical women say They are sick of the palette and fiddle-bow. Of poets that are always gay”. (Yeats, 2000: 2500). What Yeats had in mind in the third line of his poem was definitely not sexual orientation of poets but their supposedly naïve optimism and joyfulness in the face of military conflict looming over the world. One might argue however (especially if one is a follower of New Criticism and a proponent of intentional fallacy) that the new meaning of an adjective “gay” enriches the text and its connotations for a modern reader. Still, one might argue that the most probable solution adapted by a translator of Lapis lazuli into another language will be to retain first of all in his/her translation the previous, older meaning of “gay”, even at the price of losing an interesting connotation obvious for an English reader of the source text. After all, “poetry is what gets lost in translation”. Paradoxically, in biblical translation, the reverse situation is preferred as a translator is expected to transfer the new, theological meaning of lexical elements of the Bible into the target language ignoring the basic, non biblical and secular meaning of the text, obvious to those who know koine Greek and confirmed by other contemporary non-biblical texts (Crook, 2008: 28:35). Although translations of the Bible based on critical, philological reading of the source text date back to the Renaissance, it was not until the Enlightenment that the Bible started to be read as a secular, literary text, a historical record and a cultural heritage rather than God’s word (Sheehan 2005). In the 20th Pius XII in his encyclical Divine Afflante Spiritu acknowledged that the Bible is also a literary text and comprises various genres, however, for obvious reasons, this fact did not change the strategy of its translation in the Catholic Church, according to which a translator is supposed not to transfer the most basic, original meaning of some lexical elements of the source text but to see them through the prism of the “translation language”, i.e. with their theological accretions. An attempt to translate the biblical text unearthing the original, non-theological meaning of its lexis seems to be therefore not only a challenge to the existing translations but an interesting translational project as well. 2. The Gospel of Mark and its non-theological translations into Polish At first sight it seems to be unusual that a non-theological translation of one of the canonical Gospels was made into Polish as Poland is regarded (not so rightly) as a deeply religious country and Polish Catholicism is perceived (rightly) as conservative and resistant to any liberal tendencies in theology, including experimenting with biblical translation. Nevertheless, there have been two secular translations of some parts of the Bible into Polish. There was an earlier secular translation of the Gospels of Matthew and Mark by Witwicki, an outstanding Polish philosopher and psychologist who declared himself as a non-believer, (Witwicki, 1958) and a more recent one, by Węcławski, presented in this paper. The aim of both translators was to uncover the meaning of the source text hidden behind the theological “value added”. Interestingly, while Witwicki claimed that his translation was a translation of a non-believer, Węcławski published his translation as a Catholic priest (actually one of the most prominent Polish theologians) only to leave the Catholic Church in 2007 and to declare himself officially as a non-Christian, changing even his family name from “Węcławski” to “Polak”. The Gospel of Mark was neglected for many centuries, both by theologians and biblical scholars, and was even dubbed “Cinderella” among the Gospels (Bilezikian, 1979:11). It is the shortest of the four, it did not contain initially a section on the resurrection of Jesus and his death portrayed in it may be seen as a death of a pious man, maybe a prophet, who miscalculated his actions to be crushed by political intriguing of his fellow-countrymen supported by ruthlessness of the Romans ruling in a rebellious province and not as a death of someone who declared himself openly to be God. The quality of the source language is not very high, characterized by “sheer lack of literary sophistication” (France, 2002: 7). It is not possible to present here, even in a very brief form the probable origin of the text, its composition and historical background because of the sheer volume of scholarly TRANSLATION AND MEANING, PART 10, 2013, 121 - 128 123 literature on the subject. What is important in our analysis is the fact, that it is a text in which theology and beliefs of the primitive Church did not dominate completely its style and lexis and therefore it is worth seeing what a translator may excavate from it when he searches for the meaning in the source text and not in the tradition of the “translation language” accompanying it. Therefore we present below some excerpts of the Gospel of Mark in three translations: two traditional ones representing “translation language” as it is widely accepted today in the Polish Catholic Church and primarily “non-theological” translation by Węcławski. The first two translations are: Pismo Święte Starego i Nowego Testamentu. Biblia Tysiąclecia (1999) [the Millenium Bible – MB] – the most popular modern Polish translation officially approved of by the Catholic Church, regarded as rather literal rendering of the source text (Piela, 2003), although the literalness involves rendering of syntactic and idiomatic aspects of the source text and not literal i.e. non theologically oriented translation of its lexical elements; and Pismo Święte Starego i Nowego Testamentu. Najnowszy przekład z języków oryginalnych (2008) [the Paulist Bible – PB] – the most recent Polish Catholic translation in which the translators adopted principles of dynamic equivalence, including some postulates of the inclusive language. Węcławski’s translation (TW) was published in 2004 and was an object of a separate study (Szczepińska, 2005). Due to limited volume of the paper we will concentrate only on the most salient examples. Translation analysis 1: 1 Początek Ewangelii o Jezusie Chrystusie, Synu Bożym. [The beginning of the Gospel on Jesus Christ, Son of God.] MB Początek Ewangelii Jezusa Chrystusa, Syna Bożego. [The beginning of the Gospel of Jesus Christ Son of God.] PB Źródło dobrej nowiny Jezusa, Mesjasza, syna Boga. [The source of good news of Jesus, the Messiah, son of God.] TW. TW proposes his own, secular interpretation of Mark from the very first line. First of all, he decides to render the Greek term εὐαγγέλιον (euaggelion) not with the borrowed Polish form of the term “ewangelia” as almost all other Polish translations do but as “dobra nowina” (“good news”). In this way, he removes the most basic Christian connotation from the text and discloses its broader, secular meaning in the contemporary Greco-Roman world as “good news” of victory or the birth of the emperor (Kittel, 2003: 269). This solution is not uncommon in various contemporary translations. More problematic in his translation is the fact, that the original Greek Χριστός (Christos) underwent some kind of “back translation” and reads “mesjasz” (“messiah”). This is, as we know, a Jewish term, meaning “the anointed one” and translated very precisely into Greek as “Christos”. We should then expect in a Polish translation “pomazaniec” i.e. “the anointed one” rather than “mesjasz” (“messiah”). In other words, Węcławski replaces one foreignization with another, and does so against the spirit of the original. Another important detail in his translation is the separation of two terms relating to the founder of Christianity. A binominal Ἰησοῦς Χριστός, “Jesus Christ” (“Jezus Chrystus”) due to the lack of commas in the original and due to the relatively new (in cultural terms) Western tradition of identifying people by two names, is seen today by many as a forename and a surname of a Jewish prophet from Galilee and many translations of the Bible, both Polish and English, leave a reader unaware of this anachronism. In fact Χριστός is who Jesus was regarded as by his followers, and not his surname and to stress this, Węcławski places a comma between the two terms. The original allows for two interpretations of the term εὐαγγέλιον. It may be either “Christ’s good news” or “good news of/about Christ”. The latter seems to be more theologically grounded as Christ is presented as the contents of the good news, which implies seeing him as the Saviour or God. Węcławski chooses the former interpretation and to him Christ is the one who brings the good news not being necessarily the good news himself. Also PB adopts the same solution and resigns from pro- 124 ALEKSANDER GOMOLA moting this theological interpretation which differs it from MB whose translators present Christ as the contents of “good news”. Finally, a word of comment concerning the very first word of the source text. The Greek term ἀρχή (archē) is one of the key terms of Greek philosophy and means both a beginning and an origin/source. In Polish translation it is not possible to render both meanings simultaneously and Węcławski chooses the latter one that bears some philosophical connotations, unlike MB and PB, that choose the former one, thus convincing a reader that ἀρχή means here the beginning of a story and nothing else. 1: 14 (and in other passages) Królestwo Boże. [Kingdom of God.] MB Królestwo Boże. [Kingdom of God.] PB Królowanie Boga. [Reign of God.] TW. βασιλείας τοῦ θεοῦ (basileia toū theoū) – “God’s kingdom” or “kingdom of God” is one of the key terms of Jesus’ preaching and Christian theology with many fascinating cultural consequences going far beyond the domain of religion. The notion behind it was the basis of political theology by Augustine of Hippo in his Civitas Dei and we also find it at the heart of various political visions of remodeling the Western world in the last two thousand years, even at the price of bloodshed, both Christian and non-Christian ones. Today feminist theology rejects it as another form of patriarchalism bondage (king-dom). These and other interpretations of the biblical term, may be seen, paradoxically, as a misinterpretation resulting from not very precise translation of the source text. Biblical scholars argue that the term does not mean a political entity but “God’s kingly rule” (Kittel, Friedrich 2003: 100), i. e. a dynamic reality, not a static structure. TW is aware of that and he renders the Greek phrase as „królowanie Boga” („reign of God”), while MB and PB have got “królestwo Boże” (“kingdom of God”). 1: 8 Ja chrzciłem was wodą, On zaś chrzcić was będzie Duchem Świętym. [I baptized you with water, yet He will be baptizing you with the Holy Spirit.] MB Ja ochrzciłem was wodą, a On będzie was chrzcił Duchem Świętym. [I baptized you with water, yet He will be baptizing you with the Holy Spirit.] PB Ja was zanurzyłem w wodzie, a on was zanurzy w duchu świętym. [I immersed you in water but he will immerse you in holy spirit.] TW. Mark 1:8 presents John the Baptist’s opinion on Jesus. Although quite laconic, the statement is theologically very significant, and what is the most important, the theological meaning of the statement is its “value added”, not an inherent part of it. In other words, looking at its translation one can see whether a translator adopts “translation language” of the Bible or ignores it. Both MB and PB translate the passage “theologically”, i. e. they render the Greek term βαπτίσει (baptisei) as “baptized”. Yet the term βαπτίζω (baptizō) meant in koine Greek simply “to immerse”. There is a great difference between baptism as a religious concept with all its connotations and implications and a simple physical act of being immersed into water and even if the former superseded the latter in “translation language”, the latter is still there and whether it will be unearthed or remains hidden, depends solely on a translator. Węcławski decides to render more basic, non-theological denotation of this word and as it is not possible to retain its theological sense simultaneously, this changes dramatically the meaning of the passage. He is consistent in his approach since he does not employ capital letters in translation of the Greek pronoun αὐτός (autos) – “he” and ἅγιος πνεῦμα (hagios pneuma) – “holy spirit” either. Doing so, he is faithful to the Greek text in which we do not find capital letters. Both TRANSLATION AND MEANING, PART 10, 2013, 121 - 128 125 MB and PB follow here “translation language” principles and have capital letters in “On” (He”) and “Duchem Świętym” (“with Holy Spirit”). In Polish spelling system capitalization is used to distinguish common nouns from proper names or to stress a higher sociolinguistic status of a referent. Therefore both MB and PB, by using capital letters, impose a clear theological interpretation of the passage on a reader according to which Jesus is presented not as a mere mortal and holy spirit as a person. In other words, both MB and PB prepare, by means of their “translation language”, the ground for the Trinitarian dogma which would develop in the next centuries and present it as an inherent part of the biblical text, whereas TW avoids it. 1: 16-20 Idąc dalej, ujrzał Jakuba, syna Zebedeusza, i brata jego Jana, którzy też byli w łodzi i naprawiali sieci. Zaraz ich powołał, a oni zostawili ojca swego, Zebedeusza, razem z najemnikami w łodzi i poszli za Nim. [Going a little farther, he saw James son of Zebedee and his brother John in a boat, mending their nets. Without delay he called them [to a religious function - AG], and they left their father Zebedee in the boat with the hired men and followed him.] MB A gdy uszedł nieco dalej, zobaczył Jakuba, syna Zebedeusza, i jego brata Jana. Oni też byli w łodzi i naprawiali sieci. Zaraz ich powołał, a oni zostawili w łodzi swego ojca, Zebedeusza, razem z najemnikami i poszli za Nim. [ When he had gone a little farther, he saw James son of Zebedee and his brother John in a boat, mending their nets. Without delay he called them, [to a religious function - AG], and they left their father Zebedee in the boat with the hired men and followed him.] PB A uszedłszy jeszcze trochę zobaczył Jakuba, tego od Zebedeusza, i jego brata Jana, ich także w łodzi, jak sprawiali sieci. No i wezwał ich. I zostawili swojego ojca, Zebedeusza w łodzi z najemnikami odeszli za nim”. [ When he had gone a little farther, he saw James, that of Zebedee and his brother John in a boat, preparing their nets. And he summoned them. And they left their father Zebedee in the boat with the hired men and followed him.] TW The passage above presents calling the first disciples by Jesus, which is expressed by the Greek term καλέω (kaleō). Both English “to call” and the Greek καλέω are polysemic words and their meanings include a physical act of calling (shouting) and summoning somebody. Thus in English we can say of somebody as being called by God and about his/her vocation (Latin vocare – to call) as the result of this experience. In Polish these two meanings are distinguished and denoted by two different verbs: “wołać” (to call/shout), “powołać” (to summon, esp. to a religious function; Polish “powołanie” (“vocation”) is a noun derived from “powołać”). Both MB and PB use “powołać” in their translation, creating a very strong presupposition that the first disciples were like Catholic priests. TW resigns from this “value added” and decides to render the Greek term non-theologically telling a reader that Jesus just “summoned” (“wezwał” ) John and his brother. 1: 25 Lecz Jezus rozkazał mu surowo: “Milcz i wyjdź z niego!” [But Jesus told him sternly: “Be quiet and come out of him!”] MB A Jezus stanowczo mu nakazał: “Milcz i wyjdź z niego!” [And Jesus told him sternly: “Be quiet and come out of him!”] PB. A Jezus go skarcił mówiąc: “Zamknij się i wyjdź z niego!” [And Jesus rebuked him saying: “Shut up and come out of him!”] TW. What was Jesus’ idiolect? According to many scholars (e.g. Vermes, 1981) Jesus was a Galilean Jew of rather low origin. He was not using then a very refined language characteristic of prominent Christians of the later centuries like Origen or Augustine brought up in the rhetorical tradition of antiquity. However the “translation language” of the Bible very often makes him speak a very dignified language, even if the original looks different. Of course, every word in the Gospels uttered by Jesus including even famous ipsissima verba Iesu may be hardly authentic, since Gospels are not 126 ALEKSANDER GOMOLA historical records but theological texts. Nevertheless, one must agree that whenever Jesus speaks dignified language this makes him sound more like a divine figure which he, according to the Christian creed is. While MB and PB render the Greek φιμόω (phimoō) with a formal Polish imperative “milcz” (“be silent”), TW decides to use a straightforward and not very elegant phrase “zamknij się” (“shut up”). Therefore Jesus in TW’s translation seems to be one of many quick-tempered Jewish prophets famous for speaking always their minds and not a divine figure full of majesty and dignity. 1:40 Wtedy przyszedł do Niego trędowaty i upadając na kolana, prosił Go: “Jeśli chcesz, możesz mnie oczyścić”. Zdjęty litością, wyciągnął rękę, dotknął go i rzekł do niego: “Chcę, bądź oczyszczony!”. [A man with leprosy came to him and falling on his knees begged him: “If you are willing, you can make me clean.” Filled with compassion, Jesus reached out his hand, touched the man and said to him: “I am willing. Be clean!”] MB. Wtedy podszedł do Niego trędowaty, upadł na kolana i prosił: “Jeśli chcesz, możesz mnie oczyścić”. Ulitował się, wyciągnął rękę i dotknął go i powiedział: “Chcę, bądź oczyszczony” [A man with leprosy approached him and begged him on his knees, “If you are willing, you can make me clean.” Jesus felt pity for him, reached out his hand and touched the man and said: “I am willing. Be clean!”] PB I przychodzi do niego trędowaty i prosi go na kolanach tak: “Jeśli chcesz, możesz mnie oczyścić”. I uniósł się, wyciągnął swoją rękę, dotknął i mówi mu: “Chcę, bądź oczyszczony”. [A man with leprosy comes to him and begs him on his knees, “If you are willing, you can make me clean.” And he got angry, reached out his hand, touched the man and says to him: “I am willing. Be clean!” ] TW The impression one gets from 1: 25 in TW is augmented in 1: 40. Actually 1: 40 is one of the most interesting passages not only in Mark but in all the New Testament due to the intriguing discrepancy between the manuscripts. Presenting Jesus’ reaction to a leper’s plea some of them read that Jesus was filled with compassion, the others that he got angry (Ehrman, 2007: 133-135). Most modern popular translations (but not all of them, e.g. Today’s New International Version) opt for the first variant, because this reading corresponds with Jesus as the manifestation of loving God. TW follows here the basic principle of textual criticism that says that out of two versions of the text the more difficult one is proper and therefore decides to present Jesus as indignant and not feeling pity for a leper. In this way once again the image of Jesus as a mere mortal is highlighted and the “translation language” is abandoned. 8: 36 Cóż bowiem za korzyść stanowi dla człowieka zyskać świat cały, a swoją duszę utracić? [What advantage is it for a man to gain the whole world, yet forfeit his soul? ] MB Jaką korzyść ma człowiek, jeśli zdobędzie cały świat, ale straci swoje życie? [What advantage is it for a man to gain the whole world, yet forfeit his life?] PB Bo co to pomoże człowiekowi, jeśli cały świat zyska, a swoje życie straci? [What will it help for a man to gain the whole world, yet forfeit his life?] TW. This is one of the most popular biblical passages, so widely known that probably not everybody knows that it comes from the Bible and represents a part of Jesus’ teaching on the eternal life of one’s soul as the spiritual element of human nature, unique to every individual. But does Jesus really mean here what for example the modern Catholic Church means teaching on the immortality of soul and using this passage to support its doctrine? The Greek term ψυχή (psychē) may be rendered as „soul” or simply “life”. Translating it as “duszę (“soul”) like MB does, is yet another example of the “translation language” according to which the biblical text is seen not in his own right but as a resource of scriptural evidence for some elements of the Christian doctrine. This time not only TW but also PB resigns from a theological “value added” and chooses “życie” (“life”) instead of an at- TRANSLATION AND MEANING, PART 10, 2013, 121 - 128 127 tractive, yet doubtful, because of its later philosophical and doctrinal accretions, term “dusza” (“soul”). 15: 9 Jezus odrzucony przez swój naród [Jesus is rejected by his nation] Piłat im odpowiedział: “Jeśli chcecie, uwolnię wam Króla Żydowskiego?” [Pilate answered them: “If you wish I will release you the Jewish King”.] MB Skazanie Jezusa na śmierć [Jesus is sentenced to death) Piłat im odpowiedział: “Jeśli chcecie, uwolnię wam króla Żydów”. [Pilate answered them: “If you wish I will release you the King of Jews”.] PB A Piłat im odpowiedział tak: “Chcecie, to wam uwolnię króla Judejczyków?” [And Pilate answered them in this way: “If you wish, I will release you the king of Judeans?”] TW As we know, the text of the Bible preserved in the oldest manuscripts and codices is not divided into verses or chapters. It does not contain titles of sections either. However, such titles are commonly found in various translations of the Bible and as such they must be seen as a part of a translation process and should not be ignored. How dangerous they may be, forcing a reader into a specific interpretation, one can see in MB, where the trial of Jesus before Pilate and Jews’ role in it is preceded by the title “Jesus is rejected by his nation”. It is hard to escape the impression that it is an anti-Semitic title, suggesting that Jews as nation are guilty of deicide. This idea underlay the anti-Semitism among Christians for centuries. The Second Vatican Council rejected it, yet, as we can see, some traces of it remain even in the translations of the Bible approved of by the Church authorities. PB’s title is neutral and descripitive, while TW resigns from any title which seems to be the best solution. Another interesting detail concerning this passage is how translators render the Greek adjective Ἰουδαῖος (Ioudaios). MB and PB, like most traditional translations, refer it to Jews as a nation and translate “Żydowski”/”Żydów” (“Jewish”/”Jews”). TW, following some modern translations (e. g. TNIV), decides not to render it literally and to narrow its meaning. Thus in PB and MB Jesus is “the King of the Jews” while in TW he is “the King of Judeans” only (i. e. inhabitants of Judea). This decision on the part of a translator might be questioned, given the historical and political context of gospel narrative and other occurrences of this term in the Synoptic Gospels. (It is of course a good solution with regard to the translation of the Gospel of John representing the attitude of so called “Johannine community”; Brown 1979: 40–45). Yet it should be seen first of all as an attempt to avoid modern connotations of “Jews” and “Jewish” (anti-Semitism, holocaust, etc.). In other words, TW wants to present Jesus as it might have been seen by somebody living in the Roman province of Syria in the 1st century A.D. In this way he also breaks with a “translation language” characteristic for this passage for centuries. 15: 37 Lecz Jezus zawołał donośnym głosem i oddał ducha. [But Jesus cried with a loud voice and gave up his spirit.] Lecz Jezus zawołał donośnie i oddał ducha [And Jesus shouted loudly and gave up his spirit.] A Jezus wydał wielki krzyk i wyzionął ducha. [With a loud cry Jesus breathed his last.] Mk 15: 37 depicts Jesus’ death. Each Evangelist presents it in a different way making it an important part of their theological visions. In Mark the death of Jesus is a “mind-stretching antinomy” (France, 2002: 650), tragic and incomprehensible to Jesus, who in 15: 34 cries “My God, my God, why hast thou forsaken me?” Once again the human-divine ambiguity comes to the fore in Mark. He describes the moment of the death using the verb ἐκπνέω (ekpneō) – “to breathe out”, “to breathe one's last”, “to expire” which TW translates as “wyzionął ducha” (“expired”). This is one again very faithful, non-theological rendering of the original, presenting Jesus as a mere mortal. MB and PB present the 128 ALEKSANDER GOMOLA death of Jesus in a different way as their translations read “oddał ducha” (“gave up his spirit”). This is a transfer of Johannine theology onto Markan text since in the Gospel of John we read that Jesus “παρέδωκεν τὸ πνεῦμα (paredoken to pneuma), “gave up spirit” (“oddał ducha”). This is yet another example of the “translation language” helping to preserve the image of Jesus as God Incarnate, whose death is understood as his deliberate act of sacrificing his life for humankind. 3. Conclusions In this paper we have presented just a few excerpts from three contemporary translations of the Gospel of Mark into Polish. Even this short analysis proves that adopting or rejecting the “translation language” of the Bible with regard to its lexis and translating biblical terms theologically or nontheologically results in significantly different versions of the target text. The more detailed and exhaustive analysis would probably provide further evidence that the problem of the “translation lan guage” very often overlooked, seems to be of great importance in translation practice especially in the case of texts of such cultural and social meaning as the Bible. References Aichele, G., F. W. Burnett (eds.) (1995), The Postmodern Bible. New Haven: Yale University Press. Bail, U., F. Crüsemann (2006), Bibel in gerechter Sprache. Gütersloh: Gütersloher Verlagshaus. Berger, K, C Nord (1999), Das Neue Testament und frühchristliche Schriften. Leipzig: Insel Verlag. Bilezikian, G. (1979), The Liberated Gospel: A Comparison of the Gospel of Mark and Greek Tragedy, Grand Rapids: Baker Books. Brown, R. E. (1979), The Community of the Beloved Disciple. New York – Mahwah: Paulist Press. Carson, D. A. (1998), The Inclusive Language Debate. A Plea for Realism. Michigan: Baker Books. Crook, Z.A. (2008), Grace as Benefaction in Galatians 2:9, 1 Corinthians 3: 10, and Romans 12:3, 15, 15. In: Neufeld, Dietmar (ed.), The Social Sciences and Biblical Translation, 25–38. Atlanta: Society of Biblical Literature. Ehrman, B. (2003), The Lost Christianities: The Battles for Scripture and the Faiths We Never Knew. Oxford: Oxford University Press. Ehrman, B. (2007), Misquoting Jesus. The Story Behind who Changed the Bible and Why. San Francisco: HarperOne. France, R.T. (2002), The Gospel of Mark: a commentary on the Greek text. Grand Rapids, Michigan: Wm. B. Eerdmans Publishing Co. Jobling, D., T. Pippin and R. Schliefer (eds.) (2001), The Postmodern Bible Reader. Oxford: Blackwell. Kittel, F. (2003), Theological dictionary of the New Testament. Grand Rapids, Michigan: Wm. B. Eerdmans Publishing Co. Piela, M. (2003), Grzech dosłowności we współczesnych polskich przekładach Starego Testamentu. Kraków: Wydawnictwo Uniwersytetu Jagiellońskiego. Pismo Święte Starego i Nowego Testamentu. Biblia Tysiąclecia (1999). Poznań: Pallotinum. Pismo Święte Starego i Nowego Testamentu. Najnowszy przekład z języków oryginalnych (2008). Częstochowa: Edycja św. Pawła. Rohrbaugh, R. (2008), Foreignizing Translation In: Neufeld, Dietmar (ed.), The Social Sciences and Biblical Translation, 11–25. Atlanta: Society of Biblical Literature. Sheehan, J. (2005), The Enlightenment Bible: Translation, Scholarship, Culture. Princeton: Princeton University Press. Strauss, M.L. (1998), Distorting Scripture. The Challenge of Bible Translation and Gender Accuracy. Illinois: InterVarsity Press. Szczepińska, B. (2005), “Ewangelia zadziwienia. Tłumaczenie Ewangelii św. Marka ks. Tomasza Węcławskiego”. In: Szczepińska Ewangelie tylekroć tłumaczone... : studia o przekładach i przekładaniu. 276– 290 Gdańsk: Wydawnictwo Uniwersytetu Gdańskiego. Today’s New International Version, (2005) Grand Rapids, Michigan: Zondervan Publishers. Vermes, G. (1981), Jesus the Jew. Augsburg: Fortress Publishers. Węcławski, T. (2004), Dobra nowina według św. Marka. Poznań: W Drodze. Witwicki Władysław 1958 Dobra Nowina według Mateusza i Marka. Warszawa: Państwowe Wydawnictwo Naukowe. TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 129 A CORPUS-DRIVEN STUDY OF TRANSLATION UNIVERSALS: THE CASE OF ENGLISH AND RUSSIAN “LOLITA” AND REFERENCE CORPORA OF LITERARY TEXTS Łukasz Grabowski Opole University, Opole, Poland Abstract: This article falls within the scope of descriptive translation studies (DTS) and corpus linguistics, and its aim is twofold. Firstly, it attempts to compare the English (1955) and Russian (1967) version of the novel "Lolita" by V. Nabokov with corresponding English and Russian reference corpora, which include samples of typical literary language, in order to compare core patterns of lexical use (Laviosa 1998) and find traces, if any, of translation universals. Secondly, it aims to compare English "Lolita" with Russian "Lolita" in order to identify major lexical differences between the two texts. Thus, the corpus-driven methodology used in this article is text-centred, as it aims to compare the texts with each other and with reference corpora in order to describe what happened in the translations, and to determine major lexical differences between translational and non-translational language. Apart from identifying a number of stylistic differences between the two texts, the analyses revealed a number of T- and S-universals applicable to English-to-Russian literary translation. 1. Introduction Lolita (1955) is one of the best known novels by Vladimir Nabokov, which firmly established him as an outstanding American novelist. If Lolita was ever going to be translated into Russian, Nabokov wanted to ensure maintaining control over the outcome. He commenced translation in winter 1963 and in March 1965 the Russianized novel was revised and completed (Boyd, 1995: xlvii; Trzeciak, 2005: 11). Nevertheless, Nabokov's postscript to the Russian edition of Lolita contains passages which illustrate his skepticism and dissatisfaction with the self-translation, which he calls “the story of disappointment”, and claims that many fragments of his translation are “clumsy, loquacious and frequently disgusting in terms of style and rhythm” (Nabokov, 1989: 358). Therefore, in this article, the English and Russian Lolita (henceforth the ‘parallel texts’) will be compared with each other with emphasis on core patterns of lexical use as attested in the two texts. However, according to Baker (1995: 233), descriptive translation studies shall not be limited to the comparison of source texts and their translations, but they shall also be extended to the comparison of non-translated texts with translated texts, which are produced under different social, cultural and sometimes even political circumstances. Further, Baker (1995: 243) also posits a number of hypotheses on the differences between translational and non-translational language (i.e. translation universals), e.g. that translations tend to be, among others, more explicit as regards lexis and syntax than non-translated texts, their content and form is simplified if compared with non-translated texts, and that language used in translation is more conventional and less creative than the one used in nontranslated texts. According to Olohan (2004: 92), these patterns are specific to translation and are seen to be more typical of translational language than of non-translational language. In addition, these patterns are a product of constraints inherent in the translation process and do not vary across cultures. Thus, it is essential to study linguistic patterns which are specific to translated texts, irrespective of source and target languages involved (Laviosa-Braithwaite, 1995: 153). Further, Laviosa (1998: 557-570), studied distinctive features of translational English as compared with native English (represented by samples ellicited from the BNC) and found that translational English has four core patterns of lexical use: a relatively lower proportion of lexical words over function words, a relatively higher proportion of high-frequency words over low-frequency words, a relatively greater ŁUKASZ GRABOWSKI 130 repetition of the most frequent words, and a smaller vocabulary frequently used. The latter three characteristics will be also addresses in this study. Therefore, in this article, the English and Russian Lolita (henceforth ENL and RUL, respectively) will be also compared with corresponding reference corpora, which include samples of typical literary language, to compare core patterns of lexical use, find lexical differences between them as well as detect traces, if any, of translation universals. For the purposes of this study, the typology of translation universals [TUs] proposed by Chesterman (2004) was applied, who distinguished between two types of TUs: the S-universals, which are related to translation from the source to the target language, and the T-universals, which are related to comparisons of translational and non-translational language (i.e. target-language texts, which are not translations). 2. Methodology, research material, tools and stages of the analysis In this study the bottom-up corpus-driven methodology was applied. In contrast to the corpus-based approach, which always works within commonly accepted frameworks of theories of language (which implies prior classification of linguistic data), the parallel text and reference corpora were not adjusted to fit any predefined categories or theoretical schemata, and the study questions were addressed through empirical analysis of frequency distributions of words and recurrent patterns of language use as found in English and Russian Lolita and corresponding reference corpora. The parallel texts were manually scanned and OCRed. Then the scanned texts were repeatedly proofread in order to ensure spelling accuracy, and they were further verified against the paper format versions. At that stage, any cases of misrecognition of characters were edited and corrected using a spellchecker, or a search-and-replace facility of a word processor. The texts included in the English reference corpus (henceforth ENRC) were downloaded from the Brown Corpus Manual website (http://khnt.aksis.uib.no/icame/manuals/brown/). The texts included in the Russian reference corpus (henceforth RURC) were downloaded from the Maxim Moshkov's Library website (www.lib.ru). Finally, the parallel texts and reference corpora were saved in two files in a plain text format (encoded in Unicode). The structure of both reference corpora is presented in Table (1) and (2) below. 1 2 3 4 5 6 7 8 Text Category F (Popular lore) G (Belles Lettres) K (Fiction – General) L (Fiction – Mystery) M (Fiction - Science) N (Fiction – Adventure) P (Fiction – Romance) R (Fiction – Humour) TOTAL No of texts (2000 tokens each) 48 75 29 24 6 29 29 9 249 (507,408 tokens) Table (1). Structure of English reference corpus (ENRC) No 1 2 3 4 5 Author B. Balter V. Kaverin I. Kazakov K. Paustovskiy V. Tendryakov Title Do svidaniia malchiki! (1963) Sem Par Necistych (1961) Izbrannye rasskazy (1965) Zolotaya Roza (1955) Nakhodka (1965) Size (tokens) 64,100 17,301 29,381 55,596 17,271 TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 No 6 7 8 9 10 11 12 Author V. Tendryakov V. Tendryakov I. Trifonov I. Trifonov I. Nagibin V. Kataev B. Mozhaev TOTAL Title Sud (1960) Ukhaby (1960) Dom na naberezhnoy (1970-1976) Predvaritelnye itogi (1970) Bogoyar (1944-1994) Almazny moi venetz (1978) Zhivoy (1966), Naled’ (1959) 131 Size (tokens) 20,848 12,938 44,218 20,418 55,570 59,448 35,323 + 30,055 504,960 Table (2). Structure of Russian reference corpus (RURC) When compiling the above corpora, the most important criteria were representativeness and size. As regards representativeness, it was aimed that the reference corpora represent typical literary English and Russian contained in the novels published around the time when both language versions of the novel were launched onto the book market. As for the size, the reference corpora are approximately five times larger than study texts. According to Berber-Sardinha (2000: 7-13), who conducted the study aimed to estimate the ideal size of a reference corpus to be utilized in WordSmith Tools, a reference corpus which is five time larger than the study corpus generates a similar number of keywords as reference corpora which are 100 or 200 times larger than the study corpus. Subsequently, the corpus-driven analysis was completed with the use of WordSmith Tools 4.0 developed by Scott (2004), which is a suite of programs custom-designed for text analysis. The study was broken down into two successive stages. First, the ENL was compared with the ENRC, and the RUL with RURC (the latter study being comparison of translational and non-translational language). In particular, I looked at descriptive statistics, frequency profiles, frequency spectra and keywords generated for English and Russian version of the novel, respectively. Secondly, the ENL was compared with the RUL, and there I focused on descriptive statistics, frequency profiles, frequency spectra and top-frequency lexical wordlists. 3. Comparison of parallel texts with reference corpora 3.1 Descriptive statistics Descriptive statistics describe linguistic data in quantitative terms, which is commonly accepted in corpus linguistics as basic indicators of style and lexical richness (Olohan, 2004: 78-81). Hence, it provides a holistic view of both the parallel texts and reference corpora, which is presented in Table (3). Firstly, ENL will be compared with ENRC, and then RUL will be compared with RURC. Statistics Number of tokens ENL 112,230 ENRC 507,408 RUL 102,548 RURC 504,672 Number of types 13,991 30,329 29,651 71,480 Type/token ratio (TTR) (%/or x per 100 tokens) Standardized TTR (STTR) (in %/or x types per 100 tokens) Mean word length (in characters) Number of sentences 12.46 5.97 28.91 14.16 51.67 45.58 67.46 61.64 4.40 4.50 5.44 5.19 5,549 29,459 5,531 48,792 ŁUKASZ GRABOWSKI 132 Statistics Mean sentence length (in tokens) Sentence length standard deviation (in tokens) ENL 20.22 ENRC 17.22 RUL 18.54 RURC 10.34 20.63 12.57 18.30 10.02 Table (3). Descriptive statistics for parallel text and reference corpora As the reference corpora are approximately 5 times larger in terms of the number of tokens than the novels, it is necessary to filter out those quotients which are not sensitive to differences in size (i.e. STTR, mean word length, mean sentence length, sentence length standard deviation). These are printed in bold in Table (3). In general, STTR provides brief information as to the complexity/simplicity or specificity/generality of a particular text or corpus. According to Baker (2006: 52), low STTR indicates narrow range of vocabulary, or narrow scope of subjects discussed in a text or corpus, and high STTR indicates the opposite. The STTR value shows that on average in ENL there are almost 52 types per 100 tokens, whereas in ENRC there are only 46 types per 100 tokens. This perfunctory measure of lexical richness shows that Lolita is more complex and specific lexically in comparison with ENRC, and has slightly fewer repetitions. The same is the case with RUL and RULC - the former has almost 68 types per every 100 tokens whereas the latter has only 62 types. As far as mean word length is concerned, the data generated for ENL and ENRC (4.40 and 4.50 characters) as well as for RUL and RURC (5.44 and 5.19), respectively, provides no conclusive evidence of conspicuous differences. As regards the mean sentence length, for ENL it accounts for 20.22 tokens per sentence, which is 3 tokens more than in ENRC. Thus, we can conclude that longer sentences in ENL indicate higher lexical variety than ENRC (which is also confirmed by the STTR) as well as higher information load in the sentences. Moreover, the mean sentence length standard deviation (20.63) shows even higher aberration from the typical mean sentence length attested in ENRC (only 12.57), which means that generally sentences in the English-original Lolita are more challenging linguistically, and at the same time, they are more explicit and precise than sentences in ENRC. This characteristic is even more pronounced in the case of RUL and RURC, where on average sentences in Russian Lolita are 8 tokens longer than in the reference corpus. This considerable difference shows that sentences in RUL are more varied lexically, carry higher information load, are more explicit and precise than short and terse sentences in RURC. It is further confirmed by the mean sentence length standard deviation (18.30 in RUL vs. 10.02 in RURC), which indicates that Nabokov consistently used longer sentences. This difference may be also due to the fact that sentences attested in ENL have the mean length of 20.22 tokens per sentence. It may be the case that Nabokov restrained from the use of sophisticated syntactic translation strategies and did not interfere in the length of sentences translated into Russian. Thus, long sentences attested in Russianized Lolita do not conform with the typical literary Russian used in the 1950s, 1960s and 1970s as attested in RURC. To sum up, one may hypothesize that both ENL and RUL (translated text) are more complex lexically and have more explicit and precise sentences than reference corpora (non-translated texts), which in the case of Russian language material may be treated as a T-universal. 3.2 Frequency profiles In order to determine whether it is ENL or ENRC, or RUL or RURC that has more repetitions and lower lexical variety in terms of top-frequency words, a frequency profile proposed by Baroni (2006: TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 133 3-4) was used. As a rule, the frequency profile is obtained by a replacement of words in a frequency list with their frequency-based ranks, by assigning rank 1 to the most frequent word, rank 2 to the second frequent word, rank 3 to the third frequent word etc. It enables one to answer the question which frequency-based ranks (r) of words (tokens) have a particular frequency (f). However, a typical frequency profile was modified in that frequency information was substituted with information on cumulative percentage of the total word count (%cW) corresponding to frequency-based ranks. The results are presented in Tables (4) and (5). Rank 1-100 ENL %cW 49.54 Rank 1-100 ENRC %cW 49.48 Table (4). Frequency profiles for ENL and ENRC Rank 1-100 RUL %cW 35.04 Rank 1-100 RURC %cW 36.05 Table (5). Frequency profiles for RUL and RURC The data show that ENL and ENRC have similar distribution of top-frequency words, and it is the same in the case of RUL and RURC (with the latter being slightly more repetitive). Nevertheless, there is no evidence to confirm that translational language (RUL) has relatively greater repetition of the most frequent words or that a smaller vocabulary is frequently used in it than in typical Russian literary language as attested in RURC. Therefore, these core patters of lexical use, or T-universals, were dismissed. 3.3 Frequency spectra According to Baroni (2006: 3-4), frequency spectra enable one to determine how many word types in a frequency list have a particular frequency. As creative or author-specific vocabulary usually occurs in a text with low frequencies, frequency spectra can be used to study lexical variety and degree of repetitions among bottom-frequency words (i.e. a text is more varied lexically if proportion of bottom-frequency words in the total word count is higher). Therefore, the frequency information in the frequency spectra was substituted with information on the cumulative percentage of the total word count (%cW) corresponding to words types with frequencies 1-25. The results are presented in Tables (6) and (7). w (f) 1-25 ENL %cW 34.85 % w (f) 1-25 ENRC %cW 20.63 % Table (6). Frequency spectra for ENL and ENRC w (f) 1-25 RUL %cW 54.46 % w (f) 1-25 RURC %cW 37.90 % Table (7). Frequency spectra for RUL and RURC The data show that ENL has higher lexical variety among bottom-frequency words (i.e. words with frequencies 1-25) than ENRC, and the same is true in the case of RUL and RURC whereby trans- ŁUKASZ GRABOWSKI 134 lational Russian represented by RUL is more varied lexically than RURC. It follows that a Tuniversal whereby translational Russian (RUL) is more varied lexically in terms of bottom-frequency words than typical literary Russian (RURC) was revealed. 3.4 Comparison of keywords According to Scott (1996-2004), keywords may be classified into three groups, each of them being an indicator of a different aspect of a text or a corpus under investigation. The first group includes proper nouns, which are incidental or specific to the text (e.g. names of protagonists, or places where the plot unfolds). The second group are lexical items, which would be also recognized by a human being with a naked eye, and which provide information on aboutness of a text. The third group accounts for a high frequency words, which are common grammatical (function) words, which provide information on style of the author of a text rather than the text’s aboutness. Baker (2000: 245, quoted in Olohan 2004: 147) understands style as author’s or translator’s presence in a text with emphasis on characteristic use of language, individual profile of linguistic habits as compared to other writers or translators. To sum up, according to Baker (2004: 239), close observation and examination of keywords should reveal the most significant lexical differences between study corpus and reference corpus in terms of aboutness and style. As this analysis focuses on style, proper names identified as keywords were disregarded. Therefore, it is essential that keywords are not important words in a sense that they are culture-specific or symbolic1, but they are identified through ‘keyness’ (aka 'keyness score' or ‘K-score’), which is a statistical method of extraction of lexical items with high relevance to the text topic (Waliński, 2004: 51). It is calculated by comparing frequency of a word in a study corpus with the frequency of the same word in a reference corpus, taking into consideration the size of both corpora (Scott,19962004 & 1997a). As a result, keywords are not the most frequent words, but words which are unusually frequent in a study corpus in comparison with a reference corpus. As selectivity was the priority for generation of keywords, I used Dunning’s (1993) log likelihood test (LL) with a p value set at 0.000001 and the minimum frequency set at 32 in order to obtain fewer but more representative keywords. The KeyWords facility of WordSmith Tools generated 162 and 192 positive keywords for ENL and RUL, respectively. In order to isolate statistically significant number of keywords, the Pearson’s chisquare test was applied, following the method proposed by Berber-Sardinha (1999: 6), which yielded 98 and 90 statistically significant keywords in ENL and RUL, respectively. These will be presented in Tables (8) and (9), divided into lexical and function words. Nouns 1 ENL NYMPHET, CHILD, READER, CAR, TENNIS, NYMPHETS, GIRL, DAUGHTER, DARLING, LAKE, LAWN, HEART, TENDERNESS, FROCK, RACKET, DOG, WRIST, BED, CHAIR, HIGHWAY, BARE, LIPS, LIMBS RUL МИСС, МИССИС, МИСТЕР, ЧИТАТЕЛЬ, АВТОМОБИЛЬ, ШОССЕ, ГОСТИНОЙ, НИМФЕТКА, ПОСТЕЛИ, СМЫСЛЕ, ДЕВОЧКА, ЛАГЕРЬ, ДОЛЛАРОВ, ВАННОЙ, МЕРЕ, ПРИСЯЖНЫЕ, ТЕЧЕНИЕ, ЦЕЛЬЮ, ПИСТОЛЕТ, КАФЭ, Nevertheless, culture-specific or symbolic words in a text may also account for keywords, provided that they are unusually frequent in a text as compared with a reference corpus. 2 Minimum frequency value set at 3 excluded the use of chi-square test, which requires frequencies to be equal or higher than 5. TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 Others French borrowings / loan words (incl. grammatical) OH, LITTLE, AM, VERY, BROWN, MAGIC, LOVELY, QUITE, BLUE, PALE, PRESENTLY, GRAY, OLD, BLACK, GLOSSY, PUBESCENT, POOR, ROSY, AFAR JE, QUE, MON 135 МИЛЯХ, РАНЧО, ГОСПОЖИ ДОВОЛЬНО, СОВЕРШЕННО, ЗАТЕМ, СТОЛЬ, ПРОГОВОРИЛА, ПРОДОЛЖАЛА, АХ DE, TEM, EST, JE, MON, VOUS, QUE, BIEN, LA Table (8). Keywords – lexical (content words) in ENL and RUL As lexical keywords provide information on aboutness of a text, Table (8) shows that the themes discussed in ENL revolve around erotic fascination of Lolita on the part of Humbert Humbert (nymphet/ nymphets3, darling, heart, tenderness, magic, bare, rosy, lips, limbs, wrist), emphasis on Lolita’s being an innocent child (child, daughter, girl, poor), the items used by both Lolita and Humbert Humbert as the plot unfolded (tennis racket, lawn (which shows that Humbert Humbert relished in observing Lolita playing lawn tennis), dog, frock, bed, chair), modes of transportation used while roaming about the US and related vocabulary (car, highway). It is exactly the same in the case of RUL – eroticism (нимфетка/и/у/ой, постели), emphasis on Lolita being innocent child (девочка/ек), the places where the plot unfolded (гостиной, лагерь, ванной, шоссе, холл), modes of transportation used while exploring the US (автомобиль/я/е, миль, шоссе) come to the fore as major themes. However, the noun reader (used in ENL 51 times) does not refer to any particular theme, but it constitutes a linguistic habit and thus a stylistic choice of the author. It shows that the target reader of a text is addressed in a straightforward manner with the use of 2nd person singular. Moreover, the noun reader is used as a proper name, because it is never preceded with any article. As a result, the narrator (Humbert Humbert) is not distancing himself from a reader. He intends to be close, visible, as well as to absorb and involve a reader into the story, as well as provide some explanations of his controversial intentions and actions directed towards titular juvenile heroine. The same effect is made in RUL with the noun читатель (used 43 times in RUL). However, such a stylistic choice was not found in RURC. As regards linguistic habits visible in RUL, these include, among others, forms of address (мисс, мистер, миссис), which were not found in RURC4. Moreover, they serve as further evidence that Nabokov favoured literal translation and preferred translation strategy of foreignization where English forms of address are preserved to emphasize strangeness of a source text. The same is the case with measurements (миль) and names of buildings or their parts (холл). The presence of the noun автомобиль (together with all its inflectional forms, it occurred in RUL 106 times) is also indicative of the author-specific use of Russian because it is rarely used in the reference corpus RURC (all its 3 The very word nymphet was a Nabokovian coinage and due to the popularity of the novel in the world over, it can be found in most dictionaries published today. 4 However, the forms of address, such as мисс (Miss), миссис (Mrs) and мистер (Mister) are provided in contemporary Russian dictionaries, e.g. Ozhegov - Shvedova (2000) and they are defined as ‘polite forms of address in English-speaking countries’ (‘вежливое обращение в англоговорящх странах’). They were also included in older dictionaries of Russian dating back to 1950s (e.g. Blok et al 1958: 379-380), where мисс, миссис and мистер were defined as ‘polite forms of address in English-speaking countries’. ŁUKASZ GRABOWSKI 136 inflectional forms occur there only 36 times). Instead RURC shows high occurrence of the synonymous noun машина, as all its inflectional forms occur there 166 times (its frequency in RUL is only 15). The favoured adjectives and modifying adverbs in ENL are employed to enhance Lolita’s qualities (little, lovely, glossy, pubescent, quite, very, rosy); the adverb of time presently (occurred 28 times) is consistently used throughout the novel and constitutes the author-specific way of referring to time at the moment of speaking by Humbert Humbert. The last group of lexical keywords are borrowings from French (je, que, mon), which are stylistic devices used in code-switching on the part of Humbert Humbert in order to enhance his status of literary scholar born in Paris and a European sophisticate. As regards RUL, apart from the favoured modifying adverbs довольно (quite) and cовершенно (absolutely/totally), the presence of the reporting verb проговорил/а among keywords explains under-representation of a stock reporting verb cказал/а in RUL as compared with RURC. Pronouns ENL I, MY, HER, ME, SHE, WE, MYSELF, MINE, HERS, THOSE Conjunctions Prepositions Others AND WITH, UNDER A, SOME RUL Я, ОНА, МОЯ, КОТОРЫЙ, ТОГО, ИЛИ, ИБО, ЛИБО, ЧТОБЫ, ДАБЫ, НЕВЗИРАЯ МЕЖДУ, МЕЖ, СКВОЗЬ ГО, ЛИ, БЫ, Table (9). Keywords - grammatical (function) words Table (9) shows that most key pronouns are either first person singular or plural pronouns (I, my, me, myself, mine, we), which underscores the 1st person narration, or 3rd person singular feminine pronouns (she, her, hers), which are multitudinous references to eponymous juvenile heroine of the novel. The same is the case with RUL. Moreover, the co-ordinating conjunction and features as a key word in ENL, which indicates that the higher mean sentence length results from a conscious strategy of the author, who favoured long compound sentences with the said conjunction. The preposition with (cf. Table (10)), which was also revealed to be overrepresented in ENL in comparison with ENRC, is also a marker of Nabokovian style because the author frequently (40 times) starts his sentences with prepositional phrases functioning as objects or adjuncts. With people in movies I seem to share the services of the machina telephonica (…) With her right hand holding her left arm behind her untanned back, the lesser nymphet (…) With Linda Hall the school tennis champion, Dolly played singles at least twice With her brown bobbed hair, luminous gray eyes and pale skin, she looked perfect (…) With Lo's knowledge and assent, the two post offices given to the Beardsley post Table (10). Selected concordances with sentences starting with a preposition With As regards RUL, an interesting finding is revealed by the keyword status of the conjunction дабы (in order to, in order that, so that), which is synonymous with conjunctions чтобы, or phrases, such as для того чтобы, which are more contemporary and frequent in RURC. Nevertheless, according to Dictionary of Contemporary Russian Literary Language (Gorbachevitch, 1993) the conjunction TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 137 дабы is antiquated5 and it was used in the Russian literary language in the 19th century. In RUL, however, it occurs 19 times. The same is the case with the extensive use of the pronominal adverb6 столь (such), which occurs in RUL 42 times, while at the same time Ozhegov - Shvedova (2000: 770) define it as antiquated and literary. 4. Comparison of English and Russian Lolita 4.1 Descriptive statistics Hence no lemmatization was conducted on either ENL or RUL, the indicators such as the number of types, TTR and STTR are inflated for RUL and thus impossible to serve as the basis for comparison. Therefore, the analysis focused on the following indicators: number of tokens used for a wordlist, number of sentences and mean sentence length (in bold in Table (12) below). Statistics Number of tokens ENL 112,230 RUL 102,548 Number of types (distinct words) 13,991 29,651 Type/token ratio (TTR) (%/or X per 100 tokens) Standardized TTR (STTR) (%/or x per 100 tokens) 12.46 51.67 28.91 67.46 Mean word length (in characters) 4.40 5.44 Number of sentences 5,549 5,531 Mean sentence length (number of tokens) 20.22 18.54 Table (11). Descriptive statistics for ENL and RUL The data show that the Russian translation of Lolita is shorter than the original version by 10,000 tokens. In other words, on average only 914 Russian words were required to translate 1000 English words. At the same time, the number of sentences in ENL and RUL is almost identical (5,549 vs. 5,531, respectively). However, the data on the mean number of words per sentence show that on average Russian sentences are shorter than English sentences by almost two words (18.54 vs. 20.22, respectively). This difference was even more pronounced in the case of Russian and English reference corpora (10.34 vs. 17.22), which shows that sentences in RUL are untypically longer as compared with RURC. It seems that Nabokov-translator used long-form constructions to translate the novel into Russian and followed the long-sentence principle insisted on by himself in the English original. 4.2 Frequency profiles In order to determine whether it is English or Russian text that has more repetitions and lower lexical variety in terms of top-frequency words, a frequency profile proposed by Baroni (2006: 3-4) and described in greater detail in section 3.2 was used. Again a typical frequency profile was modified in that frequency information was substituted with information on cumulative percentage of the total word count (%cW) corresponding to frequency-based ranks. The results are presented in Table (12) below. Gorbachevitch (1993:125) provides example of use of conjunction дабы in Dead Souls („Мёртвые души” 1842) by Nikolai Gogol, and in letter to V. Polenov written by Alexander Pushkin on 25 Aug 1835. 5 ŁUKASZ GRABOWSKI 138 Rank 1 100 105 ENL %cW 4.593316 49.54556 50.05869 Rank 1 100 593 RUL %cW 3.3032 35.04949 50.01022 Table (12). Modified frequency profile for ENL and RUL Table (12) shows that in ENL 105 word types account for 50 % of the total word count, whereas in the case of RUL - 593 word types. As a result, there are 488 more word types in RUL which account for 50 % of the total word count. This text-specific data can be further generalized as the difference between distribution of the top-frequency words in English and Russian texts. English texts have more repetitions and lower lexical variety in terms of top-frequency words, whereas their Russian translations (and Russian texts in general) have fewer repetitions and higher lexical variety among top-frequency words, which seems to be another S-universal as regards English-to-Russian translation. 4.3 Frequency spectra In order to determine whether it is English or Russian text that has fewer repetitions and higher lexical variety among bottom-frequency words, a frequency spectrum proposed by Baroni (2006: 3-4) and described in greater detail in section 3.3 was used. Again a typical frequency spectrum was modified in that the frequency information was substituted with information on the cumulative percentage of the total word count (%cW) corresponding to words types with frequencies 1-25. The results are presented in Table (13) below. w (f) 1-25 ENL %cW 34.85 % w (f) 1-25 RUL %cW 54.46 % Table (13). Frequency spectra for ENL and ENRC The data show that RUL has fewer repetitions and higher lexical variety (i.e. words with frequencies 1-25 account for almost 55 % of the total word count) among bottom-frequency words than ENL (i.e. words with frequencies 1-25 account for 34.85% of the total word count). Although this relationship may be treated as another S-universal in English-to-Russian literary translation, it is not known how far that result is influenced by the lack of lemmatization conducted on English and, in particular, Russian language data (each inflectional form of a particular word type (e.g. genitive, accusative or locative case of the noun, in either singular or plural, feminine or masculine) is treated as a single occurrence of a type, which accounts for a typical problem of operating with non-lemmatized types and tokens in highly-inflectional languages, such as Russian or Polish). 4.4 Lexical wordlist comparison A major problem with information extracted from frequency lists is that raw frequencies highlight very common grammatical words, such as the, a, and, of. However, their occurrence is unlikely to provide evidence of any specifically used vocabulary across source and target language texts or characteristic features of translation. To remedy that, all grammatical words were deleted from the topfrequency items, and the most common content (lexical) words used in ENL and RUL, respectively, were isolated. Such a filtered out frequency list with 25 top-frequency content words is presented in Table (14). TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 Rank 9 30 37 39 42 51 61 62 63 81 84 87 89 93 95 100 101 102 103 108 113 114 115 116 120 ENL Word WAS HAVE SAID WERE LITTLE LO OLD LOLITA TWO KNOW HAZE WAY CHILD ROOM GIRL AM CAR GOOD HUMBERT EYES HAND MADE DAY FIRST LET Freq. 1486 388 344 305 287 236 197 193 192 139 137 135 132 121 120 117 117 116 115 109 104 104 103 103 98 Rank 31 32 38 41 43 58 62 64 74 87 88 95 101 102 103 106 108 111 117 122 127 131 134 139 141 RUL Word БЫЛО БЫЛ ТОЛЬКО БЫЛА ЛОЛИТА БЫЛИ МОЖЕТ МОГ СКАЗАЛ ВРЕМЯ СКАЗАЛА ДОЛЛИ ДВА ЛЕТ ГУМБЕРТ ГЕЙЗ СОВЕРШЕННО ДЕНЬ ЛО ДОВОЛЬНО ЕСТЬ ДОЛОРЕС ОДИН ВРЕМЕНИ ЛОЛИТЫ 139 Freq. 322 308 222 214 202 143 137 132 115 98 96 91 87 86 85 82 79 77 74 70 67 64 63 60 60 Table (14). 25 top-frequency content (lexical) words in ENL and RUL The data show, among others, the high rank and frequency of the reporting verb said, with 344 occurrences in ENL, which is more than aggregate frequency of the equivalent Russian reporting verbs сказал (masc.) and cказала (fem.), with 211 occurrences both. This observation shows that repetitions in Russian texts are regarded as a bad style, and therefore Nabokov used, among other techniques, synonymous reporting verbs to render said into Russian7 (e.g. подсказала, рассказала, высказал, заговорила, говорила, произнес/ла, проговорил/а, заявила, говорил, поговорил), or employed more specific and concrete verbs, such as выразился (‘expressed / voiced’), ответил (‘answered/replied’), объяснил/a (‘explained’), обещала (‘promised’), кричал (‘screamed’), подхватила (‘picked up’), объявила (‘announced’), обратилась [ко мне] (‘turned [to me]’), cпросила (‘asked’), процедил (‘muttered/mumbled’), воскликнула (‘shouted’), шепчет (‘whisper’) etc. It seems that the English verb said is not semantically determined and it is only context which allows for specification of its exact meaning. This observation refers to one of the specific problems of translation between English and other Slavic languages, namely the fact that Russian (and other Slavic languages) is more explicit semantically (i.e. the words have more specific meaning distinctions) than English, which in turn is more ambiguous and vague in its surface forms (Comrie, 1981: 7 The data presented below come from the analysis of parallel concordances. 140 ŁUKASZ GRABOWSKI 31-79), which can be treated as another S-universal in English-to-Russian translation. Hence, English language is more dependant on pragmatic and contextual information in specifying exact interpretation of its linguistic forms (e.g. a past tense verb said), which are broad in meaning. According to Piotrowski (1994: 95-96), users of English tend to choose lexemes which are broad in meaning rather than specific (even though English language has both broad-meaning and narrow-meaning words), whereas users of Russian (and probably other Slavic languages) tend to choose narrow-meaning lexemes, and therefore they regard texts with multiple repetitions as plain, simple, or even bad style. The side-effect to that is that proper translation of English reporting verbs (or broad-meaning English lexemes in general) requires that more lexical (content) words be used in order to produce a natural and acceptable text or translation into Russian. Moreover, it also shows that translators translate not words, but texts with words, which means that the number of possible translations is huge, perhaps even infinite. 5. Conclusions To conclude, the following T- and S-universals were revealed throughout the study: 1) T-universals: (i) RUL is more complex and specific lexically than non-translational language (RURC); (ii) sentences in RUL are untypically long as compared with RURC; (iii) RUL has fewer repetitions and higher lexical variety among bottom-frequency words, 2) S-universals: (i) Russian translation (RUL) is shorter than the English original (ENL) by almost 10,000 tokens; (ii) RUL has fewer repetitions and higher lexical variety among both top-frequency and bottomfrequency words than ENL; (iii) broad-meaning English words are frequently translated into Russian (in RUL) with words of more specific meaning. Nevertheless, it is essential to emphasize that the study presented herein was based on one translation only (RUL), which, to make matters worse, is the self-translation completed by Nabokov. Thus, the generalization of the findings presented above to all literary texts seems at best dubious unless the corpus of translated literary texts includes more translations completed by different translators (to avoid the translator's idiolect effect). Moreover, the compilation of corpus of translational Russian, with texts representing numerous different genres, may provide more comprehensive results. All in all, it is hoped that this study signals the need for more corpus-based or corpus-driven research on translation universals concerning Slavic languages, such as Russian or Polish, in line with the studies conducted by Scarpa (2006) or Xiao (2009). Such studies would enable one to examine other T- and S-universals, e.g. explicitation, simplification, normalization or conservatism, sanitization and leveling out, co-occurrence of features (as quoted in Olohan (2000: 90-104)), applicable to translation of texts of various genres into Slavic languages. References Baker, M. (1995), “Corpora in Translation Studies: An Overview and Some Suggestions for Future Research”. In: Target 7(2): 223-243. Baker, M. (2004). ,“A corpus-based view of similarity and difference in translation”. In: International Journal of Corpus Linguistics. Vol. 9/2: 167-193. Baroni, M. (2006), “Distributions in text”. In: Anke Lüdeling and Merja Kytö (eds.), Corpus linguistics: An international handbook, Berlin: Mouton de Gruyter. 1-6. Berber-Sardinha, T. (1999), “Using Key Words in Text Analysis: Practical Aspects”. In: DIRECT Papers 42, LAEL, Catholic University of São Paulo. Berber-Sardinha, T. (2000), “Comparing corpora with WordSmith tools: how large must the reference corpus be?”, Proceedings of the Workshop on Comparing Corpora - Volume 9. Hong Kong, 7-13. Boyd, B. (1993), Vladimir Nabokov. The American Years. Princeton: Princeton University Press. TRANSLATION AND MEANING, PART 10, 2013, 129 - 141 141 Chesterman, A. (2004).,“Beyond the particular”. In A. Mauranen and P. Kuyamaki (eds.) Translation Universals: Do they exist?. Amsterdam: John Benjamins, 33-49. Comrie, B. (1981), Language Universals and Linguistic Typology: Syntax and Morphology. Oxford: Basil Blackwell. Dunning, T. (1993), Accurate Methods for the Statistics of Surprise and Coincidence. In: Computational Linguistics, 19/1: 61-74. Gorbachevitch, K. (1993), Dictionary of Contemporary Russian Literary Language. Moscow [Горбачевич, К. 1993. Словарь современного русского литературного языка. Москва] Laviosa-Braithwaite S. (1995), “Comparable corpora: towards a corpus linguistic methodology for the empirical study of translation”. Translation and Meaning Part 3: 153-163. Laviosa, S. (1998), “Core patterns of lexical use in a comparable corpus of English narrative prose”. In: Meta Translators' Journal, vol. 43, n 4. 557-570. Nabokov, V. (1955 / 2000), Lolita. London: Penguin Books. Nabokov, V. (1989), Lolita: a novel. Moscow: Izvestia [Набоков, В. В. 1989. Лолита.Роман. Москва: Известия]. Olohan, M. (2004), Introducing Corpora in Translation Studies. Routledge: London. Ozhegov, S, Shvedova, N. (2000), Great Reference Dictionary of Russian. Moscow: Nauka [Ожегов, C., Шведова Н. 2000. Толковый Словарь русского языка. 4-е издание, дополненное. Москва: Наука]. Piotrowski, T. (1994), Problems in bilingual lexicography. Wrocław: Wydawnictwo Uniwersytetu Wrocławskiego. Scarpa, F. (2006), “Corpus-based Quality-Assessment of Specialist Translation: A Study Using Parallel and Comparable Corpora in English and Italian”. In: M. Gotti, S. Sarcevic (eds.), Insights into Specialist Translation. Frankfurt am Main: Peter Lang Verlag. Scott, M., (1996, 1997, 1999, 2004), Wordsmith Tools, Oxford: Oxford University Press. Trzeciak, J. (2005), Nabokov as self-translating author. Unpublished PhD dissertation. University of Chicago. Waliński, J. (2004), “Acquisition of terminology at the phrase level from comparable corpora”. In: B. Lewandowska-Tomaszczyk (ed.), Practical Applications in Language Corpora (PALC 2003). Frankfurt am Main: Peter Lang. 49-65. Xiao, R. (2009), “How different is translated Chinese from native Chinese? A corpus-based study of translation universals”. In: Proceedings of the Corpus Linguistics 2009 conference. Liverpool University, 21-23 July 2009. Retrieved on 20 August 2010 from: http://ucrel.lancs.ac.uk/publications/cl2009/105_FullPaper.doc. 142 TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 143 THE SEMANTICS OF NORTHERN SOTHO VERBS AS TRANSLATED FROM ENGLISH: COMMENTS ON THE AFRICAN LANGUAGES WORDNET PROJECT Mampaka Lydia Mojapelo University of Sourth Africa (UNISA), Pretoria, South Africa Abstract: African Languages WordNet is an ongoing project which is based on the English WordNet. WordNet is an electronic lexical database that groups words in synonym sets (synsets). In this project, words are translated from English to African Languages. As such, this paper comments on the semantic aspects of verbs in one of the African Languages, namely Northern Sotho, that are translated from English. Verbs are generally understood as expressions of action or state of being. These two languages are typologically dissimilar, with different cultural-historical backgrounds. Northern Sotho is a Bantu1 language, agglutinating with extensive and productive use of affixes while English is not. In the first place, structural differences between these two languages pose equivalence challenges, both linguistically and computationally. Secondly, verb equivalents may not be affected by the same collocational restrictions in the source and target languages. Another issue is that a verb in one language may invoke certain connotations, which may not apply to its equivalent in another language. Finally, some concepts may be foreign and others culture-specific to one language and not the other, thus resulting in omission of some target language concepts. Attention to these equivalence challenges may enhance technological development of the target language lexicon. 1. Introduction This paper is based on an ongoing African Languages WordNet project, which was started in South Africa in 2007. It is a collaborative project between Unisa’s Department of African Languages and Northwest University’s Centre for Text Technology. The project is based on existing global networks such as the English WordNet, Princeton (cf. http://wordnet.princeton.edu; Fellbaum, 1998). In this project, the synonym sets (henceforth referred to as synsets) are translated from English into African Languages. The envisaged result of the project is the WordNets of official South African languages – the lexical database that will support automatic text analyses and other applications. The languages currently involved in the project are Northern Sotho, Setswana, Zulu and Xhosa. The synsets translated and quality-controlled to date include nouns and verbs. The discussion in this paper, however, is limited to verbs, that is, the semantic aspects of Northern Sotho verbs within the context of WordNet. In other words, although WordNet is multidisciplinary in nature, the paper focuses specifically on the linguistic aspects of the project. A verb is simply defined as a word that expresses action or state of being. Yule (2003: 88), for example, defines this part of speech as “words used to refer to various kinds of actions and states involving the ‘things’ in events”. Semantically, verbs may be classified under those expressing change, change of possession, communication, existence, experience, contact, motion, creation, weather, bodily care and function, perception and social interaction (Du Plessis, 1999). Under each of these classes, there are narrower classifications according to semantic contents. Verbs can further be classified into main and deficient verbs. Lexical items that are used as deficient verbs require main verbs to form a complete predicate. Ziervogel and Mokgokong (1969) affirm that often a deficient verb is derived from an independent verb and it assumes a different meaning and a different role of modifying the main verb. Deficient verbs provide additional information to the basic 1 Bantu is used in this paper as a linguistic classification term. It bears no reference to the South African sociopolitical connotation. 144 MAMPAKA LYDIA MOJAPELO meaning of the main verb which is similar to what the adverb does. For instance, the verb stem -šala (remain, past tense: -šetše) can be used as a main verb or as a deficient verb, as illustrated below: 1) (a) (b) Main verb: i. Baratho ba Modupi [ba [šetše]. (Modupi’s younger siblings remained) ii. Baratho ba Modupi [ba [šetše] thabeng. (Modupi’s younger siblings remained on the mountain.) Deficient verb: Baratho ba Modupi [ba [šetše] [ba [boile]. (Modupi’s younger siblings are already back.) In example 1) (a) the main verb expresses “remained” and the main verb in 1) (b) expresses “are back”. -šetše in 1) (b) is a deficient verb stem, providing more information about the main verb. Furthermore, the verb in Bantu languages has a complex structure because of the languages’ agglutinating nature. Besides the basic meaning of the verb, morphological units that may be affixed to the verbal root or stem carry with them some additional semantic content. For instance, the causative verbal extension -iš- has two basic semantic interpretations: cause to and assist to, such as in: -loga (knit; crotchet; plait) > -log-iš-a > -logiša (cause to/assist to knit; crotchet; plait). The causation interpretation excludes the causer from the action while assistance implies the involvement of the assistor. However, the two semantic interpretations do not apply in every context, as illustrated by the following underlined translations: -rekiša < -rek-iš-a (sell), -sepediša < -sepel-iš-a (cause to walk; assist to walk - includes accompanying - ; walk hurriedly) and -botšiša < -botš-iš-a (ask). With regard to -bolediša < -bolel-iš-a (cause to talk; talk to), the causation interpretation is indirectly implied in talk to, in the sense that when you talk to someone you cause them to talk back. Verbs also assign thematic roles to their arguments. Jackendoff (1992:18) cites selectional restrictions as an example of semantic well-formedness conditions, and adds that knowledge of the language and/or of the real world are the bases on which violation or compliance to these restrictions occur. A verb’s subcategorisation conditions are therefore sensitive to semantic alignment with their arguments. For instance, if the verb stem -ya (go) acquires an internal argument, by virtue of its semantic content, its argument has to have a thematic role of location. The main aim of translation is to provide the semantic equivalent of the linguistic item in the source language, in the target language. However, according to Baker (1992), translators are often faced with challenges of non-equivalence. Given the above scenarios regarding the Northern Sotho verb and the verb in general, it becomes inevitable that challenges relating to equivalence will be encountered during a translation process. While differences in morphology and syntax are bound to impact on semantic equivalence, drawing from morphological and syntactic relevance of a particular language may inform decisions regarding equivalence. Owing to these and many other factors, the general-specific conceptual representation may not be linguistically in unison in the source and target languages. Languages have different grammatical structures, the speakers of languages have different historical and cultural experiences and all of these reflect on their languages. 2. Material and methods The source language in the African Languages WordNet is English and, in this case, Northern Sotho is the target language. The source language provides lexical entries, definitions and examples of usage. The domain of use and suggested upper merged ontology (SUMO) tags help to clarify the concepts. The examples used in this paper are sampled from 750 (three batches of 250 synsets each) verb entries. The envisaged end product will be equivalent entries in the target language, which are semantically related. The synsets will be edited, made available for inputs and, finally, the database will be open source; aimed at language development. TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 145 Apart from definitions and usages provided, available resources include the project manual detailing procedures for the translation of synsets and documented lexical support in the form of dictionaries. The constraints include the lexical limitation of dictionaries or of the target language, cultural-linguistic and structural dissimilarity as well as specific-general conceptual dissimilarity between the source and target languages. Although the ideal for good translation is that a translator should be both bilingual and bicultural, it is not always easy to find that kind of a match. However, in the case of this paper, both translation and quality control are done by mother tongue speakers with reasonable knowledge of the target language and culture. 3. Translation and equivalence Etymologically speaking, translation (from Latin translatus) is to carry over, to bring over, to remove from one place to another, to turn from one language to another (Online Etymology Dictionary). According to Nida and Taber (2003: 19), the primary aim of translation must be to reproduce the message. The process must reproduce in the target language the closest natural equivalent of the source language message in terms of meaning and style (Nida & Taber, 2003). They further differentiate between equivalence and identity by explaining that the process of obtaining equivalence carries over the message while obtaining identity preserves the form. This conforms to Nida’s two types of equivalence, namely: formal equivalence and dynamic equivalence. While the former preserves the form thereby bringing across the text literally, the latter carries over the central meaning expressed in the source language. The process of providing the semantic equivalent of the source text/lexical unit while preserving the natural features of the source language requires a lot of grammatical and lexical adjustment (Nida & Taber, 2003). Although Nida’s main translation involvement was with the bible, and therefore text translation, his theory is very relevant to lexical translation such as in the case of WordNet. Morphological and syntactical differences between the source and the target languages need to be observed so that, in conveying the semantic equivalent, the target language is not shortchanged in terms of grammatical structure. Equivalence is known to be the main challenge for translation. Al-Kasimi (1983:63) maintains that “absolute equivalents which have exactly the same semantics and grammatical functions in both languages are rare”. On equivalence, Newmark (1995: 90) asserts that it is “an unfortunately named term implying approximate equivalence”. According to Culler, the problem with equivalence is that “each language articulates or organises the world differently” (Baker, 1992: 10). Over the years proponents and opponents of the Sapir-Whorf hypothesis debated the issue of the connection between language and cognition. Jackendoff (1992), for instance, points out that the semantic representations are tightly integrated into the cognitive system of the human mind. The way in which different societies experience the world affect how their language is structured and presented, and vice versa. Equivalence challenges can be attributed to a number of factors. Among those, this paper comments and provides illustrations on factors relating to collocations, connotation, specificity, grammatical structure, foreign concepts and culture-specific concepts. Among sense relations that exist between linguistic items are broad categories of paradigmatic and syntagmatic relations. Paradigmatic relations, on the one hand, exist between linguistic items which are substitutable by one another in the same syntactic environment. Syntagmatic relations, on the other hand, exist between linguistic items which co-occur in the same syntactic relationship. These sense relations form the basis for semantic aspects that are under discussion in this paper. For instance, collocational restrictions apply at syntagmatic level while hyponymy is a paradigmatic relation of inclusion (cf. Cruse, 2000: 150). With regard to verbs, the term “troponym” was introduced and is used instead of “hyponym” (cf. Fellbaum, 1990; Fellbaum & Miller, 1990). Fellbaum (op. cit.: 285) explains the troponymy relation between verbs with the formula: “To V1 is to V2 in some particular manner”, whereby V1 is a troponym of the superordinate V2. 146 MAMPAKA LYDIA MOJAPELO The complexity of language is such that, as far as semantic content is concerned, one lexical item may be affected by more than one of the aspects mentioned above. Regarding language structure and expression, one language may need an adverb to express that which is inherently contained in the semantics of a verb stem in another language. Therefore, while meaning itself is a complex concept, not every verb stem will necessarily be translated as just a verb stem. Some verbs’ semantic contents inherently express manner and direction. By way of illustration, the following set of verb stems have the basic meaning of -sepela (walk) but each one also inherently carries additional information as to how or in which direction the act is being carried out – there is no need for an adverb to express that. -sepela is the superordinate or hypernym and the rest of the verbs in 2) (a) – (l) are troponyms: 2) (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) -nanya (walk slowly) -phakiša (walk hurriedly) -sepediša (walk hurriedly) -gwataka (walk fast defiantly or arrogantly) -nanabela (walk with the aim of catching or reaching for someone or something you stalk) -babaela (walk or tread like someone with sore feet; walk like a cat on hot bricks) -khukhuna (walk about or roam around as if a predator; crouch; prowl; slink) -nyonyoba (walk stealthily; walk smartly; stalk) -katakata (walk backwards) -tsatsanka (walk boastfully; swagger; strut) -sesella (walk about slowly and aimlessly as if ashamed of something; loiter) -tsherema (walk slowly as if making a great effort against psychological reluctance or physical incapacitation, without lifting feet) The superordinate -sepela entails troponyms (a) to (l), but is not entailed by them. Fellbaum (1990: 45) maintains that “lexical entailment is a unilateral relation: if a verb V1 entails another verb V2, then it cannot be that case that V2 entails V1”. She further explains that if the verbs are mutually entailing, then they are in a synonymous relationship and not in a superordinate-troponym relationship. In support of this view Croft and Cruse (2004: 142) suggest the following formulation for lexical entailment: “X is a hyponym of Y iff F(X) entails, but is not entailed by F(Y)”. The meaning of a troponym is therefore more specific and has additional meaning. While all of the verb stems have the semantic content expressed by walk, each of them also has an additional meaning that is not contained in walk and that they do not necessarily share. For the verb stems listed in 2) (a) – (l) above, some languages may need an adverb to capture some of added specific semantic contents inherently contained for Northern Sotho. Northern Sotho may also use a word group to express the meaning of a single lexical item from another language. With reference to the examples above, there are potentially many different manners of walking, but each language will encode some and not the others, such that no two languages encode exactly the same. As evident from 2) (j) above -tsatsanka corresponds to English “swagger/strut” while there may be no direct lexical equivalent of -babaela. Some of the semantic aspects that have a bearing on equivalence challenges, as were mentioned earlier in this section, will be illustrated in the following paragraphs. 3.1 Collocations Collocations are defined by Croft and Cruse (2004: 249) as “combinations of words that are preferred over other combinations that otherwise appear to be semantically equivalent.” Leech (1981: 17) explains collocational meaning as “what is communicated through association with words which tend to occur in the environment of another word.” Collocation restriction therefore refers to a situation where one lexical item may be barred from co-occurring with another because of the existence of an- TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 147 other one with the same conceptual meaning that is semantically suited to appear in that environment. In this project, lexical equivalents inserted are verb stems. In the grammar of the target language, verb stems are preceded by a hyphen (as reflected in the verb stems in 2) (a) – (l) above) to show that the prefixal morphemes that complete the predicate have been left out. This is so because the morphemes will be extensively varied, consistent with the nominal class system. For computational convenience in the project, the hyphens are left out of Northern Sotho verb stems. Therefore, from this point on, the verb stems will appear without hyphens. The examples in tables 1-5 relate to collocational restrictions. Table 1 shows two separate entries of the source language, dress and wear, defined as put on clothes and put clothing on one’s body, respectively. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v dress, get dressed apara, tšwara, tšwala, rwala put on clothes Can the child dress by herself? Fashion Putting v wear, put on, get into, assume apara, tšwara, tšwala, rwala put clothing on one’s body He got into his jeans. Fashion Putting Table (1) Both entries have the target language equivalents: apara, tšwara, tšwala and rwala. The equivalents represent the same conceptual meaning as the source language synsets, as their definitions seem to require. These Northern Sotho verb stems are, however, subject to collocational restrictions. For instance, rwala cannot substitute tšwara in a syntactic environment. The semantic content of the verb may select a specific noun as internal argument and other nouns may be prohibited from appearing in the syntactic environment. Tšwara and tšwala are dialectal variants. While apara is a superordinate, the other three (or two) are collocationally determined troponyms – their usage will depend on the type of garment and, more specifically, the part of body being covered or clothed. Tšwara/tšwala collocate with pants and underwear; rwala with hat, earrings and shoes. For the upper body, the general word is used. Figure 1 illustrates the semantic relation of the three lexical entries. English may not encode the concepts in a similar way. [superordinate] apara (wear; dress) tšwara/tšwala rwala [troponyms] Fig. (1): Superordinate-troponym relation Table 2 illustrates other two separate entries. POS Eng_Synset Tgt_Language Synset Eng_Definition Eng_Usage Domain SUMO v reproduce; procreate; multiply tswala, belega, hlatša, phaphaša have offspring or young The Catholic Church tells people to procreate, no matter what their economic situation may be. biology Sexual Reproduction 148 MAMPAKA LYDIA MOJAPELO POS Eng_Synset Tgt_Language Synset Eng_Definition Eng_Usage Domain SUMO v give birth; deliver; bear; birth; have tswala, belega, hlatša give birth (to a newborn) My wife had twins yesterday. zoology Sexual Reproduction Table (2) Northern Sotho equivalents tswala; belega; hlatša and phaphaša are conceptual equivalents; but they differ in terms of collocations. Phaphaša is excluded from the second synset because it means hatch. It applies to fowls and egg-laying reptiles. Tswala applies to cows, goats and other such animals, belega to human beings and hlatša to dogs and cats – they litter. Therefore, all the Northern Sotho equivalents appear for the first synset because the concept is general. Other source language entries whose equivalents are subject to collocational restrictions are in table 3. POS Eng_Synset Tgt_Language_ Synset Eng_Definition Eng_Usage Domain SUMO v have na le; ba le; tšwa suffer from; be ill with She has arthritis. factotum ProbabilityRelation Table (3) The target language’s first two equivalents, na le and ba le, are subject to grammatical structure, more specifically the verbal form or mood. Example 3 illustrates the use of the two verbal forms: 3) (a) (b) (c) (d) Ke na le maphone. (I have blisters.) – [Indicative mood] Ke na le bolwetši bja pelo. (I have heart disease.) – [Indicative mood] Go ba le maphone go palediša motho go šoma. (To have blisters makes one unable to work.) --- [Infinitive mood] Go ba le bolwetši bja pelo go a tšhoša. (To have heart disease is frightening.) – [Infinitive mood] The equivalents in table 3 translate back as have, but it is not the only way of expressing having an illness. Other ways of expressing the same concept, subject to collocational selections, are the following: bolawa ke/bolaya ke*2, swarwa ke/swara ke*, opša ke/opa ke*, rengwa ke/rema ke*, longwa ke/loma ke* and tswenywa ke/tshwenya ke*. They are linguistically represented in the passive construction, with the body part or type of illness serving as “agent”. For example, 4) (e) Ke swerwe ke mokgohlane. (I have flu – literally “I am caught by flu”; not “I caught flu”) The choice of any of these verb stems will be determined by the affected body part or the type of illness. Bolawa ke/bolaya ke* and swarwa ke/swara ke* are hypernyms with troponyms opša ke/opa ke, which will select head (headache) or tooth (toothache), rengwa ke/rema ke* will collocate only with head (headache), longwa ke/loma ke* will select tummy (tummy ache). Tšwa (literally come out) is not used with the name of the body part affected, it is rather used with the name of the disease which can be visually attested. The past tense of tšwa is tšwile, also written as tšwele. The past tense is the most commonly used in this case. It expresses the state of the patient. For example, 2 The asterix sign is used to mark the forms prevalent in spoken language – cf. Kosch (2003). TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 5) 149 Kedibone o tšwile/tšwele mauwe. [Kedibone (she) came out mumps*], expressing: [On Kedibone, mumps came out*], simply meaning: (Kedibone has mumps). The following illnesses/conditions can replace mauwe in example (5) above: digaroga (rash), sekaku/lentshwe/lentsho/sekutu (boil/abscess), sepšhatlapšhatlane (chicken pox), mooko (measles), and other visual types of illnesses and conditions. Very closely related to the expressions in the previous paragraph are the following, which in a very narrow sense express the type or nature of pain or discomfort, issues which will inevitably link to the body part likely to be so affected: 6) (a) (b) (c) (d) thunywa ke; thunya ke* (as of tooth ache or painful bone) tšhatšhamelwa (ke); tšhatšhamela (ke)* (burning pain on the skin) hlohlonwa (ke); hlohlona (ke)* (itch) hlabja ke; hlaba ke* (literally stabbed by, which is figurative in nature but never substituted by other forms. It selects madi (literally blood) and mašimatho as “agents”.) The following examples, with transliterations, illustrate syntactic representations of (6) (a) – (d) above: (aa) (bb) (cc) (dd) (ee) Ke thunywa ke leino [I am pained by tooth]. Ke tšhatšhamelwa ke menyabidi [I am burnt by allergic reaction]. Ke hlohlonwa ke seatla [I am itched by hand]. Ke hlabja ke madi [I am stabbed by blood]. (to experience internal muscular or organic pain in the torso). Ke hlabja ke mašimatho [I am stabbed by ‘stitch’]. (to experience athletics-related muscular pain on the sides of one’s abdominal wall – ‘stitch’). It becomes clear from the examples above that, in such contexts, Northern Sotho does not lexicalise possession of the body part involved (for body parts syntax, cf. Mojapelo, 2007: 123-124). Table 4 contains other Northern Sotho equivalents, goela; feka; kopa; robala (le); robalana; nyoba; nyobana, that are subject to collocational restrictions. POS Eng_Synset Tgt_Language_ Synset Eng_Definition Eng_Usage Domain SUMO v copulate; mate; pair; couple goela, feka, kopa, robala, robala le, robalana, nyoba, nyobana make love Birds mate in Spring. sexuality Sexual Reproduction Table (4) The first three verbs would select examples such as kgomo (cattle), mpša (dog), kgogo (chicken), respectively. The last four (which are actually two basic verbs) apply to human beings. Syntactically, all of these verbs can only select a [+male] external argument while the internal argument is normally [-male]. As far as human beings are concerned, the verbs can also express reciprocation by adding extensional suffix -an-. 150 MAMPAKA LYDIA MOJAPELO Table 5 is the last example illustrating collocational selections/restrictions. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v chew, masticate, manducate, jaw sohla, sotla, hlahuna, phura, phuphura, otla, sesena chew (food) The cows were masticating the grass. gastronomy Eating Table (5) The concept concerns a step in the digestive process. Northern Sotho sohla, sotla, hlahuna, phura, phuphura and otla’s collocational selections involve also the texture of the object and the stage of the digestive process. sohla/sotla/hlahuna generally collocate with soft objects. Sesena is when the front teeth are used to chew without putting the whole object in the mouth. Phura and phuphura collocate only with hard objects like bones, hard candies and pits of fruit and vegetable. Only when the object has been gnawed and crunched can it collocate with sohla/sotla/hlahuna. Otla means to chew the cud. The situation, therefore, is that one or more English synsets will have a number of same Northern Sotho words as equivalents, but which will not necessarily substitute one another in all contexts. They are collocationally restricted to certain arguments. Another factor, with the exception of what Baker (1992: 13) terms propositional meaning, is “what other semantic implications do speakers of a language as a group or as individuals attach to a particular linguistic item?”. The next section will address those kinds of associations. 3.2 Connotation A word’s basic meaning is termed its conceptual, connotative, denotative or logical meaning (cf. Leech, 1981: 9, 23), similar to Baker’s (1992) propositional meaning. According to this kind of meaning assignment, a word is defined by semantic features that make its referent differentiable from others. That is, the meaning can be conceptualised by componential analysis. It covers the “basic, essential components of meaning which are conveyed by the literal use of a word” (Yule, 2003: 114). Over and above the conceptual meaning of a word, there are other implications or associations present. “Connotations are relatively unstable … they vary according to culture, historical period and the experience of an individual” (Leech, 1981: 13). Examples in table 6 illustrate this type of meaning. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v reduce; melt off; lose weight; slim; slenderize; thin; slim down phophothega, fokotšega, fela, ota, gwamelwa take off weight [empty] body_care Removing v gain; put on oketšega, akola, šušumoga, nona, increase (one’s body weight) She gained 20 pounds when she stopped exercising. body_care Increasing Table (6): Connotation Equivalents of the first synset phophothega; fokotšega; fela; ota; gwamelwa are subject to connotations, and in table 6 they are arranged from the neutral or the least negative down to the most in- TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 151 clination towards negative connotation. Fela is subject to how the speaker uses it; it can carry either a positive or a negative connotation. To induce an unambiguously non-negative connotation, often loan words are used. Compare the following sentences of example 6 with the same conceptual meaning, namely “Mosima has reduced weight”: 7) (a) (b) Mosima [o [slimile]. (From English slim) Mosima [o [gwametšwe]. In contrast to 7) (a), (b) paints an unsightly picture of Mosima’s weight reduction. 7) (a) depicts Mosima’s weight reduction as pretty, unless the remark is meant to be sarcastic. The antonym synset gain; put on defined as increase (one’s body weight) has in the target language oketšega, šušumoga, akola and nona. In the first place it depends on the cultural-historical orientation or the mental orientation of the speaker, the hearer and the society; whether having a big body is viewed in a positive light as compared to having a small body. Whether a big body is a reflection of good life, healthy life, neglect or unhealthy life will call for selection of a word which will either be neutral or reflect envy, a compliment, disgust, pity or sarcasm. The situation triggers a mental continuum ranging from obese to skinny, with either the neutral, positive or negative connotation attached. The connotation also reflects the speaker’s feelings or attitude either towards the person talked about or towards the particular body size or condition. In giving an equivalent, all of them are relevant but the selection for use will be determined by connotations. 3.3 General-specific conceptual (mis)alignment A concept may be general in one language while it is specific in another language and vice versa. Another scenario is that a general or specific concept may possibly not be lexicalised or may be represented as a word group in another language. Tables 7-9 illustrate this point. In table 7 the source language synset has target language equivalents imiša, senya and roba leoto. While imiša is general, senya’s meaning is specific. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v impregnate; knock up; bang up; prang up imiša, senya; roba leoto make pregnant He impregnated his wife again. biology Sexual Reproduction Table (7):generality or specificity in the source and target languages may not necessarily correspond The semantic content includes the relationship between the parties involved as well as the state of the female prior to the condition. Senya happens only to a first time unmarried mother-to-be; roba leoto is its figurative counterpart. Table 8 illustrates a case where the target language lacks lexical expression of the general concept expressed in the source language. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v administer, dispense fa, nweša, noša, tlotša, tsikitla, tšhela give or apply (medications) [empty] medicine TherapeuticProcess Table (8): The target language may lack a lexical expression of the general concept expressed in the source language 152 MAMPAKA LYDIA MOJAPELO Equivalents fa, nweša/noša, tlotša, tsikitla and tšhela can be back-translated as give; make to drink (including pills); apply for example, cream or ointment, rub and pour, respectively. Northern Sotho does not lexicalise the general concept administer. Fa (give) will be inappropriate since it will exclude apply and pour. In table 9 the target language lacks lexical expression of the specific concept expressed in the source language. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage v widow bolaela Cause to be without spouse The war widowed many women in the former Yugoslavia. Domain SUMO Death Table (9): The target language may lack a lexical expression of the specific concept expressed in the source language In this case, Northern Sotho does not lexicalise the specific concept. Lexicalisation is general, and the internal argument will shed light on who is it that has been killed. Northern Sotho word for kill is bolaya. The equivalent will require affixation of applicative -el- > bolaela followed by the word for husband, which will perfectly be represented by a word group rather than a single lexical item. Applicative -el- in this case means “to the detriment of”’. Alternatively, a phrase which will be backtranslated as “change/turn her into a widow”, fetola/fetoša mohlologadi, will convey the same meaning. Another scenario is that the target language verb may require assistance of an adverb to express the concept from the source language, as illustrated in table 10. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v revolutionize; revolutionise; overturn fetoša kudu, fetoša ka maatla change radically Email revolutionised communication. factotum Process Table (10): The target language may need an adverb to express the concept only by a verb in the source language In this case, Northern Sotho requires the assistance of an adverb to modify a word for change, fetoša, in order to capture the scale of the act or process, with phrases [fetoša [kudu] and [fetoša [ka maatla]. The fact that an adverb would be needed to express the scale of the act or that a concept may be expressed differently is an indication that languages are structured differently. The next section concerns grammatical structure, which also reflects on equivalence challenges. 3.4 Grammatical structure While Northern Sotho is a Bantu language of the Niger-Congo family, English is an Indo-European language. The grammatical structures of the two languages differ on many levels, including the agglutinating nature of Northern Sotho. The verbal affixes in Northern Sotho are extensively productive, including the shades of meaning that they add to the basic meaning of verbs. From the English definition and usage in table 11, the Northern Sotho equivalent tsentšha/ tsenya will apply only to the context of the example of usage provided. TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 153 POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v get; let; have [empty] cause to move; cause to be in a certain position He got a girl into trouble. factotum Process Table (11): Lack of one-to-one corresponding lexical expression due to difference in grammatical structure On the other hand, the causative morpheme -iš- affixed to whatever verb that expresses the act that the subject is being caused to do seems to include what is represented in the definition. The challenge in entering affixes as equivalents is that the equivalent is expected to be a verb stem, which should not have hyphens around it. Secondly, while leaving out the hyphens may misrepresent the entry as a verb stem, retaining the hyphens may interfere with the computational part of the project (the programming script that converts the Excel file into XML must be able to read the data properly). To avoid misinterpretation, the target language synset slot is left open. 3.5 Foreign concepts Another challenge in finding an equivalent is when the entry in the source language represents a concept that is foreign to the target language. An example of this is in table 12. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v tease; fluff [empty] ruffle (one’s hair) by combing towards the ends towards the scalp, for a full effect. [empty] body_care ShapeChange Table (12): Lack of lexical expression due to the concept being foreign to the target language Hair can be ruffled to make it untidy with equivalent: hlahlamola, but it does not seem to be what this entry is about. This seems to imply some relative length between the scalp and the end of the hair. One wonders what the relevance of the concept is for Northern Sotho: what full effect? Language reflects the experience of the speakers, be it anatomic, cultural or historical. The concept seems foreign to the target language. 3.6 Culture-specific concept According to Al-Kasimi (1983: 62), there is a close relationship between language and culture. Baker (1992: 21) identifies culture-specific concepts as one of the challenges for finding an equivalent in translation, but her point of departure in this respect is the source language. She explains culture-specific challenges as situations where the linguistic item in the source language is culture-specific, and is not culturally relevant to the target language. In this sub-section, though, culture-specific concepts are being viewed from the target language perspective. Firstly, a concept may be familiar, but due to culture-specific factors can be expressed only idiomatically as far as humans are concerned, or else it can be perceived as an insult or abhorrent to say. Secondly, as translation progresses, some gaps emerge regarding some senses that are not necessarily peripheral in the target language, but have not been catered for, probably because they are not central in the source language. The case of ‘miscarry’ in table 13 illustrates this point. 154 MAMPAKA LYDIA MOJAPELO POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v miscarry folotša , goma tseleng, boa tseleng suffer miscarriage [empty] biology Death Fig. (13): Non-lexicalisation or idomatisation due to cultural observance Northern Sotho equivalents are folotša, boa tseleng and goma tseleng. Folotša is the only nonidiomatic one, but it is never used for persons. Only animals folotša. The idiomatic ones used for persons literally mean “turn back while on the way” and this implies that the subject, namely the pregnant woman, has not been able to reach her destination. In terms of pregnancy, boa/goma tseleng (turn back while on the way) means not running full term. Another factor is that the translation process may reveal a gap that may never be filled in the target language because the concept is culture-specific. Multiple appearance of a specific verb as the target language equivalent of various source language entries brings several concepts involving the verb to mind. Most of these concepts may find the way in, but some may never appear. While swara appears several times as the target language equivalent, as illustrated in table 14, it becomes clear that some senses linked to the verb may never be included in the project. POS Eng_Synset Tgt_Language_Synset Eng_Definition Eng_Usage Domain SUMO v catch khetšha; swara be the catcher Who is catching? baseball Sport v catch; grab; take hold of swara take hold of so as to seize or restrain or stop the motion of Grab the elevator door! factotum Touching v handle; palm swara touch, lift, or hold with the hands Don't handle the merchandise. factotum Touching v seize; prehand; clutch swara take hold of; grab Birds of prey often seize small mammals. factotum Grasps v seize thopa; swara take or capture by force The rebels threaten to seize civilian hostages. factotum Unilateral Getting v collar; nail; apprehend; arrest; pick up; nab; cop swara; golega take into custody The police nabbed the suspected criminals. law Imprisoning Table (14): The same verb as an equivalent of various source language entries For example, among the meanings of -swarega < -swar-eg-a there are cases wherein the conventional meaning of neuter -eg- “potential” does not apply. Affixed to the verb stem -swara, the neuter extension brings out different meanings such as “to be held up and be unable to perform a certain duty or to go somewhere, to be very ill, to have a need to respond to the call of nature, and to be caught in the act and be unable to move”. The last semantic interpretation is also associated with witchcraft – likened to being caught with a hand in the cookie jar and freezing when caught. Whether or not the situation is possible is beside the point; the point is that it is part and parcel of the idiom of the language. A similar situation involves the verb stem -bona (see) and its extensional affixes. While the verb stem and its extensions would serve as equivalent to senses such as see, visible, understand and real- TRANSLATION AND MEANING, PART 10, 2013, 143 - 156 155 ise, there may not be a chance to capture, for instance, all senses of -bonela < -bon-el-a. Applicative affix -el- has semantic interpretations: on behalf of, for the benefit of, to the detriment of, onto and towards. One of the senses, which is culturally significant, may not be lexicalised in the source language as it is in the target language. It means “seeing private parts or undergarments of . Traditionally, a child would know that he/she must look away when, for instance, he/ she sees an older person standing next to a footpath relieving him/herself so that he/she does not -bonela them – and the myth attached to the taboo is that should anyone not abide by that they will turn blind. Young girls are also taught correct postures for sitting down, bending and standing up from the sitting position so that no one -bonela them. This interpretation is culture-specific, being part of traditional informal curriculum for raising children with respect for themselves and others in a society that upholds a certain level of moral standards. One may argue that such concepts are outdated and that societies have moved on with times, but leaving out cultural fundamentals is tantamount to destroying a language. Taking note of such gaps may ensure that in one way or another some target language gaps get filled. Culture-specific concepts or words like the two discussed above do not hamper the translation process, but they are conspicuously missing. 4. Conclusion With the challenges of equivalence discussed and illustrated in this paper, it will be interesting to see how the pre-editing stage is going to unfold; and how the resultant trial ontologies will prove and provide. It remains to be seen whether semantic aspects such as collocational restrictions, connotations, general-specific, structural differences, foreign concepts and culture-specific and other (non)equivalence issues will enrich or impoverish the languages the project seeks to advance. As translation progresses, some lexical items and/or concepts are clearly missing, a fact which points to the two languages’ different cultural-historical backgrounds, which in turn affects their linguistic make-up. The spin-off of this is that it may provide an opportunity to have original words of the target language also documented and preserved. It may also be interesting to experiment with a situation where the target and source languages swop places. While Northern Sotho keeps technologically abreast with other languages, it is also important that the language’s idiom does not fall through the cracks. The project should ideally be a useful tool to also document, preserve and contextualise words and concepts that are original to the target languages. Acknowledgements I acknowledge with gratitude the helpful comments from of the following scholars, who read earlier drafts of this paper: 1. Prof James Mafela (Unisa): first draft 2. Prof Christiane Fellbaum (Princeton): pre-final draft. References Al-Kasimi, A.M. (1983), Linguistics and bilingual dictionaries. Leiden: EJ. Brill. Baker, M. (1992), In other words: a coursebook on translation. London & New York: Routledge. Chow, I.C. & Webster, J.J. (2008), “Supervised clustering of the WordNet VerbHierarchy for systemic functional process type identification”. ICGL 2008 International Conference on Global Interoperability for Language Resources, 51 – 58. Croft, W & Cruse, D.A. (2004), Cognitive linguistics. Cambridge: Cambridge University Press. Cruse, D A. (2000), Meaning in language. Oxford: Oxford University Press. Du Plessis, J.A. (1999), Lexical semantics and the African languages. Stellenbosch: Stellenbosch University. Fellbaum, C. (Ed.) (1998), WordNet: An electronic lexical database. Cambridge, Mass: MIT Press. Fellbaum, C. (1990), “English verbs as a semantic net”. International journal of Lexicography 3(4):278-301. Fellbaum, C & Miller, G.A. (1990), „Folk pshychology or semantic entailment? A reply to Rips and Conrad (1989)”. Psychological review 97:565-570. 156 MAMPAKA LYDIA MOJAPELO Jackendoff, R. (1972), Semantic interpretation in generative grammar. Cambridge: MIT Press. Kosch, I.M. (2003), “Sandhi under the spotlight”. South African Journal of African Languages 23(3) 144153. Leech, G. (1981), Semantics. Middlesex: Penguin Books. Mojapelo, M.L. (2007), Definiteness in Northern Sotho. Unpublished D.Litt. dissertation. Stellenbosch: University of Stellenbosch. Newmark, P. (1995), A textbook of translation. New York: Phoenix. Nida, E.A & Taber, C.R. (2003), The theory and practice of translation. Leiden: Koninklijke Brill NV. Poulos, G & Louwrens, L.J. (1994), A linguistic analysis of Northen Sotho. Pretoria: Via Afrika. Yule, G. (2003), The study of language. Cambridge: Cambridge University Press. Ziervogel, D, Lombard, D.P & Mokgokong, P.C. (1969), A handbook of the Northern Sotho language. Pretoria: Van Schaik. Ziervogel, D & Mokgokong, P.C. (1975), Pukuntšu ye kgolo ya Sesotho sa Leboa. Pretoria:Van Schaik. Web references Online Etymology Dictionary Available at http://www.etymonline.com/ accessed on 2010-05-28. http://en.wikipedia.org/wiki/Linguistic_relativity accessed on 2010-08-31. http://wordnet.princeton.edu accessed on 2010-04-02 TRANSLATION AND MEANING, PART 10, 2013, 157 - 163 157 TRANSLATION AND FORMS OF ADDRESS: AN EXPERIENCE IN TSHIVENḒA Munzhedzi James Mafela University of South Africa, Pretoria, South Africa. Abstract: Throughout the entire world, languages are characterized by different forms of address, forms of address that are based on both linguistic and cultural dissimilarities. As a result, languages have different ways of showing respect. Addressing elderly people in the Vhavenḓa society involves the use of quite different forms of address from those used to address younger people. In this community, an elderly person will always be addressed in a respectful way, whether the addressor knows the person or not. In Tshivenḓa, for example, this respect is shown by the use of the honorific ‘Vho-’, which is prefixed to the person’s name. In many instances the use of the honorific Vho-, however, leads to translation problems when it comes to translating from Tshivenḓa into, for example, English. The aim of the paper is to highlight the translation problems caused by the forms of address used in Tshivenḓa. 1. Introduction The use of language differs from one language group to another. Languages often have different forms to refer to one and the same thing. This could be due to linguistic and cultural differences. Whatever the reasons, such differences are the source of a number of translation problems. Translation problems become evident when translators translate documents or literary texts from one language into another. The basic aim of translation is to coordinate the lexical units of one language with those of another language, which are equivalent in their lexical meaning (Zgusta, 1971). However, there will always be differences between the source and target language, as this article will show. The differences become more visible, for example, when one translates forms of address from one language into another. The problem of translating forms of address lies in cultural awareness. Azarkowska (2005:1) states: “It comes near to stating the obvious that translation difficulties regarding address form stem from the differences between the two languages and culture.” Dickey (2007: 7) cites Mehotra (1981), who argues that the rules governing address usage in various cultures are often complicated, and it is difficult to work out which factors do or do not influence the choice of address. Parkinson (1985) defines forms of address as words used in a speech event that refer to the addressee of that speech event, words that may convey extremely important social information. Parkinson (1985: 3) further writes: Naturally, different speech communities may make different uses of terms of address as a whole, some using them only occasionally as relatively minor conveyors of social information, and others using them constantly and having them carry a relatively large portion of the social communicative load. According to Akindele (2008), terms of address are significant linguistic mechanisms by which a speaker’s attitude toward, and interpretation of, his or her relationship with a speaker is reflected. The author further writes that inappropriate choice of address hinders good communication between the speaker and the hearer. On the other hand, Warren-Rothlin (2007) sees forms of address as strategies for communicating politeness. However, this scholar views the strategies for communicating politeness as highly culture-specific and as relating closely to broader cultural norms. Forms of address play an important role in communities. Brown/Levinson (1978), as cited by Akindele (2008), state that the importance of address forms cannot be overestimated in the use of language in any human society, because they serve as an indicator of the social relationship between a 158 MUNZHEDZI JAMES MAFELA speaker and a listener in terms of status and social distance. Forms of address constitute an important part of everyday communication in every culture, and they should not, therefore, be disregarded in translation and translation training (Szarkowska, 2005: 1). Adler (1978: 172) states: The forms of address give us valuable insight into a particular society. It is assumed that all speech communities have grammatical and lexical differentiation in forms of address (e.g. pronouns, names, titles) the use of which directly symbolizes differences in status, role, setting and affective factors. Translators do not always know when certain address forms are used in another culture. Languages have different ways of showing politeness; for example, some use plural pronouns, while some may use honorific terms, titles or even first and last names. Tshivenḓa has its own way of showing politeness. Of all the African language communities in South Africa, the Vhavenḓa are the most sensitive about address forms, especially as far as politeness is concerned. A younger person needs to know that an elderly person should be addressed in a respectful manner, because addressing the elderly is quite different from addressing the younger people in that community. An elderly person will always be addressed respectfully; it does not matter whether the addressor knows the person or not. The aim of this paper is to highlight the translation problems caused by certain forms of address in the translation of Tshivenḓa texts into a language such as English. The discussion will focus on the honorific Vho-, the pronoun, and concord as forms of address in Tshivenḓa. According to Ziervogel, Wentzel and Makuya (1972), the prefix vho- of class 2(a), as well as the concords and pronouns of class 2(a) (both plural), are often used in the singular to show respect. 2. The honorific VhoThe honorific is a term used to confer honour or show respect to the addressee. Adler (1978) states that honorific speech is the distributive style of speech, and involves special vocabulary and often special features of delivery, which social inferiors are expected to use toward their social superiors in certain societies. In Tshivenḓa, the plural prefix vho- is used to refer to a single person in class 2(a) and is an honorific term. The honorific Vho- is a general term used to show respect. It is equivalent to Mr and Mrs in English, and it does not imply any specific profession. When prefixed to a name, that is, the first name or the last name (surname), it shows respect. According to the Tshivenḓa culture, it is unheard of for a younger person to address an elderly person without using the honorific Vho-. As indicated above, the honorific Vho- is prefixed to the name of the addressee to show respect; for example: Vho-Munaka, Vho-Muelelwa, Vho-Rasila (Mr/Mrs Munaka, Mr/Mrs Muelelwa, Mr/Mrs Rasila). Many Tshivenḓa names can be bestowed on both boys and girls (i.e. they are not gender-specific). The equivalent of the honorific Vho- in this regard could be Mr or Mrs, depending on the context. It should be emphasized here that Tshivenḓa is an exception when compared with other African languages, because the honorific Vho- can also be prefixed to the first name to show respect; for example, Vho-Maria, Vho-Fulufhelo, Vho-Mashudu, and so forth. However, there are no English equivalents of the honorific Vho- prefixed to the first name. The equivalent will therefore be Maria, Fulufhelo and Mashudu respectively. The situation sketched above becomes problematic if Tshivenḓa texts entailing honorific Vho- are translated from Tshivenḓa into a language such as English. It will be difficult to identify whether the equivalent of the honorific Vho- is Mr or Mrs if the text itself does not make this clear. Without this knowledge, Vho-Masindi could be either a man or woman. The translator will have to depend on the context in order to find the correct equivalent. The extracts below will be analysed to illustrate this argument. Uḽa Mugivhela vho mbo ḓi swika … Vho-Masindi na murathu wavho na khotsi Vho-Rasila na mme Vho-Muhanelwa khevha vho dzula. Vho dzula nḓuni ya mukalaha. Vha vuwa vhe mukalaha vha ri: “Masindi, hezwi ni tshi vhona ndi tshi tou vhuya nda dzhena nḓila nda ḓa hangei haṋu, he nda ri ni ye TRANSLATION AND MEANING, PART 10, 2013, 157 - 163 159 ni vhidze murathu ni ḓe ri dzule ri tou rali, ndi uri nṋe a thi na ṅwana wa mme, ndi ndoṱhe. Zwino mashaka anga ane nda amba nao mafhungo ndi voinwi. Zwino hafhano ndo ni vhidza nga fhungo ḽa uri no zwi vhona uri khaladzi aṋu o lovha, ro mu swiṱa. A re nṋe ndo fulufhuwa zwauri o ṱuwa tshoṱhe. Ha tsha ḓo dovha a vhuya. Zwino sa izwi mafhungo o tou rali, uyu ṅwana wa vhathu u tou itwa hani? Anga ndi aneo Masindi ṋwananga.” Vho-Muhanelwa vhe fhaḽa vhe, “Khezwo ni tshi khou zwi pfa, a ni fhinduli muṋe waṋu? “Mmawe, habe hayo mafhungo ndi mahulwane ha ṱoḓi u skou fhufhelwa lini musi muthu a tshi fhindula. Na Sundani khoyu e hone, u a ḓi thoma u fhindula, na nṋe nda ḓo fhindula. Khezwo Sundani.” Ndi Vho-Masindi vha tshi khou kumedza murathu wavho iḽi fhungo ḽo vhewaho nga mukalaha VhoRasila (Sigogo, 1995: 37 – 38) Mukalaha zwino vha vho vhonala sa muthu o renguluwaho. Vha pfa iḽi ḽa mukegulu ḽi ḽone ḽine ḽa nga shumea ngaḽo. Vho no ḽi haseledza ndi hone vha tshi ruma avha Vho-Sundani uri vha ye vha vhidze Vho-Maria. Nga tshifhinganyana Vho-Sundani na Vho-Maria na vhone khevha vho no dzula nḓuni (Sigogo, 1995:40). (It was on a Saturday when they arrived … Masindi, her younger sister, their father Mr Rasila and their mother Muhanelwa are seen seated in the old man’s hut. The old man said: “Masindi, when you see me undertaking a journey and come to your homestead, where I said that you must go and invite your younger sister so that we can come together like this, it is because I do not have relatives, I am alone. Therefore, my relatives with whom I can discuss matters are you people. I have invited you here because of the matter concerning the death of your brother which you witnessed, and who has been buried. As for me, I am convinced that he has gone forever. He will never come back. Since the matter is like this, what do we do with this child (his widow)? That is all from me, Masindi my child.” Over there Muhanelwa says, “You have heard. Don’t you respond to what your father has just said?” “Mother, this is a serious matter, it needs to be attended to carefully. Sundani is also present, she can be the first respondent, I will also respond. There you are Sundani.” This is Masindi telling the younger sister about the matter raised by old man, Mr Rasila. Now the old man is seen as a person who is free. He thinks that the old lady has said something worthwhile. After discussing it, they asked Sundani to go and call Maria. Within a short period of time, Sundani and Maria joined them in the hut.) In Tshivenḓa literary works, as in a real-life situation, elderly people are addressed with the use of honorific Vho- prefixed to their name in order to show respect. This leads to translation problems for translators who translate from Tshivenḓa into, for example, English. In the passage above, readers learn about the characters Masindi, Muhanelwa, Rasila, Sundani and Maria. All these characters are members of the Rasila family. Old man Rasila, his wife Muhanelwa, and their daughters Masindi and Sundani, are gathered in Rasila’s hut to discuss a serious matter. Maria, Rasila’s daughter-in-law, is seated outside the hut. As we can see, in certain instances the names of these characters are prefixed with the honorific Vho-, whereas in other cases they are not. It should be noted that the honorific Vho- cannot be used by itself; it is prefixed to a name. All members of the family address the old man with the honorific Vho- prefixed to Rasila (Vho-Rasila) to show him respect – because he is elderly and he is the head of the family. The name Rasila is the family’s last name (surname). The equivalent of Vho-Rasila is Mr or Mrs Rasila, because Vho- does not denote gender, but refers to people of either sex. In this regard, because we are aware of the gender of the addressee, the equivalent is ‘Mr Rasila’. The element of gender brings about translation problems when translators try to find the equivalent of names prefixed with the honorific Vho-. If the text above did not specifically mention the fact that Mr Rasila was an old man, it would be difficult to provide the English equivalent of the name (because the translator would not know whether Vho-Rasila was supposed to be translated as ‘Mr’ or ‘Mrs’). Both men and women can be addressed as Vho-Rasila (Mr Rasila or Mrs Rasila), because Rasila is the last name. In languages such as English, politeness is shown by the use of ‘Mr’ and ‘Mrs’ prefixed to the last name. 160 MUNZHEDZI JAMES MAFELA In the passage cited above, younger persons are addressed without the use of the honorific Vho-. This becomes clear when Mr Rasila addresses his children, Masindi and Sundani, without prefixing the honorific Vho- to their first names. In Tshivenḓa, the honorific Vho- is not only prefixed to the last name to show respect – it can also be prefixed to the first name. Mashiri (1999) states that the honorific marker can be used together with the first name or the last name to show respect. Nor is it uncommon to hear children addressing or referring to neighbours with their first names prefixed with an honorific marker (Mashiri, 1999: 96). Muhanelwa is Rasila’s wife and she is addressed by her first name, prefixed with the honorific Vho(Vho-Muhanelwa) to show respect. The same applies to Mr Rasila’s children who are, at times, addressed as Vho-Masindi and Vho-Sundani. The equivalents of Vho-Masindi and Vho-Sundani are difficult to find, because English does not use the form of the first name to denote respect. The English equivalents of Vho-Masindi and Vho-Sundani will therefore remain as ‘Masindi’ and ‘Sundani’. However, the fact that the element of respect has been left out in the English equivalents means that an incomplete meaning has been rendered. In short: the translated version fails to bring out the full meaning of the original Tshivenḓa equivalents. The prefixing of the honorific Vho- is not solely restricted to Tshivenḓa names: to show respect, even non-Tshivenḓa names are prefixed with the honorific Vho-. In the text cited above, Maria, Mr Rasila’s daughter-in-law, is addressed as ‘Vho-Maria’. Vho-Maria cannot be translated into English as ‘Mrs Maria’, because Maria is the first name. The solution here will be to write the first names without the prefix Vho-. As I said at the beginning of this article, how to address people correctly is something that is culturally determined. What is considered polite in one language may be perceived as impolite in another (Emma, 2009: 1). In our case, effective translation from the Tshivenda language into English, which involves the honorific Vho- prefix to names, can sometimes prove difficult – as the example shows. 3. The pronoun A pronoun is generally used either to refer back to a noun that has been used earlier in the discourse or one that is implied (Poulos, 1990: 92). Fromkin and Rodman (1998: 192) state that pronouns may be used in the place of noun phrases or may be used to refer to an entity that is, presumably, known to the discourse participants. This means that pronouns may occur anywhere that a noun phrase may occur. Dickey (2002: 5) makes the point that, in a linguistic sense, the concept address is a speaker’s linguistic reference to his or her collocutor(s), which includes not only nouns, but also pronouns. All speech communities have grammatical and lexical differentiation in their forms of address. Besides prefixing the honorific Vho- to the first and last names of persons, Tshivenḓa also uses pronouns. As in the case of the honorific Vho-, the Tshivenḓa pronouns are characterised by translation problems when translated into a language such as English. The pronouns of class 2 are usually used in the plural form, but are also used, in class 2(a), in a singular form to show respect to the elderly. The pronouns that will be discussed in this section are absolute pronouns, relative pronouns and possessive pronouns. The absolute pronoun of class 2(a) in Tshivenḓa is characterised by translation problems because, in class 2, the absolute pronoun denotes the plural whereas, in class 2(a), it denotes the singular (as illustrated below). Vho-Maria vha tshi bva afha nḓuni na vhone vho vho ḓo ita zwa u reshana na avha mueni vho imaho henefha muṱani (Sigogo, 1995: 41). (When Maria came out of the hut, she also greeted the visitor who was standing in the courtyard.) TRANSLATION AND MEANING, PART 10, 2013, 157 - 163 161 The vhone written in bold is an absolute pronoun referring to Vho-Maria. However, in the English translation the equivalent of vhone is she, which is a singular form. However, vhone is an absolute pronoun in the plural form and refers, for example, to vhathu (Vhathu vhone vha khou ḽa - People they are eating). But in this case it refers to the name ‘Maria’, which has been prefixed by the honorific Vho- to show respect. This type of respect is difficult to translate into English, hence the equivalent of vhone is she. In short, there is no English equivalent for vhone which, in Tshivenḓa, is a respectful form of address. English uses she which, as we know, does not reflect the person’s social status. As in the case of the honorific Vho-, vhone can refer to persons of either gender. This brings in other translation problems because, in Tshivenḓa, the same name can be bestowed on both boys and girls. The context will tell the reader whether the character is a man or woman. If the context does not make it clear, then it will be difficult for the translator to provide the English equivalent of the absolute pronoun vhone. In the text cited above, it is easy to identify the character as a female, because Maria is known to be the name given to girls and women. The translation problems encountered in the absolute pronoun also exist in the relative pronoun of class 2(a) in Tshivenḓa. The relative pronoun of class 2(a) is vhane which, again, is a respectful form of address. The extract below is a good illustrative example of this problem. “Naa ndi thitshara? Hu ita hani?” Ndi vhone mukalaha vhane vha vhonala vha tshi nga vho takalela u ḓa ha vhenevha Vho-Shandukani (Sigogo, 1995: 41). (“Is that you teacher? How are you?” It is the old man, who looks as if he is happy to see Mr Shandukani, who has paid them a visit.) Mukalaha (the old man) refers to Mr Rasila, and vhane refers to mukalaha. The relative pronoun vhane therefore refers to Vho-Rasila. The equivalent of vhane is who in English. In Tshivenḓa, vhane also refers to more than one person. However, in the extract above, it refers to one person, and denotes respect. Unlike in the absolute pronoun, the equivalent of vhane, which is who in English, does not lead to problems in the translation, because who, in English, can be either singular or plural. However, in the English translation, the element of respect is missing: we just have the expression ‘the old man’. The English translation, in fact, fails to capture the complete meaning of the original text. The same holds for the translation of the Tshivenḓa possessive pronoun of class 2(a). The Tshivenḓa possessive pronoun of class 2(a) is avho, which is illustrated in the example below. Vho-Masindi, mafhungo avho a a lemela sa ngweḓi vhukuma (Sigogo, 1995: 40). (Masindi, your matter is indeed as heavy as iron.) Note that Vho-Masindi is not translated as ‘Mrs Masindi’ in English, because Masindi is a first name and, as we know, English first names cannot be prefixed by Mr or Mrs to show respect. The equivalent of the possessive pronoun avho is your in English. Whereas avho can be a possessive pronoun in both the plural and singular forms, the English your also denotes both the singular and plural forms. However, once again, the English term – your in this case – lacks the element of respect. The possessive pronoun avho refers to Vho-Masindi. As a result, the English translation of the Tshivenḓa sentence does not reflect the respectful form of address shown to the character Masindi and, once again, the English translation therefore only partially captures the meaning of the Tshivenḓa text. 4. Concord Concord is the word category in a language that is grammatically related to the noun because it shows agreement with the class of the noun. Like the Tshivenḓa pronouns of class 2(a), the Tshivenḓa concords of class 2(a) are characterised by translation problems when translated into langua- 162 MUNZHEDZI JAMES MAFELA ges such as English. The Tshivenḓa concord of class 2(a) shows respect to the addressee (rather than the plural form in class 2). The extract below will illustrate this point. Ndi hone Vho-Shandukani vha tshi kumedza mafhungo kha Vho-Masindi. Vho-Masindi ndi hone vha tshi a swikisa kha muvhuye. Nga murahu ha tshifhinga ndi hone Vho-Maria vha tshi kona u fhindula (Sigogo, 1995: 40). (It is then that he (Mr Shandukani) tells Masindi the matter which she (Masindi), in turn, relates to her sister-in-law. After some time, she (Maria) is able to respond.) The concord vha in the extract refers to three people: the first vha refers to Vho-Shandukani, which is the last name of Aifheli, whereas the second refers to Vho-Masindi and the last to Vho-Maria, both of which are first names. In the translated version of the extract, Vho-Masindi and Vho-Maria are translated without the prefix ‘Mrs’, because first names in English are not accorded the respectful form of address. Vha- is known to be a concord that refers, for example, to more than one person in class 2. For example, in Vhathu vha a tshimbila (People they are walking), the equivalent of vha is they, which represents the plural form. However, in this extract it has been used to refer to an individual in class 2(a) – again, to show respect. In the extract, the equivalent of vha is he/she, which represents the singular form. It is therefore impossible to provide the equivalent respectful expression in English. The English translation version does not, therefore, reflect the complete meaning of the source text because, once again, it lacks the element of respect. Unlike the name ‘Maria’, Shandukani and Masindi can be names of people of either gender. If the translator does not know the character’s gender, he or she will find it difficult to provide the equivaent of the concord vha, which could be either he or she. Translators usually depend on the context in which the concord appears in order to provide the correct equivalent. 5. Conclusion The discussion above shows that translating forms of address from Tshivenḓa into English is by no means an easy task. The problem is made worse by the fact that Vhavenḓa are very sensitive about address usage: forms of address play a large part in social interaction of the Vhavenḓa. Languages of different cultures will always use forms of address in dissimilar ways. In Tshivenḓa, one does not have to know the addressee in order to show respect: all elderly addressees are shown respect through the use of the polite form of address. In Tshivenḓa, the honorific Vho- of class 2(a), and the accompanying concords and pronouns, are used to show respect to the elderly. When these words are translated into English, the element of respect always disappears, which means that the English text will reflect only an incomplete meaning. Also, in Tshivenḓa, the respect shown to the addressee by the use of the form of address is also applied to first names and, again, it is difficult, if not impossible, to translate this satisfactorily into English. Sometimes it is difficult to identify the gender of the addressee because, unlike in English, the honorific Vho- and the accompanying concords and pronouns do not distinguish gender. The same applies to names that are given to both boys and girls. In this case, the translator will have to depend on the context. References Adler M.K. (1978), Naming and Addressing: A Sociolinguistic Study. Hamburg: Helmut Buske Verlag. Akindele, D.F. (2008), “Sesotho Address Forms”. Linguistic Online. http://www.linguistic online.de/ 34_08/akindele.html. Dickey, E. (2002), Latin Forms of Address: From Plautus to Apuleius. Oxford: Oxford University Press. Emma. (2009), “Forms of Address in English”. http://www.lexiophiles.com/English/forms-of-address-inenglish. Fromkin, V. and Rodman, R. (1998), An Introduction to Language. 6th ed. Fort Worth: Harcourt Brace College Publishers. Mashiri, P. (1999), “Terms of Address in Shona: A Sociolinguistic Approach”. Zambezia XXVI(i), pp. 93 – 110. TRANSLATION AND MEANING, PART 10, 2013, 157 - 163 163 Parkinson, D.B. (1985), Constructing the Social Context of Communication: Terms of Address in Egyptian Arabic. Berlin: Mouton de Gruyter. Poulos, G. (1990), A Linguistic Analysis of Venda. Pretoria: Via Afrika. Sigogo, N.E. (1995), Ho felwa nnyi. Pretoria: Kagiso Publishers. Szarkowska, A. (2005), „On Teaching Forms of Address in Translation”. Albaglobal. http:www.alaglobal.com/article 1390.html Ziervogel, D., Wentzel, P.J and Makuya, T.N. (1972), A Handbook of the Venda Language. Pretoria: University of South Africa. Warren-Rothlin, A. (2007), “Politeness Strategies in Biblical Hebrew and West African Languages”. Journal of Translation. http://www.sil.org/siljot/abstract.asp. Zgusta, L. (1971), Manual of Lexicography. Mouton: The Hague, Paris. 164 TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 165 CALQUES IN POLISH TRANSLATIONS OF ENGLISH PRESS ARTICLES: LINGUISTIC INNOVATIONS OR MISTRANSLATIONS? Mirosława Podhajecka University of Opole, Poland Abstract: The present paper focuses on words and expressions (e.g. certyfikat urodzenia) encountered in Polish (TL) translations of English (SL) press articles available from Internet portals and electronic newspapers, which are seen as idiosyncratic from the point of view of codified language use. Under close scrutiny, most of the innovations have turned out to be examples of calquing, so the question that needs to be asked is whether all types of calques should be allowed in translation. The analysis of the samples suggests that calques of SL patterns can be applied fairly successfully when there are no adequate means in TL to express the original concepts, whereas the use of calques violating principles of TL grammar or stylistics should, at least theoretically, be highly restricted. Nonetheless, it is clearly not always the case, as the latter are replicated quite frequently, apparently not having been recognized as ‘alien’ by TL readers. One of the findings of my study is therefore that modern users whose language awareness is increasingly shaped by the mass media may, and in fact often do, uncritically adopt deviant expressions as instances of correct language. This is bound to lead to shifts in stylistic preferences of the TL users. 1. Introduction A significant influence of English onto the Polish language and culture over the last few decades has been a widely acknowledged fact (e.g. Fisiak, 1970; Walczak, 1983; Mańczak-Wohlfeld, 1994a; Chłopicki and Świątek, 2000; Mańczak-Wohlfeld, 2006). It has manifested itself, in the first place, in the massive influx of English words, whose changes in form and meaning are now documented in detail (e.g. Mańczak-Wohlfeld, 1994b, Mańczak-Wohlfeld, 2009). Linguistic influences, however, can also appear in a more camouflaged form, and there is substantial evidence that calques have had as long a tradition in most European languages as borrowings. It should be explained that a calque, also referred to as ‘loan translation’ (e.g. Katamba, 2005), ‘loanshift’ (e.g. Haugen, 1950) or ‘through translation’ (Newmark, 1988), is a lexical item translated part by part (i.e. morpheme for morpheme) on the basis of a foreign model, e.g. Pol. drapacz chmur < Eng. sky-scraper, Eng. Holy Spirit <Lat. Spiritus Sanctus, Pol. listonosz <Ger. Briefträger or Eng. superman <Ger. Übermensch. Items created in this way are morphologically complex, remaining at the same time semantically transparent, because both the constituent elements as well as rules governing their combination are native (Bynon, 1996: 232). Calques are usually subdivided into lexical and semantic ones, of which the former are created via the substitution of morphemes (Eng. mother tongue <Lat. lingua materna), while the latter appear due to a graphic/ phonological and semantic resemblance with a foreign word (US Spanish grados ‘degree’ extended to include the meaning of English grades (Spanish ‘notas’) (Field, 2002: 9). Still, this classification seems to be over-simplified, insofar as the existence of various types of hybrid calques, especially multi-word ones (cf. Silva-Corvalán 1995: 255-265), makes any sharp distinctions between them fairly difficult. Being typical of active bilingualism, calquing has always been part and parcel of the speech of immigrant communities. The following expressions, for instance, represent characteristic features of American-Polish vocabulary: Pol. robić przyjaciół < Eng. to make friends; Pol. ludzie angielskie < Eng. English people; Pol. leniwy pies < Eng. a lazy dog; Pol. pomóż sobie sam <Eng. help yourself; Pol. mieć na ręce < Eng. to have at hand; Pol. szkoła internatowa < Eng. boarding school; Pol. market owocowy < Eng. fruit market; Pol. egzaminacja lekarska <Eng. doctor’s examination, etc. (cf. Sękowska, 1991). The above examples are a clear proof that what we are dealing here with must be contact-induced change, or linguistic interference. Conceived of as a rearrangement of patterns or rules in one language under the influence of another one (Weinreich, 1970: 1, cf. Thomason and Kaufman, 1988: 35; Heine and Kuteeva, 2005: 6-8), interference has traditionally been related to lin- 166 MIROSŁAWA PODHAJECKA guistic transfer, either positive or negative (e.g. Odlin, 1989). Transfer can be found in all areas of language structure: phonology, morphology, syntax, etc.; in fact, when human creativity comes into play, there are virtually no limits to the possibilities of transferring linguistic features from one language into another (cf. Thomason, 2001). By involving mediation between two languages, translations turn out to be classic instances of language contact situations with the same mechanism of linguistic transfer. Interestingly, a certain regularity has been observed here: transfer in translation is assymetrical, because it is the source language (SL) that influences the target language (TL). Named the law of interference of the SL, it is based on the premise that “phenomena pertaining to the make-up of the source text tend to be transferred to the target text” (Toury, 1995: 275), and one of the hypotheses proposed by Teich (2003: 222) is that the SL “shines through” the TL patterns. There are more researchers who have noticed this apparent one-way interaction between languages in translational contexts. According to Koller, for example, translations have always been responsible for introducing some SL properties into German as a TL, triggering two types of language variation: innovations on the level of the language system (Systeminnovationen), and innovations on the level of usage norms or stylistic innovations (Norm-/Stilinnovationen) (qtd. in Baumgarten and Özçetin, 2008: 293-294). As might be expected, usage and/or stylistic preferences are more susceptible to language change than the underlying linguistic system, which is, after all, more stable than unstable a structure. Importantly, what may enhance the tendency to introduce innovations into the TL is the fact that great stress is usually laid in university curricula for future translators on improving proficiency in the foreign language (i.e. the translator’s ST), while excellent command in the mother tongue (i.e. the translator’s TT) is often wrongly taken for granted (Rommel qtd. in Hervey and Higgins, 1992:1). To ask a rhetorical question, if one has not mastered his/her own native tongue, how can we expect him/ her to translate successfully? 2. Calques in translation studies The present paper looks at the use of calques in translation, which has already received attention in the literature, primarily practice-oriented (e.g. Vinay and Darbelnet, 1995; Silva-Corvalán, 1995; Sewell, 2001). In it, calques have been discussed from different research angles, such as equivalence (e.g. Sievers, 2006), stylistics (Vinay and Darbelnet, 1995) or cultural issues (Hervey and Higgins, 1992: 28). Being originally a linguistic concept, in translation studies calques are usually associated with translation procedures proposed by Vinay and Darbelnet (1995:31), which, together with borrowing and literal translation, belong to the direct methods of translation. Vinay and Darbelnet point to two distinct classes of calques: (1) lexical ones, which respect the syntactic structure of TL, while introducing a new mode of expression; and (2) structural calques, which introduce new constructions into TL. Providing the translator with effective means of expressing SL concepts, both types are recognized as convenient in the translation process. It is worthy of mention that calques as constituent elements of the TL are not always identified with a unit of translation, at least in the process-oriented approach, but they fit Malmkjaer’s (1998: 286) definition “the target text unit that can be mapped onto a source-text unit”, which is applied in product-oriented translation studies. Notwithstanding Vinay and Darbelnet’s claim that new calques can serve to fill a lacuna (1995: 32), not all calques obey the law of the TL standardization. More exactly, while items created in accordance with the TL system are practically invisible to the reader (e.g. Eng. professional translator → Pol. profesjonalny tłumacz or Eng. Polish presidency → polska prezydencja), more controversial are those which violate codified practices of the TL. Translations of some English commercials and advertisements show that the latter have already permeated into Polish: “na widok Corsy Twoje serce zabije szybciej” (SL your heart) instead of “na widok Corsy serce zabije Ci szybciej”; “meble dla TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 167 domu/kuchni/ogrodu (SL for home) instead of “meble do domu /kuchni /ogrodu”; “wanny masażowe (SL massage baths)” instead of “wanny do masażu”; or “namioty imprezowe (SL party tents)” instead of “duże namioty, które można wykorzystać na imprezach plenerowych” (Chłopicki & Świątek, 2000: 275-278; Tabakowska, 2002: 31). Such grammatical and lexical innovations are bound to be applied by other TL users, the more so because the very fact of appearing in the media grants the texts public recognition, which translates directly into high acceptability rate. No wonder the innovations mentioned above, having been introduced merely a decade ago, are apparently treated as standard Polish phrases (e.g. the phrase namioty imprezowe is found on 3,290 websites, and meble dla domu occurs as many as 888,000 times). There are reasons to think that the mass media, including the World Wide Web, have now become authoritative sources of language use in Poland. Since translations are part and parcel of daily news reports, the translational language, with all its inevitable inadequacies, must also be treated as such. The users, while not being completely insensitive to the problem of language correctness, often do not have the time, the skills or the resources to evaluate lexical, stylistic or grammatical features of the TTs, so the translator appears to be the primary agent in the importation of innovative patterns into the TL. Whether or not this hypothesis can be tested empirically – there are some qualitative and quantitative methodological constraints, like difficulties in ascertaining the first occurrence of a given pattern in the TL, that may affect the reliability of the results – translations seem to increasingly affect both spoken and written domains of the Polish language. As studies of the language of the news usually ignore the role of translation (Baker and Saldanha, 1998: 186), this issue undoubtedly deserves more in-depth research. 3. Rationale for the study and the research framework My interest in calques was triggered by poor quality of a Polish translation of English press articles that I once came across on the Web, and which might have been expected of a novice in the field, not a professional translator. With this experience in hand, I started collecting translations that looked idiosyncratic due to the use of awkward TL equivalents. In the present paper I provide an analysis of such items, all of which are apparently calques of the SL models, attempting to determine whether they should be interpreted as linguistic innovations or utter mistranslations. In doing so, I have relied primarily on my native speaker’s intuition and experience; I hope that having been involved in teaching translation to students for several years now allows me to evaluate translation products in an unbiased way. However, to support my decisions with empirical data, I conducted a study on the recognition of the calques using corpus methodology. More exactly, in line with the recent developments in corpus research – despite its chaotic and uncontrollable content, the Web is claimed to be an important source of linguistic evidence – I treat the World Wide Web as a corpus (cf. Kilgarriff and Grefenstette, 2003). To retrieve calques from the Web, I applied the most popular Google search engine (www.google.pl); although Google frequency data does not always provide a correct usage index (Kilgarriff, 2007), it nonetheless seems relevant for a general assessment of linguistic trends. Explaining my research framework in as simple terms as possible, the number of occurrences of a given pattern is its frequency. For the context-derived calque raportował zyski, for example, raportować zyski becomes the canonical form, but because of the rich inflectional system of Polish, various retrieved variants of the given pattern, sometimes broader than one would expect, are also recorded (e.g. raportuje zyski, raportowała zyski, raportowali zyski, etc.). The overall frequency of a calqued pattern, i.e. the sum of the frequencies of the particular variants, is regarded as rudimentary for determining the genuine currency of the item (cf. section 4). Since it is hard to estimate the relative frequency of conventional Polish structures on the Web, let alone deviant ones, due to the lack of an official yardstick I introduced my own categorization: the frequency range 0-500 is perceived as low, that of 501-1000 as medium, and that over 1001 – as high (cf. Podhajecka, 2006: 270). 168 MIROSŁAWA PODHAJECKA My project is based on the analysis of 18 English press articles (STs) and their translated Polish versions (TTs). The English articles are drawn primarily from the electronic version of Financial Times, a prestigeous British daily, as well as the international editions of CNN (www.cnn.com), BBC (www.bbc.co.uk) and FoxNews (www.foxnews.com). The Polish translations, some of which are no longer accessible on the Web (in such cases their URLs are not specified in the list of references), have been collected mainly from the Internet portal www.onet.pl and the electronic editions of Gazeta Wyborcza at www.gazeta.pl, arguably two best known sources of information in Poland, and another source of the translations is the Internet portal of a Polish broadcasting station TVN24 (www.tvn24.pl). It needs to be mentioned that the process of collecting texts was not as smooth as might be expected; Polish portals and e-newspapers do not always provide information on the ST, so I often had to search for the original texts myself. It can be fairly difficult, because the same news gets reported by various portals and newspapers, often in slightly altered forms; there is admittedly a lot of give-andtake among news agencies. Another difficulty in identifying STs was that some of the TTs should be treated as adaptations rather than translations proper (e.g. selected passages are skipped, reported speech is used instead of direct speech, etc.), which Hervey and Higgins (1992: 17) call ‘gist translation’. All in all, there may be some inconsistency in the matching of SL and TL texts, yet the dependence of the Polish expressions on the English models is undeniable. As calques escape consistent classifications, in this paper I use a fairly broad definition of the term. In other words, I take into account patterns in which at least one element is a calque, either a lexical (Eng. birth certificate – Pol. certyfikat urodzenia) or a semantic one (e.g. Eng. company – Pol. Kompania). I also treat as calques items in which only a specific linguistic feature, such as aspect in verbs or word order in a phrase, has been transplanted into the TL (e.g. Eng. two advantages – Pol. dwie przewagi). Speaking of syntactic constructions, it is easy to see that they are not always identical replicas of the SL patterns in the strict sense of the word (e.g. Eng. a contraction in credit – Pol. kurczenie się kredytu), but since they mirror the foreign model to a greater or lesser degree, I regard them as legitimate objects of my analysis. 4. Analysis of calques in Polish translations 4.1 Calques violating the category of number Jednak moje biurko ma dwie przewagi nad Yet my desk has two advantages over Lord biurkiem lorda Lucana.TT10 Lucan’s. ST10 While English advantage is a countable noun, its Polish equivalent przewaga, as singularia tantum, can only appear in the singular form. The translator had two simple solutions: either to look for another equivalent in the plural form (e.g. zalety, plusy) or apply the noun przewaga preceded by an adjective (e.g. moje biurko ma pewną przewagę / sporą przewagę / dużą przewagę). As changes allowed by the linguistic system are not always reflected by usage, it may be surprising that the Google search for dwie przewagi yields 1310 hits (cf. Podhajecka, 2007). W szczególności, po upadku Lehmanów [sic], delewarowanie i wymuszone sprzedaże tak się nasiliły, że doszło do całkowitego załamania się tradycyjnych stosunków cenowych i rozchwiania modeli prowadzenia interesów. TT7 Most notably, in the wake of the Lehman collapse, the pattern of deleveraging and forced sales has been so intense that traditional price relationships have completely broken down, sending trading models haywire. ST7 What should grasp the TL reader’s attention is that the English noun sales should not be rendered into Polish in the plural form, because sprzedaż conventionally occurs in the singular only (cf. USJP). TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 169 Therefore, wymuszona sprzedaż seems a fairly resonable option, and another alternative, przymusowa sprzedaż, seems even better, fitting the context more closely (“przymusowa sprzedaż aktywów, akcji, środków trwałych”, etc.). Illustrating deviation from linguistic norms, the phrase wymuszone sprzedaże has nevertheless a relatively high frequency in the TL (117 hits, compared with as many as 6680 hits for przymusowa sprzedaż). 4.2 Calques violating TL stylistics – semantic extensions Bank uznał, że po obcięciu stóp procentowych w okolice zera wykorzystał całą dostępną amunicję i musi podjąć dodatkowe, niekonwencjonalne działania, aby zapobiec osuwaniu się gospodarki w deflację. TT1 The Bank conceded that it had run out of ammunition on interest rates after cutting rates as close to zero as possible and needed to take additional unorthodox measures to prevent a slide into deflation. ST1 In this context, the SL ammunition may be treated as a military metaphor, since it has been used figuratively rather than literally. Still, equivalent forms such as wszelkie dostępne środki or wszelkie argumenty sound more natural, which means that the metaphorical extension, by being incomprehensible in a text related to banking, is superfluous in the TL. Yemenia wydala komunikat wyjaśniający przyczyny zawieszenia lotów z Francji na Wyspy Komorskie. Decyzja została podjęta ze względu na "bardzo poważne wydarzenia, jakie zaszły w ostatnich dniach" oraz "zagrożenie, na jakie pewne osoby narażają pracowników lotnisk, naszej kompanii oraz pasażerów". TT11 "In light of serious incidents in recent days and major risks that some passengers posed to airport staff, our company and passengers, Yemenia has decided to no longer serve Moroni for an indefinite period from July 3 until the situation eases," it said in a statement. ST11 According to USJP, kompania ‘a trade association … widespread in Western Europe in the 17th and 18th centuries’ [translation mine] is a historical term with no references to the modern times. If this semantic calque is an attempt to stylize the TL text, such lexical stylization is fairly inconsistent, as no other elements are comparable to it in terms of stylistic value. Interestingly, contrary to to the lexicographers’ assessment, there are more instances of kompania on the Web, especially in contexts referring to airlines (“szwajcarska kompania lotnicza Swiss otrzymała oficjalną propozycję”, “australijska kompania lotnicza Qantas zamierza”). Uwolnieni przez wojsko chłopcy byli miesiącami trenowani przez terrorystów, by przeprowadzić samobójcze ataki. - Wyprano im mózgi. Wyprano im mózgi przeciwko nam wszystkim – mówią zajmujący się chłopcami psychologowie. TT18 They had all been kidnapped by the Taliban and taken to camps where they would be trained to kill; trained to be suicide bombers. […] “They have been brainwashed. Brainwashed against people like you and me,” she said. ST18 Though the TT is not identical with the ST, the area of overlap is evident at first sight. The wordform trenowany, seen as the Polish equivalent of trained, is absolutely unacceptable, at least to me as a translator trainer; trenowany brings to mind positive sporting connotations, which are by no means associated with military trainings, usually given to terrorists by ex-army officers. As a result, the only reasonable equivalent that can be taken into account here, and one ignored by the translator, is szkolony. A search for trenowani przez terrorystów has yielded limited results (14 hits, mistakenly calculated by Google as 9700!), but some variations of this expression are used more frequently (cf. section 4). 170 MIROSŁAWA PODHAJECKA 4.3 Calques violating TL stylistics – syntactic constructions Podczas gdy Demokraci w Kongresie chcą, aby obie strony zredukowały swe arsenały do 1.000 głowic, administracja Obamy nie wytyczyła jeszcze celu dla rozmów. TT4 While congresssional Democrats are pushing for both sides to cut their arsenals to 1,000 warheads, the Obama administration has not yet agreed on a goal for the talks. ST4 It is difficult to predict what the kernel of this phrase may have been, be it agree, goal or talks, yet the ultimate result is clearly unsatisfactory. One could accept the collocation wytyczyć cel, even though, in this particular context, there are more suitable verbs (określić or wyznaczyć), but the main problem is the use of the preposition; it is celu rozmów that sounds natural in Polish. Google hits (384000) for a wild card search (celu dla *) are largely irrelevant, as they cover plenty of syntactic constructions that conform to the TL norms (e.g. “w tym celu dla każdego osiedla przygotowała osobną ulotkę” / “została sformułowana w innym celu dla potrzeb innej ustawy”). W miarę jak rosną sumy odpisywane na straty, finansiści niemal każdego pokroju wahają się przed wchodzeniem w długoterminowe transakcje. TT7 As fresh write-offs keep mounting, financiers of almost any stripe are nervous about committing to long-term transactions. ST7 What has been disregarded in this otherwise adequate TL term is that, in this particular context, transakcje długoterminowe would be more adequate. It is not the only case in which the standard pattern of English noun phrases (a modifier in the attributive position followed by a noun) is taken into Polish, whereby there is a great deal of variability in the use of pre- and postmodifiers, depending on the stylistic properties of the text. The internal arrangements of words in phrases is rarely the subject of serious translation research, because the dividing line between acceptable and unacceptable forms is somewhat tenuous, being dependent on style rather than on grammar. W środę, w londyńskiej siedzibie firmy aukcyjnej Bonhams, licytowano biurko lorda Lucana. Minęło 35 lat odkąd lord Lucan zabił opiekunkę dzieci swojej rodziny, myśląc że to jego żona. TT10 On Wednesday, at Bonhams’ salerooms in London, Lord Lucan’s desk will be put up for auction. It is 35 years since Lord Lucan killed his family’s nanny, thinking she was his wife. ST10 This translation is clearly defective: the translator rendered SL nanny as TL opiekunka, but s/he decided to enhance the translation with the Polish equivalent of his family, which has resulted in idiosyncratic wording. The easiest way to translate this passage would be by saying zabił opiekunkę swoich dzieci; the SL context does not specify whether all the children in the family were actually lord Lucan’s, but comprehensibility in the TL translation has a price to pay. Interestingly, as can be seen in the TT above, even punctuation is subject to calquing – in Polish, for instance, time adverbials are never separated from the clause with a comma, which has to precede the conjunction że. Po dniu spędzonym głównie na pozyskiwaniu sobie Europejczyków w Brukseli, Clinton kładła nacisk na pragnienie administracji Obamy pracowania wspólnie z partnerami na całym kontynencie. TT4 After a day largely spent wooing Europeans in Brussels, Mrs Clinton emphasised the Obama administration's desire to work with partners throughout the continent. ST4 There are several factors that contribute to the awkwardness of this TL expression. Firstly, even if one treats the collocation kłaść (położyć) nacisk as a proper equivalent of the verb emphasize, the SL word-form suggests both repetitive and non-repetitive actions, so the translator should have noticed that położyła nacisk would fit the context better. Moreover, this grammatical construction is made of components that do not go with one another: nacisk na pragnienie is unnatural in Polish, and so is TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 171 pragnienie administracji [Obamy] or pragnienie [administracji Obamy] pracowania. To convey the meaning rather than the form, the translator should have opted to render this passage as follows: Clinton podkreśliła, iż administracja [prezydenta/ Baracka] Obamy pragnie współpracować z partnerami na całym kontynencie. Jak można upierać się przy kosztownym programie rozszerzenia opieki zdrowotnej albo zachęt podatkowych dla najuboższych pracowników, kiedy deficyt budżetowy już przebijał barierę biliona dolarów? TT5 How could he press ahead with his costly schemes to widen healthcare and offer better incentives for the working poor when the budget deficit was already bursting through the $1,000bn barrier? ST5 The TL syntax requires either the imperfective aspect of the present tense form (przebija) or past tense in the perfective aspect (przebił), even though – this is an important proviso – English past progressive tense is not, as a rule, translated into Polish in the latter way (cf. Kaczmarski 1983). Apart from that, the collocation przekroczyć próg appears to be more natural in the TL (and English billion should obviously be translated as miliard). Despite the fact that the Web data is hard to analyse because of a range of potential forms and different contextual uses, one can consider several better alternatives, such as przekroczyć barierę (2210 hits) or przekroczyć próg (13300 hits), of which the last one is clearly most common in Polish. Natomiast Greenwell żaliła się, że musi sprawować opiekę nad trojgiem dzieci, których biologiczni rodzice starają się o separację, lecz ich wychowywanie jest dla niej uciążliwe ze względu na pracę jako kierowca tira. TT13 Greenwell then told the Romeros that she was taking care of three children whose biological parents were going through a separation […] saying that her job as a truck driver made it hard to take care of the children… ST13 My intuition tells me that the TL grammatical construction in the last clause is unaccceptable; there are a few solutions which would not require much effort on the part of the translator, such as ich wychowywanie jest dla niej uciążliwe ze względu na fakt, iż pracuje jako kierowca tira / ich wychowywanie jest dla niej uciążliwe, ponieważ jest kierowcą tira. One should not have any objections to the naturalness of the text chunks uciążliwe ze względu na (1760 hits) or ze względu na pracę (7570000), but the pattern ze względu na pracę jako kierowca has only been found in 6 websites, all of which include the same translation shown above. Testament jest datowany na 7 lipca 2002 r. i The will, dated July 7, 2002, estimated his majątek Jacksona jest w nim oszacowany na estate at that time at more than $500 więcej niż 500 mln dolarów. TT14 million. ST14 Więcej niż, a calque of two possible English patterns – more than and in excess of – should be replaced by Polish ponad, which reflects better the TL linguistic habits. The phrase więcej niż has perfectly correct uses in Polish („dzieci, które zarabiają więcej niż wielu dorosłych” / „więcej niż przepisy kulinarne” / „ropy wycieka więcej niż przypuszczano”), so the translator needs to be aware of semantic nuances that affect the TL usage. In my experience, it is a frequent mistake made by translator trainees. The wildcard search for “więcej niż * dolarów”, many of which are apparently irrelevant, yields over 14 million hits. 172 MIROSŁAWA PODHAJECKA 4.4 Calques violating collocatability Głęboka recesja skłoniła 315-letnią instytucję do decyzji o nabywaniu aktywów, w większości obligacji państwowych, za pieniądze tworzone przy pomocy komputerowego guzika. TT1 The deep recession has led the 315-yearold institution to embrace the policy of quantitative easing – the purchase of assets, mostly gilts, with more money created at the stroke of a central bank computer key. ST1 Had the translator kept the reference to the central bank, komputerowy guzik would be more comprehensible, though still not recommended, as an equivalent of the SL computer key. The Polish version przez jedno naciśnięcie klawisza (przycisku) komputera seems more appropriate; klawisz, which immediately connotes the computer keyboard, is suggested by all bilingual dictionaries available to me. Komputerowy guzik, with various inflectional endings, is rather infrequent on the Web. Można to częściowo przypisywać kurczeniu się In part, this can be blamed on a kredytu, gdy banki zaczynaja się dobierać do contraction in credit as banks get tough swoich wierzycieli. TT7 with their creditors. ST7 In this context, it is not any specific bank credit that is contracting, so we should explicate the expression under analysis by translating it accordingly, for example, as kurczenie się możliwości kredytowych or kurczenie się rynku kredytów. To my surprise, the phrase kurczenie się kredytu has nonetheless been found in the Web materials (92 hits). Firma bada szczególnie straty wygenerowane przez Alexisa Stenforsa, pracującego w londynie [sic!] tradera walutowego, który, jak pisze New York Times, wcześniej raportował zyski z transakcji za rok 2008 rzędu 120 mln dol. TT6 The New York Times reported the investigation yesterday and said that Mr Stenfors had earlier reported a personal trading profit of $120m for 2008. ST6 Since the Polish expression raportować zyski belongs to the jargon of financiers and brokers –unlike the English counterpart to report profits used both in specialist as well as general contexts – the question the translator needs to answer is whether it should be resorted to in a journalistic text targeted at non-specialist audience. In my opinion, collocations such as odnotować zyski or osiągnąć zyski are more adequate here, even though their meanings overlap only to a certain extent (i.e. acquired profits may not necessarily be reported by dishonest businesspersons). As to the Web data, raportować zyski has 10 hits, odnotować zyski – 126 hits, while osiągać zyski – 17700 hits, so the standardized form is self-evident. Z kolei władze Hawajów potwierdziły, że Hawaiian officials have confirmed they have posiadają w swoich aktach certyfikat urodzenia a birth certificate on file for Obama… amerykańskiego prezydenta. TT15 ST15 Certyfikat has made a brilliant career in the Polish language over the last two decades, hence certyfikat energetyczny, certyfikat księgowy, certyfikat językowy, certyfikat podpisu elektronicznego, certyfikat inwestycyjny, certyfikat ISO and many others. In most cases, however, this term designates new types of documents, whereas SL birth certificate has had its traditional Polish equivalent for years. The inability to come up with one of two legitimate terms, świadectwo urodzenia or slightly old-fashioned metryka urodzenia, shows that the translator has not created an adequate target culture situation, having resorted to maximizing the sameness of form only. The frequency of certyfikat urodzenia (6800 hits) on the Web is impressive (cf. akt urodzenia – 56200 hits), which is indicative of problems faced by Polish immigrants in English-speaking countries (e.g. “czy mając angielski certyfikat urodzenia dziecka będę mogła wyrobić polski paszport?”, “child benefit – do tego bedzie Ci potrzebny certyfikat urodzenia”, “revenue odsylalo nam certyfikat urodzenia naszego bobasa”). TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 Także obligacje korporacyjne rosły najmocniej od ponad dekady. Rynek ma nadzieję, że działania banku centralnego odblokują rynek prywatny, zupełnie przyblokowany przez kryzys kredytowy. TT1 173 Sterling corporate bonds also rose by the most in a decade, as the Bank’s move raised hopes that its decision to buy gilts would help unstick private sector markets gummed up by the credit squeeze. ST1 This expression concerns Polish financial terminology, which has been shaped vigorously under the influence of English. Interestingly, in the TL one cannot say obligacje rosły; the act of growing cannot be perceived literally, so the expression needs to be changed into ceny akcji korporacyjnych rosły. While the phrase obligacje rosły is infrequent (10 hits), obligacje rosną occurs more often on the Web (423 hits). Setki ataków miało miejsce od czasów wyborów prezydenckich, o wiele więcej niż zazwyczaj powiedział Mark Potok z pozarządowej organizacji Intelligence Project w Southern Poverty Law, która zajmuje się monitorowaniem zbrodni nienawiści. TT19 There have been “hundreds” of incidents since the election, many more than usual, said Mark Potok, director of the Intelligence Project at the Southern Poverty Law Center, which monitors hate crimes. ST19 This TL passage comes from gist translation, which includes a handful of weird TL expressions (e.g. incydenty nietolerancji rasowej), so it is easy to guess that zbrodnie nienawiści is likewise a SL calque. Hate crime is a concept in American federal law, introduced in 1968 after the assasination of Martin Luther King, which nowadays covers crimes related to the victim’s race, color, religion, national origin, and gender identity. As to the Polish version, there are a few hate-based terms with adequate Polish equivalents (Eng. hate campaign – Pol. nagonka; Eng. hate mail – Pol. listy z pogróżkami), which is a clear sign that one does not need to express nienawiść overtly in the TL. Przestępstwo związane z prześladowaniem lub dyskryminacją fills the communicative gap fairly well, but one has to admit that the term is longer and less transparent than its English model. Perhaps that is why zbrodnie nienawiści has high frequency in the Web resources (zbrodnia nienawiści – 3110 hits). 4.5 Calques violating idiomaticity Po swym spotkaniu z Ławrowem w Genewie, które określiła mianem ‘dobrego początku z perspektywy USA’ stwierdziła, że ‘nie ma czasu do stracenia. Niezwłocznie zaczniemy pracować nad przekładaniem naszych słów na czyny’. TT4 “There is no time to waste. We will begin working immediately to translate our words into deeds,” she said after her meeting with Mr Lavrov in Geneva, describing it as “a good beginning from the US perspective”. ST4 One could come up with more adequate TL collocations to choose from, such as przekuć słowo w czyn, obrócić słowo w czyn, zamienić słowo w czyn or wprowadzić słowa w czyn. Since przekładanie słów w czyny looks fairly idiosyncratic to me, the last TL sentence should have been replaced by zaczniemy niezwłocznie wprowadzać nasze słowa w czyny. The search for przełożyć słowa na czyny yields 218 hits, yet wprowadzić słowa w czyn is five times as frequent. Mówiono by o sukcesie lub klęsce w zależności od tego, czy poradziłby się z posprzątaniem cudzego bałaganu. Ale Obama nie jest tego rodzaju politykiem. TT5 Success or failure would be defined on whether he had managed to clear up some else’s mess. Mr Obama is not that sort of politician. ST5 In this case, we are faced with a SL idiomatic expression that remains comprehensible to TL readership even though it has lost its idiomatic character – posprzątanie cudzego bałaganu connotes, literally, an act of cleaning. Perhaps the translator should have considered a Polish idiom that would 174 MIROSŁAWA PODHAJECKA better fit the context alluding to George W. Bush’s policy, naprawić to, co sknocił ktoś inny, the more so because s/he would not need to compensate for any loss in meaning. The search for the shortened form cudzy bałagan yields 66 hits. Kiedy stałem w laboratorium, zajmując się kośćmi rodzonej siostry królowej, i myślałem, że ona dotykała kiedyś Kleopatry, Cezara czy Marka Antoniusza, jeżyły mi się włosy na szyi wyznaje inny członek ekipy, Neil Oliver. TT16 “When I stood in the lab and handled the bones of Cleopatra's blood sister - knowing that in her lifetime she touched Cleopatra and perhaps Julius Caesar and Mark Antony as well - I felt the hairs go up on the back of my neck.” ST16 It is argued that the knowledge of SL and TL idiom is a prerequisite for a good translator, so mistranslations of this sort should be avoided at any price. The SL idiomatic expression to feel one’s hairs go up on the back of one’s neck can be translated in several ways, all of which express the same communicative situation of being frightened: włosy się jeżą na głowie / włosy się jeżą na karku / ciarki przechodzą po plecach / czuć ciarki na plecach / włosy stają dęba. One of these, włosy się jeżą na głowie, is a modified version of the original pattern włosy się jeżą na karku, which shows that susceptibility to change is an inherent feature of many phraseological units. What may be of interest is that Pol. szyja is synonymous with kark, but the two content words are not interchangeable in this idiom, despite the fact that the concept of “włosy na szyi” is, quite literally, becoming increasingly common (“mialam włosy na szyi, podbrodku”, “mam włosy na szyi z tyłu głowy”, “mam dość mocno widoczne i długie włosy na szyi”). Uważają bowiem, że 58-letni radiowiec jest personifikacją brzydkiego, stereotypowego Republikanina: małomiasteczkowego wyrzutka z college'u, wściekłego białego mężczyzny. TT17 “They think that Limbaugh, 58, is the very personification of an ugly Republican stereotype: he’s a small-town college dropout, an angry white man… ST17 Apart from the phrase wściekły biały mężczyzna, which is a mistranslation that conjures up a completely different image in the TL than angry white man in the SL (wściekły ‘furious’ has no references to the background phrase angry young man ‘młody gniewny’), what is an unfortunate calque is the TL wyrzutek z college’u. It is an alien expression in Polish because wyrzutek has connotations and uses (e.g. wyrzutek społeczeństwa) that are non-existent in the case of a college drop-out ‘a student who has dropped out of college’. As to the way in which this mistranslation could be improved to convey the original meaning, one might suggest the following alternatives: facet z małej mieściny wyrzucony niegdyś ze studiów / facet z małej mieściny, który wyleciał kiedyś ze studiów. Wyrzutek z college’u is found in 10 websites, all of which refer to the same film review. 5. Discussion of the research results In my study, I have shown 22 calques encountered in Polish translations of English press articles, which deviate from the TL norms in one way or another by violating the stylistics, collocatability, idiomaticity or various grammatical aspects of the TL. Because a handful of them are not fully applicable for the Web searches, for further analysis I took into account 14 calques retrieved, in the required meanings, from the Web materials. Based on my intuition, I would say that a predominant majority of the calques have turned out to be mistranslations in the TL, which should be rejected both by translators and the TL users. By contrast, only a few of them, like zbrodnia nienawiści, are apparently associated with a new communicative value, hence their introduction into the TL is at least comprehensible. It may therefore be interesting to see what the collected evidence, i.e. fully empirical data, can tell us about the status of these alien formations in Polish. TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 No Calque and frequency Variants and frequency 1 dwie przewagi 1310 2 wymuszone sprzedaże 117 3 kompania lotnicza 597 4 trenowani prez terrorystów 14 5 pragnienie * pracowania 0 6 przebić barierę 2190 7 komputerowy guzik 19 8 kurczenie się kredytu 102 9 raportować zyski 10 10 certyfikat urodzenia 6800 11 zbrodnia nienawiści 3110 12 przełożyć słowa na czyny 218 13 cudzy bałagan 66 14 jeżyły * się włosy na szyi 5 dwóch przewag 219 dwoma przewagami 3 dwóch przewagach 33 wymuszonych sprzedaży 74 wymuszonym sprzedażom 1 wymuszonymi sprzedażami 3 kompanii lotniczych 1590 kompanii lotniczej 3790 trenuje się terrorystów 244 terroryści trenują 25 terroryści trenowali 39 pragnienie pracowania 28 pragnienie pracowania 3 pragnienia pracowania 1 przebijać barierę 34 przebił barierę 7000 przebiła barierę 678 przebije barierę 116 komputerowego guzika 2 komputerowe guziki 95 kurczenia się kredytu 48 kurczeniem się kredytu 2 raportował zyski 4 raportowała zyski 12 certyfikatu urodzenia 637 certyfikatem urodzenia 159 certyfikacie urodzenia 103 zbrodnie nienawiści 482 zbrodni nienawiści 2290 zbrodniom nienawiści 450 zbrodniach nienawiści 302 przekładać słowa na czyny 5 przekładają słowa na czyny 1 przełożył słowa na czyny 213 przełoży slowa na czyny 9 przełożą słowa na czyny 1 przełożysz słowa na czyny 1 przełożyli słowa na czyny 1 cudzego bałaganu 94 cudzym bałaganem 7 zjeżyły się włosy na szyi 3 175 Overall frequency 1675 Usage index high 194 low 5977 high 322 low 32 low 7928 high 116 low 152 low 26 low 7699 high 6634 high 449 low/medium 77 low 8 low Out of the 14 items, five have been retrieved with frequency significant enough to evaluate the currency of the calques as high: dwie przewagi, kompania lotnicza, przebić barierę, certyfikat urodzenia and zbrodnia nienawiści. The pattern przełożyć słowa na czyny is the only one that can potentially be seen as falling into in the medium-frequency range, while eight others have been retrieved with very low to low frequencies (wymuszone sprzedaże, trenowani przez terrorystów, pragnienie pracowania, komputerowy guzik, kurczenie się kredytu, raportować zyski, cudzy bałagan and jeżyły * się włosy na szyi), hence their usage index is relatively insignificant. How should we assess these results in the light of corpus research? Firstly, the most frequent and recognizable patterns have a good chance to remain in use, and their currency is a clear proof that the TL norms have shifted. These are apparently examples of linguistic innovations. Secondly, a number 176 MIROSŁAWA PODHAJECKA of novelties introduced into the TL apparently function in it, but their range of use is limited. One can say that medium-frequency calques can behave in a dual way, that is, in conducive circumstance they can be adopted by a larger number of users and their frequency will consequently rise, but in unfavourable conditions they will probably go out of use. The status of such calques, in my study represented by one instance only, is therefore difficult to predict. Thirdly, those SL-based expressions that occur infrequently on the World Wide Web must be regarded as more or less creative renditions of the SL elements into the TL, which are inevitably of ephemeral status in the TL. It is easy to see that they must be treated as mistranslations. Interestingly, while I would disapprove of most of the calques shown above, the TL users’ preferences are clearly less prescriptive than mine. This might suggest that the TL users take conscious choices to use means that violate traditionally perceived language norms, which does not seem to apply here due to limited social interest in linguistic matters. Therefore, it is more probable that the modern users, whose language awareness is increasingly shaped by the mass media, unconsciously adopt atypical expressions as instances of correct language, which, in the long run, is bound to lead to shifts in the TL usage. 6. Conclusions As results from the present pilot study, calques as a standard translation procedure and one of the popular direct methods of translation should be approached with caution: if used uncritically, they will introduce into the TL awkward patterns of the SL models, which may then be reproduced, equally uncritically, by other TL users. Intuitively, I would expect most calques which gain some currency in the TL to conform to the principles of Polish grammar and stylistics. Still, this is clearly not always the case, inasmuch as expressions which are fairly alien from the point of view of linguistic norms come into use in spite of their atypicality and the universal warning against “unintentional calquing resulting from too slavish a simulation of the grammatical structures of the ST” (Hervey and Higgins 1992: 34). Thus, one implication of my research is that translators are invariably exposed to the danger of calquing as a form of negative transfer, while the TL users are faced with the danger of being affected by it. Low quality of the analysed translations inevitably poses the question of the translators’ competence, since “in this vast world of communication and information overload, we need competent translators who have both the theoretical knowledge and practical skills to do their jobs well” (Razmjou, 2004). Why so many weird calques have found their way into translations published by well known Internet portals and e-magazines is hard to explain, but at least some are indicative of the lack of the knowledge and skills referred to above. Importantly, while in most cases “translation is a highly creative activity in which the translator’s personal responsibility is constantly to the forefront” (Hervey and Higgins, 1992: 2), the responsibility for the quality of translations published in the media is apparently taken by the editors, which seems to contribute to the diluting of responsibility and greater acceptance of imperfect translations. As for the question asked in the title of my paper, there is clearly no ready-made answer, which depends on the adopted point of view. If we take into account the TL norms as the authoritative parameter, most of the calques analysed here, by violating these norms, will be prescriptively treated as mistranslations. By contrast, if we lay emphasis on language usage, the most important criterion in descriptive studies, then those calques which are frequently replicated by the TL users will enjoy the status of stylistic innovations. The borderline between these two approaches points to the ubiquitous tension between descriptivism and prescriptivism in translation studies. TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 177 References Primary sources ST1 Bank pumps £75bn into economy, Chris Giles, published 5 March 2009. At: http://www.ft.com/cms/s/0/2240b7ce-09ce-11de-add8-0000779fd2ac.html. TT1 Bank Anglii wpompuje w gospodarkę 75 mld funtów, 6 marzec 2009. At: http://ft.onet.pl/11,22617,bank_anglii_wpompuje_w_gospodarke_75_mld_funtow,artykul.html ST2 Protection by another name, Christopher Caldwell, published 6 March 2009.At: http://www.ft.com/cms/s/0/7c0df13c-0a81-11de-95ed-0000779fd2ac.html TT2 Ochrona, ale pod inną nazwą. At: http://ft.onet.pl ST3 EU calls for crisis meeting on GM, Nikki Tait, published 6 March 2009. At: http://www.ft.com/cms/s/0/c4eabe50-09ee-11de-add8-0000779fd2ac.html TT3 Unia zapowiada rozmowy na temat kryzysu GM. At: http://ft.onet.pl ST4 US and Russia vow to conclude arms control deal, Frances Williams, Tony Barber,and Daniel Dombey, published 7 March 2009, http://www.ft.com/cms/s/0/d548a7de-0ab7-11de-95ed-0000779fd2ac.html TT4 Rosja i USA zapowiadają nowe porozumienie rozbrojeniowe. At: http://ft.onet.pl ST5 The change he believes in: Obama shrugs off storm, Philip Stephens, published 5 March 2009. At: http://www.ft.com/cms/s/0/4b3dd940-09ba-11de-add8-0000779fd2ac.html TT5 Zmiana, w którą uwierzył: Obama prowadzi przez zawieruchę. At: http://ft.onet.pl ST6 Merrill Lynch currency trader suspended, Adrian Cox, Peter Garnham, and Greg Farrell, published 7 March 2009. At: http://www.ft.com/cms/s/0/ffd12eee-0ab8-11de-95ed-0000779fd2ac.html TT6 Merrill Lynch odkrywa “nieprawidłowość”. At: http://ft.onet.pl ST7 Financial world is stumbling blindly through a cognitive fog, Gillian Tett, published 6 March 2009. At: http://www.ft.com/cms/s/0/da26af7c-09ee-11de-add8-0000779fd2ac.html TT7 Świat finansów potyka się i błądzi. At: http://ft.onet.pl ST8 Terror and greed combine to restore old colonial order, Simon Kuper, published 6 March 2009. At: http://www.ft.com/cms/s/2/900592a6-0a88-11de-95ed-0000779fd2ac.html TT8 Czy sport wraca do epoki kolonialnej? At: http://ft.onet.pl ST9 Developing nations outperform as equity markets drop, David Oakley, published 12 March 2009. At: http://www.ft.com/cms/s/0/a35c55f0-0e6c-11de-b099-0000779fd2ac.html TT9 Siła wschodzących rynków akcji. At: http://ft.onet.pl ST10 Lord Lucan and the missing charm of modern desks, Lucy Kellaway, published 9 March 2009. At: http://www.ft.com/cms/s/0/2efd9ff2-0c4b-11de-b87d-0000779fd2ac.html TT10 Lord Lucan i znikający urok biurka. At: http://ft.onet.pl ST11Yemenia suspends all flights to Comoros, published 4 July 2009. At: http://www.shebacss.com/en/media-center-25295.html TT11 Samoloty Yemenii na Komory nie polecą, protesty odniosły skutek, published 5 July 2009. At: http://wiadomosci.gazeta.pl/Wiadomosci/1,80708,6789260,Samoloty_Yemenii_na_Komory_nie_poleca__ protesty_odniosly.html ST12 Election spurs 'hundreds' of race threats, crimes, published 16 November 2008. At: http://www.suntimes.com/news/elections/1281749,election-racial-obama111608.article TT12 Obama wygrywa. Plaga rasizmu w Stanach, published 17 November 2008. At: http://www.tvn24.pl/12691,1573345,0,6,obama-wygrywa-plaga-rasizmu-w-stanach,wiadomosc.html ST13 Woman answers ad for bird by offering kids as payment, 27 February 2009. At: http://edition.cnn.com/2009/CRIME/02/27/kids.for.bird/index.html TT13 USA: zamieniła dwójkę dzieci na papugę i 175 dolarów, published 7 March 2009, At: http://wiadomosci.gazeta.pl/Wiadomosci/1,80708,6354974,USA__zamienila_dwojke_dzieci_na_papuge_i _175_dolarow.html ST14 Michael Jackson's Will Filed in Court, Excludes Ex-Wife Debbie Rowe, published 1 July 2009. At: http://www.foxnews.com/entertainment/2009/07/01/michael-jacksons-filed-court-excludes-ex-wifedebbie-rowe/ TT14 Testament Jacksona złożony w sądzie w Los Angeles, published 1 July 2009. At: http://wiadomosci.gazeta.pl/Wiadomosci/1,80708,6779998,Testament_Jacksona_zlozony_w_sadzie_w_Lo s_Angeles.html ST15 Born in the USA? Justice, Supremes confirm getting eligibility challenge, Bob Unruh, published 24 March 2009. At: http://www.wnd.com/index.php?fa=PAGE.view&pageId=92810 178 MIROSŁAWA PODHAJECKA TT15 Prezydentura Baracka Obamy wisi na włosku? Published 25 March 2009. At: http://www.tvn24.pl/12691,1592602,0,1,prezydentura-baracka-obamy-wisi-na-wlosku,wiadomosc.html ST16 Cleopatra's mother ‘was African’, published 16 March 2009. At: http://news.bbc.co.uk/2/hi/also_in_the_news/7945333.stm TT16 Kleopatra - Mulatką? Published 25 March 2009. At: http://turystyka.interia.pl/ciekawostki/news/kleopatra-mulatka,1276034,238 ST17 Comment: Anti-Obama retains radio power, Jonathan Mann, published 7 March 2009. At: http://edition.cnn.com/2009/POLITICS/03/06/anti.obama/index.html?iref=allsearch TT17 Nie jest czarny, ani szczupły - oto anty-Obama, published 7 March 2009. At: http://wiadomosci.onet.pl/regionalne/warszawa/nie-jest-czarny-ani-szczuply-oto-antyobama,1,3498976,wiadomosc.html ST18 Kidnapped boys ‘brainwashed’ to die as suicide bombers, Stan Grant, published 4 August 2009. At: http://edition.cnn.com/2009/WORLD/asiapcf/08/03/pakistan.boys/index.html TT18 “Wyprano im mózgi – przeciwko nam wszystkim”, published 5 August 2009. At: http://wiadomosci.onet.pl/cnn/wyprano-im-mozgi-przeciwko-nam-wszystkim,3522010,7352457,artykulfotoreportaz-duzy.html Secondary sources Baker, M. and G. Saldanha (eds.) (1998), Routledge encyclopedia of translation studies. London-New York: Routledge. Baumgarten, N. and D. Özçetin (2008), “Linguistic variation through language contact in translation”. In: Peter Siemund and Noemi Kintana (eds.), Language contact and contact languages. Amsterdam-Philadelphia: John Benjamins. 293-316. Berman, A. (2000), “Translation and the trial of the foreign.” In: Lawrence Venuti (ed.), The translation studies reader. London-New York: Routledge. 284-297. Bynon, T. (1996), Historical linguistics. Cambridge: Cambridge University Press. Chłopicki, W. and J. Świątek (2000), Angielski w polskiej reklamie. Warszawa: Wydawnictwo Naukowe PWN. Field, F. (2002). Linguistic borrowing in bilingual contexts. Amsterdam-Philadelphia: John Benjamins. Fisiak, J. (1970). “The semantics of English loanwords in Polish”. Studia Anglica Posnaniensia, 2. 41-50. Haugen, E. (1950). “The analysis of linguistic borrowing”. Language, 26(2): 210-231. Heine, B. and T. Kuteeva (2005), Language contact and grammatical change. Cambridge: Cambridge University Press. Hervey, S. and I. Higgins (1992), Thinking translation: A course in translation method, French-English. London-New York: Routledge. Kaczmarski, S. (1983), A glossary of Polish and English verb forms. Warszawa: Państwowe Wydawnictwo Naukowe. Katamba, F. (2005), English words: Structure, history, usage. London-New York: Routledge. Kilgarriff, A. (2007), “Googleology is bad science”. Computational Linguistics, 33(1): 147-151. At: http://www.kilgarriff.co.uk/Publications/2007-K-CL-Googleology.pdf Kilgarriff, A. and G. Grefenstette (2003), “Introduction to the special issue of Web as corpus.” Computational linguistics, 29(3): 1-15. Malmkjaer, K. (ed.) (1998), Translation and language teaching: Language teaching and translation. Manchester: St. Jerome. Mańczak-Wohlfeld, E. (1994a), Tendencje rozwojowe współczesnych zapożyczeń angielskich w języku polskim. Kraków: Universitas. Mańczak-Wohlfeld, E. (1994b), Angielskie elementy leksykalne w języku polskim. Kraków: Universitas. Mańczak-Wohlfeld, E. (2006), Angielsko-polskie kontakty językowe. Kraków: Wydawnictwo Uniwersytetu Jagiellońskiego. Mańczak-Wohlfeld, E. (ed.) (2009), Słownik zapożyczeń angielskich w polszczyźnie. Warszawa: Wydawnictwo Naukowe PWN. Newmark, P. (1988), A Textbook of Translation. New York: Prentice Hall. Odlin, T. (1989), Language transfer: Cross-linguistic influence in language learning. Cambridge: Cambridge University Press. Podhajecka, M. (2006), “Using the Web in the translation process.” Poznań Studies in Contemporary Linguistics, 41. 257-274. TRANSLATION AND MEANING, PART 10, 2013, 165 - 179 179 Podhajecka, M. (2007), “Czy rzeczownik ryzyko należy wciąż do singularia tantum?” Poradnik Językowy, 2. 54-63. Razmjou, L. (2004), To be a good translator. http://accurapid.com/journal/28edu.htm (DOA 12 October 2010). Sewell, P. (2001), “The occurence of calque in translation scripts”. Meta, 46. 607-615. Sękowska, E. (ed.) (1991), Słownictwo anglopolskie (zakres i zasady opracowania leksykograficznego). Warszawa: Instytut Języka Polskiego UW, Pracowania Językoznawstwa Stosowanego. Sievers, M. (2006), No translation is perfect: Gain and loss of linguistic and cultural features. Munich: Grin. Silva-Corvalán, C. (1995), “Lexico-syntactic modeling across the bilingual continuum.” In: Jacek Fisiak (ed.), Linguistic change under contact conditions. The Hague: Mouton de Gruyter. 253-270. Tabakowska, E. (2002), “Bariery kulturowe sa zbudowane z gramatyki”. In: Roman Lewicki (ed.), Przekład – Język – Kultura. Lublin: Wydawnictwo Uniwersytetu Marie Curie-Skłodowskiej. 25-34. Teich, E. (2003), Cross-linguistic variation in system and text. Berlin: Mouton de Gruyter. Thomason, S. (2001), Language Contact. An Introduction. Edinburgh: Edinburgh University Press. Thomason, S. and T. Kaufman (1988), Language contact, creolization, and genetic linguistics. Berkeley-Los Angeles: University of California Press. Toury, G. (1995), Descriptive translation studies and beyond. Amsterdam-Philadelphia: John Benjamins. USJP = Uniwersalny słownik języka polskiego (2003), (ed.) Stanisław Dubisz. Warszawa: Wydawnictwo Naukowe PWN. Walczak, B. (1983), “The earliest borrowings from English into Polish”. Studia Anglica Posnaniensia, 16. 121-133. Weinreich, U. (1970), Languages in Contact. Findings and Problems. The Hague: Mouton. Vinay, J. and J. Darbelnet (1995), Comparative stylistics of French and English: A methodology for translation (trans. and ed.), Juan C. Sager and M.-J. Hamel. Amsterdam-Philadelphia: John Benjamins. 180 TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 181 GENOLOGIE CONTRASTIVE AU SERVICE DE LA TRADUCTION DES TEXTES DE SPECIALITE Barbara Walkiewicz Adam Mickiewicz University, Poland Abstract: Every text belongs to a genre. Every genre in turn belongs to some sort of discourse. Every specialist field within its type of discourse has developed a range of genres based on a form which best suits its function in a given communication situation. These genres, related by family resemblance, form a discursive polysystem characteristic for a given culture. Specialist translations need to be written into the framework of the target discursive system, which requires the knowledge of its genre cartography. This paper aims to present the role of linguistic genology in specialist translations. 1. Traduction des textes de spécialité La traduction des textes de spécialité est une opération requérant de la part du traducteur un savoir faire à plusieurs niveaux, dans la mesure où il est censé aboutir à une équivalence non seulement au niveau sémantique et référentiel, mais aussi à celui pragmatique, relatif aux normes discursives régissant la communication propre au domaine de spécialité donné (Walkiewicz, 2009c, sous presse). Chaque texte est déterminé par des règles propres au genre dont il relève, qui sont codifiées socioculturellement dans le cadre d’un type de discours donné (Grzmil-Tylutki, 2007: 50). Ces règles normalisent les réalisations du génotype dans deux dimensions: celle de la structure verticale (superstructure) et celle des structures horizontales (microstructure). La normalisation est la plus stricte dans les textes de spécialité qui se caractérisent par la prédominance de la fonction informative assurant une haute précision référentielle. Celle-ci est possible grâce à une rigidité rédactionnelle fondée sur la relation optimale de la forme à la fonction, qui statue sur l’indépendance de l’interprétation des textes de spécialité du contexte de leur production, ce qui n’est pas sans incidence sur la traduction. En effet, la codification générique des textes de spécialité implique la connaissance des conventions rédactionnelles assumant une triple fonction: de « marqueurs de différenciation », de « déclencheurs d’attentes déterminées chez le récepteur » et de « balises sur le parcours hermeneutique » (Reiss, 2009: 134). Sans les connaître, le traducteur risque « d’appliquer indûment des conventions spécifiques à des spécimens de textes plus génériques ou de plaquer abusivement des conventions génériques sur des textes particuliers » (ibidem:132). Car le texte de spécialité ne peut être interprété que par rapport au genre dont il relève, mais considéré dans le cadre du type de discours donné où il coexiste avec les autres genres propres au même domaine de spécialité. Il s’ensuit que les genres et les texte qui les actualisent ne sont jamais isolés mais se définissent par rapport aux genres avoisinants. A cela s’ajoute la migration de certains genres qui peuvent desservir plusieurs domaines de l’activité humaine, p. ex. la lettre (migration horizontale) ou même faire partie de genres plus complexes (migration verticale ou transgénérique). Dans la plupart des cas, les textes de spécialité sont des unités textuelles complexes, comportant des genres plus simples, ceux-ci pouvant se décomposer en genres simples ou premiers (au sens bakhtinien du terme), décomposables en stratégies ou conventions rédactionnellement codifiées (Walkiewicz, 2010b, sous presse). Le texte de spécialité est donc une structure fondée sur des éléments préconstruits de taille variable qui peuvent faire partie aussi d’autres textes relevant du même type de discours mais d’un autre genre, tissant ainsi un réseau de ressemblance familiale1 qu’illustre le schéma suivant: 1 Terme proposé par Wittgenstein (Wittgenstein 1953 / 2000). 182 BARBARA WAŁKIEWICZ Fig. (1): Interférences intergénériques ou phénomène de ressemblance familiale L’interférence des genres desservant différents types de discours ainsi que les relations de ressemblance familiale qui les relient en font un polysystème discursif à part. Traduire un texte de spécialité revient à confronter le génotype qu’il réalise avec les normes génériques propres au discours de spécialité analogue dans la culture réceptrice. Toutefois, il est rare que les genres s’équivalent d’une langue à l’autre. Il peut arriver qu’au niveau structurel les genres se ressemblent tout en renvoyant à deux relations socio-communicationnelles différentes. C’est pourquoi le parcours traductif devrait passer par l’identification du genre au niveau du discours compris comme l’ensemble des relations communicationnelles, relatives à un domaine de spécialité déterminé, comme le montre le schéma cidessous: Fig. (2): Parcours traductif (Walkiewicz 2009c, sous presse) Si le discours détermine communicationnellement un domaine de spécialité qui, en tant que tel, reste extra-verbal, on peut admettre que dans chaque culture à un degré d’évolution comparable il existe TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 183 un discours qui s’y rapporte, bien qu’il soit différemment configuré au niveau socio-linguistique, ce qui se manifeste par une autre constellation des genres qui le desservent (Walkiewicz, 2009c, sous presse). Ceci est d’une importance capitale pour la traduction car déplace le niveau d’équivalence du générique (du dire) au discursif (au faire), ce qui veut dire que même si dans le discours d’arrivée il n’y a pas de genre équivalent ou les genres ne se recouvrent pas, on peut arriver à traduire le texte de départ tout en respectant les règles discursives propres à la culture d’arrivée au niveau de toutes les instances normalisatrices: NIVEAUX D’ANALYSE INSTANCE DE NORMATIVITÉ TEXTE PROJET TEXTUEL STYLE IDIOLECTE GENRE NORMES DE GENRE CHAMP GÉNÉRIQUE2 NORMES DE CHAMP GÉNÉRIQUE DISCOURS NORMES DE DISCOURS LANGUE RÈGLES LANGUE SOURCE / D’ARRIVÉE Fig. (3) Niveaux d’analyse du texte vs instances de normalisation selon Rastier (Rastier, 2007) C’est possible grâce au phénomène de ressemblance familiale fondé sur les interférences intergénériques, dans la mesure où chaque genre complexe est décomposable en genres simples qui à leur tour sont décomposables en stratégies de base. Nous avons donc affaire à une sorte de double articulation discursive (des stratégies et genres simples dans les genres complexes et des genres dans les discours) qui garantit une équivalence discursive même en cas d’absence d’équivalence générique. En effet, si un genre complexe n’a pas d’équivalent direct en LA, on peut trouver les équivalents de ses composants (indépendamment de leurs taille et nature) qui peuvent être distribués dans différents genres plus ou moins apparentés discursivement (Walkiewicz, 2010b, sous presse). Le traducteur peut y puiser des conventions rédactionnelles tout en respectant l’équivalence discursive qui se traduit par l’analogie de la relation forme / fonction (R = F / f). De la sorte, on peut restituer la superstructure de départ en reverbalisant les segments de la microstructure conformément aux normes discursives propres à la langue et à la culture d’arrivée, qui permettront de déclencher le mécanisme d’interprétation correspondant à l’intention communicationnelle de l’auteur du TD. Pour y arriver, le traducteur est censé connaître la configuration des genres dans le polysystème cible, descriptible en termes de « cartographie » générique dont l’élaboration est l’un des postulats de la génologie linguistique. 2. Génologie linguistique La génologie linguistique c’est la sous-discipline de la linguistique qui s’occupe de l’analyse et de la description des genres de textes. Partant du principe qu’un genre de texte constitue une norme de comportement verbal admettant des matérialisations linguistiques qui l’actualisent différemment, les chercheurs en postulent la description en termes de champ générique3 à géométrie variable en fonction « des gradients de typicalité » (Adam, 1999: 93). Celui-ci est compris comme l’espace dis2 Selon Rastier « un champ générique est un groupe de genres qui contrastent voir rivalisent dans un champ pratique: par exemple, au sein du discours littéraire, à l’époque classique, le champ générique du théâtre se divisait en comédie et tragédie » (Rastier, 2007). 3 Par « champ générique » nous comprenons, après Wojtak, toutes les variantes d’un genre donné. 184 BARBARA WAŁKIEWICZ cursif qui s’articule dans le cadre des trois variantes génériques: canonique, alternative et adaptative (Wojtak, 2009:355-356). La variante canonique renvoie au modèle générique qui comporte les éléments statuant sur l’identité générique, qui sont présents dans toutes les réalisations d’un genre donné. Il y a des genres n’existant qu’en tant que variante canonique – il s’agit surtout des genres à statut hautement normatif, tels genres juridiques, judiciaires et certains genres administratifs (p.ex. textes de loi, arrêtés, cartes d’identité, passeports) qui se caractérisent par la plus haute rigidité rédactionnelle. La variante alternative concerne les genres qui sont plus flexibles structurellement, admettant des modifications quantitatives (suppression ou ajout d’un ségment supplémentaire par rapport à la variante canonique) ou qualitatives (changement de l’ordre d’agencement des segments faisant partie de la superstructure du texte), par exemple deux contrats de location peuvent différer par le nombre et l’ordre des paragraphes qui les constituent. Ce qu’il faut souligner c’est que la variante alternative n’admet que des modifications dans le cadre d’un même genre. En revanche, la variante adaptative se rapporte aux genres à faible degré de petrification, sujets au phénomène de transgression générique (migration transgénérique ou « chaînes génériques » de Beacco (2004:117)) qui consiste en assimilation de genres simples (au sens bakhtinien du terme) ou de certains de ses éléments empruntés à d’autres genres. Les modèles adaptatifs sont assez fréquents dans le discours des médias où il y a des genres dont l’hétérogénéité structurelle résultant d’une labilité discursive en constitue un des traits distinctifs majeurs, par exemple la publicité (Wojtak, 2009:356). Etant donné que le genre de texte est une norme de comportement verbal plus qu’une norme linguistique en tant que telle, l’analyse générique est censée se fonder sur l’observation de réalisations d’un génotype à quatre niveaux: structurel, pragmatique, cognitif et stylistique, en vue d’y observer des écarts quantitatifs et qualitatifs par rapport au modèle canonique. La description ainsi obtenue devrait permettre de connaître la structure du champ d’un genre donné, y compris les périphéries où sa variante adaptative constitue une zone transitoire avec d’autres genres (Wojtak, 2009:357), comme le montre le schéma ci-dessous: Fig. (4): La structure du champ générique Une telle description, élargie sur tous les genres desservant un même type de discours, aboutirait à une cartographie générique démontrant les interférences des genres qui sont tous impliqués dans le réseau de ressemblance familiale. Comme telle, elle a l’avantage de présenter les genres dans leur conditionnement réciproque ce qui, contrairement aux textes parallèles, tirés ponctuellement de leur contexte discursif, minimise le risque d’identification discursive erronée. Comparant les cartographies gé- TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 185 nériques relatives au même type de discours en langues source et cible, on pourrait voir comment s’y distribuent les genres, dans quelles variantes génériques et comment s’enchaîne le réseau de ressemblance familiale à travers d’autres discours. 3. Cartographie générique vs. traduction Les polysystèmes discursifs n’étant pas isomorphes, il peut arriver que dans les cultures mises en contact par la traduction les genres fonctionnellement analogues soient tout à fait similaires, partiellement similaires ou complètement différents (Walkiewicz, 2010b, sous presse). Il n’est pas rare que le texte traduit relève d’un paradigme générique inexistant dans la culture réceptrice dotée d’une autre distribution générique. Mais il est rare que dans la culture cible à un degré d’évolution comparable il n’y ait aucun élément de microstructure qu’on ne puisse repérer. Car les genres ne sont pas des monolythes mais se décomposent en genres simples desservant aussi d’autres génotypes, tissant ainsi un réseau d’interdépendances intergénériques. Pour se rendre capable d’en appercevoir les différences dans le cadre d’un discours de spécialité donné, le traducteur doit connaître les pratiques discursives qui se matérialisent à travers les genres. D’où l’importance de la connaissance préalable du fonctionnement discursif et des paradigmes génériques qui en relèvent dans le cadre d’un domaine de spécialité donné dans deux cultures concernées. L’élaboration de cartographies génériques relatives au même discours dans le cadre de deux communautés discursives permettrait aux traducteurs de considérer les cultures source et cible en termes de polysystèmes discursifs non isomorphes, où les pratiques langagières concernant le même domaine extra-linguistique revêtent des formes génériques différentes. Cela paraît nécessaire surtout dans la traduction des textes qui n’ont pas d’équivalent isomorphe ou qui n’ont aucun équivalent en langue cible. Avec une telle description contrastive, moyennant une identification discursive correcte, le traducteur est en mesure de réussir à trouver des unités de ressemblance familiale interlinguale pour s’en servir dans la reconstruction de la microstructure du TD conformément aux normes discursives en LA, comme le montre le schéma (fig. 5). 186 BARBARA WAŁKIEWICZ Fig. (5):. Ressemblance familiale vs traduction Sans connaître les normes discursives propres au polysystème cible, le traducteur encourt le risque de ne pas assumer l’un des principaux critères de réussite qu’est « l’acceptabilité dans la culture cible » (Lavault-Olléon, 2007:55). La méconnaissance en menace de fautes graves, ce que nous allons démontrer sur l’exemple de traductions vers le français d’un projet architectural polonais, effectuées par des traducteurs professionnels. 4. Etude de cas Aussi bien en France qu’en Pologne, pour pouvoir réaliser un bâtiment, il faut déposer un dossier de demande de permis de construire au service d’architecture et d’urbanisme auprès de la commune du terrain d’implantation, en vue d’obtenir un permis de construire. Dans les deux cultures on a affaire avec le même schéma opérationnel: celui qui veut se faire construire une maison doit avoir un permis de construire. Pour l’obtenir, il doit s’adresser à un architecte pour lui demander d’élaborer un projet conforme aux règlements en vigueur. Aussi la réalité architectonique est tout à fait comparable : la construction d’une maison nécessite des projets relatifs non seulement à la construction mais aussi à toutes les installations : électrique, sanitaires, d’eau, de chauffage, de ventilation, etc. Mais au niveau discursif les genres relatifs au même discours se distribuent autrement, ce qui se reflète dans les différences structurelles paradigmatiques (champ générique) et syntagmatiques (réalisation linguistique des traits génériques). La connaissance des premières est indispensable pour bien situer le texte traduit par rapport aux système des genres desservant le type de discours analogue en culture cible, et celle des secondes permet de restituer les éléments de la microstructure du TD conformément aux normes génériques adéquates, sans quoi il est impossible d’aboutir à une traduction discursivement acceptable. TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 187 5. Paradigmatiquement Dans les deux cultures en question le projet qui fait partie du dossier de permis de construire relève de la documentation architecturale qui comporte toutes les étapes d’élaboration du projet d’un bâtiment, à partir des premières esquisses de conception jusqu’au dossier des ouvrages exécutés. A chaque étape les projets concernent la même construction mais se rapportent à des relations et à des partenaires de communication différents, se qui se répercute sur la forme du projet ainsi que sur le degré de précision des plans (dessins). Malgré la logique commune qui sous-tend l’exécution du projet dans toutes les phases d’élaboration, les genres constituant la documentation architecturale ne sont pas tout à fait équivalents, comme le montre le schéma (fig.6). Fig. (6):. Documentation architecturale en Pologne et en France Ce non-isomorphisme paradigmatique, résultant d’une distribution asymétrique de genres dans le cadre d’une même famille de textes (documentation architecturale), se double de divergences structurelles du dossier de demande de permis de construire. Dans les deux pays le document en question est bipartite, comportant un projet architectural qui en constitue le noyau central et des documents supplémentaires qui ne font pas partie du projet. En Pologne il s’agit du projekt budowlany – projet qui en dehors de la demande de permis de construire ne trouve pas d’application, alors qu’en France il est question d’un extrait de l’avant-projet définitif. Au niveau discursif ils représentent des relations communicationnelles tout à fait équivalentes: dans l’un et l’autre cas, les partenaires du discours sont l’architecte, le client et l’organe d’administration publique responsable de l’architecture et de l’urbanisme. Les deux projets – projekt budowlany et avant-projet définitif sont élaborés dans le but de montrer au client « les plans, coupes et façades, de déterminer les surfaces détaillées de tous les éléments du programme, de définir les principes constructifs, les matériaux et les installations techniques, ainsi que d'établir l'estimation définitive du coût prévisionnel des travaux» (Décret n° 93-1268 du 29 novembre 1993, http://www. legifrance.gouv.fr/texteconsolide/UBHCR.htm). Les deux font partie du dossier de permis de construire annexé à la demande de permis de construire. La seule différence réside donc en ce que l’avant-projet définitif n’est pas annexé à la demande de permis de construire dans son intégralité, ce qui reflète les divergences réglementaires au niveau de l’administration publique. En effet, en France le permis de construire est conçu comme « l’acte administratif par lequel les pouvoirs publics vérifient que la construction projetée, ou la modification d’un édifice existant, respecte bien les règles qui ont été créées dans l’intérêt de tous » (Fèvre & Chailloux, 2008:7). Il en résulte 188 BARBARA WAŁKIEWICZ que les services instructeurs s’y occupent de la vérification de la conformité du bâtiment projeté avec les règles de l’urbanisme concernant le terrain d’implantation. D’où le dossier de permis de construire ne comporte que les éléments de l’avant-projet définitif qui montrent le volume et la forme du bâtiment ainsi que son implantation dans le terrain : plan de masse, plan en coupe du terrain et de la construction, plans des façades et des toitures, document graphique permettant d’apprécier l’insertion du projet de construction dans son environnement, photographies visualisant le terrain dans l’environnement proche et dans le paysage lointan, etc. (Fèvre & Chailloux, 2008:107). En revanche, en Pologne, l’organe d’administration compétent vérifie en outre la conformité de la construction projetée avec la loi de construction et les principes techniques en vigueur, ce qui explique la présence dans le projekt budowlany des parties que la procédure française ne requiert pas (tels détails de construction, spécification de la menuiserie, projets techniques relatifs aux installations nécessaires, etc.). C’est pourquoi, ayant constaté l’analogie de l’identité discursive ainsi que du rapport de la forme à la fonction qui s’y fonde (F / f), le traducteur aurait dû traduire « projekt budowlany » par « avant-projet définitif ». Cependant, dans les traductions analysées, nous n’avons trouvé que deux solutions mises en place: « dossier de permis de construire » et « projet de construction », la dernière étant sûrement puisée dans l’un de deux dictionnaires franco-polonais spécialisés4 disponibles sur le marché polonais. Or, les deux témoignent d’une méconnaissance de la cartographie générique du discours relatif à la construction, ce qui s’est soldé par une localisation éronnée du texte traduit dans le polysystème discursif cible. En effet, le dossier de demande de permis de construire revnoie à l’annexe à la demande de permis de construire dont l’avant-projet définitif n’est qu’une partie. En revanche, le terme « projet de construction », étant équivalent du terme polonais « zamierzenie budowlane », se réfère à « l’ensemble des données d’une construction à prendre en compte avant qu’elle ne soitmatérialisée » (http://www.urcaueidf.archi.fr/abcdaire/fiche.php? fiche=363&chapitre= definition), c’est-à-dire: sélection du terrain constructible, étude de faisabilité, documentation architecturale avec toutes les étapes d’élaboration du projet (documentation architecturale), réalisation de la construction. L’inadéquation des traductions proposées par rapport au découpage générique cible illustre le schéma (fig. 7). Fig.( 7): Inadéquation discursive des traductions proposées. 4 Il est à noter que le seul distionnaire franco-polonais relatif à la construction, accessible sur le marché polonais, ne recense aucun terme correspondant à la documentation architecturale (il existe le terme général « avant-projet » qui, sans les épithètes « sommaire » ou « définitif » ne correspond pas à l’acception architecturale qui nous intéresse). TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 189 6. Syntagmatiquement La cartographie générique du discours relatif à l’architecture et à la construction permet non seulement de vérifier l’identité discursive du genre de départ par rapport au polysystème discursif cible, mais aussi de retrouver des éléments de microstructure autrement distribués. Cela est d’une utilité incontestable surtout dans le cas où les genres fonctionnellement analogues ne se ressemblent pas structurellement, comme c’est le cas du projekt budowlany et de son homologue français – avant-projet définitif. L’avant-projet définitif polonais est un hypergenre recouvrant des textes relevant de différents genres complexes composés de genres simples, eux-mêmes décomposables en unités communicationnelles encore plus simples. L’hyperstructure s’en présente comme suit: Fig. (8): Structure du projekt budowlany La structure du genre en question est dynamique ce qui veut dire que le contenu de chaque réalisation en varie en fonction du type de construction et de la localisation (Walkiewicz, 2009a: 169-170; 2010b, sous presse). A la dynamicité de la structure s’ajoute la dynamicité des genres constitutifs due à leurs variations génériques (variantes canonique, alternative et adaptative). Par contre, le projet faisant partie du dossier de demande de permis de construire en France est décidément plus modeste et comporte les éléments suivants (fig.9) 190 BARBARA WAŁKIEWICZ Fig.(9): Structure de l’avant-projet définitif. On voit donc clairement que la disproportion structurelle et discursive est considérable. Ce qui ne veut pas dire que les genres faisant partie de l’avant-projet définitif polonais n’existent pas en France. En France il y a aussi des réseaux de distribution d’électricité, de gaz, de chauffage, d’eau et d’assainissement auxquels il faut raccorder les bâtiments conformément aux normes techniques, mais les documents y relatifs ne sont pas prévus dans le scénario du même genre qu’en Pologne. C’est pourquoi l’analyse comparative d’un seul genre ne suffit pas – seule la cartographie générique du discours relatif à l’architecture et à la construction pourraot montrer la distribution en genres complexes de genres simples, fondés sur les actions communes aux cultures en question. Sans la compétence discursive en la matière, le traducteur n’est pas en mesure de restituer en français un texte hautement spécialisé comme l’avant-projet définitif, conformément aux normes discursives propres à la culture réceptrice, surtout dans le cas où celle-ci ne dispose pas de genre équivalent, comme le montrent les versions françaises du document « zestawienie stolarki », faisant partie du projet soumis à la traduction. Le document en question relève du genre apparenté formellement aux autres genres graphiques: plan de masse, plans des niveaux, coupes verticales, dessins de façades et dessins de détails (Calvat, 2009:21) et présente la spécification des menuiseries extérieures et intérieures (fenêtres et portes) prévues dans la construction qui constitue l’objet du projet (fig. 10). Fig. (10): Le document « Zestawienie stolarki » soumis à la traduction. En France, l’avant-projet définitif ne contient pas de genre analogue, mais le dossier de constitution des entreprises (DCE), parmi les pièces écrites qui le constituent, comporte un chapitre du Cahier des TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 191 Clauses Techniques Particulières (CCTP) consacré aux menuiseries, qui détaille les types de fenêtres et de portes pris en compte, où le traducteur peut puiser la terminologie et des expressions adéquats, permettant de restituer la microstructure de façon discursivement correcte. Aussi les genres graphiques fonctionnant dans la documentation architecturale française sont utiles dans la traduction des éléments communs à tous les dessins, notamment du tableau précisant les informations sur le dessin, le projet, l’échelle, la maîtrise d’oeuvre ou la maîtrise d’ouvrage. Il y a donc des genres relevant du même type de discours, offrant des unités textuelles de taille variable, correspondant à la microstructure du TD. Il s’agit alors d’une ressemblance familiale interlinguale indirecte, fondée sur l’analogie des genres simples configurés dans le discours de spécialité cible autrement que dans le discours analogue en Pologne, comme l’illustre le schéma (fig. 11). Fig. (11): Distribution des éléments de la mictrostructure du TD dans le polysystème cible De cette manière on aurait pu rendre les séquences « zestawienie stolarki okiennej », « okno stałe + P », « inwestor » et « numer uprawnień » conformément aux normes discursives propres au projet architectural en France, en les traduisant comme « menuiserie extérieure : fenêtres », « fenêtre à deux vantaux dont le vantail droit oscillo-battant », « maître d’ouvrage » et « numéro d’habilitation à la maîtrise d’oeuvre en son nom propre ». Cependant, les auteurs des traductions analysées ont rendu la séquence « zestawienie stolarki okiennej » comme « juxtaposition de la menuiserie » et « juxtaposition des fenêtres », transgressant ainsi la condition d’analogie discursive forme / fonction (F/f). Effectivement, la première solution réfère à l’aspect concret des menuiseries, renvoyant à leur montage5, alors que la dernière embraie la traduction sur le discours informatique. Non moins maladroites discursivement sont les traductions proposées pour la séquence « okno stałe + P » : « fenêtre stable + P », « fenêtre fixe + P » et « fenêtre fixe à ventail ouvrant ». Dans les deux premiers cas, l’élément « + P » (‘avec le ventail droit mobile’), transféré intact en français, perd tout son sens, constribuant à la dénaturation du message, tout comme le dernier qui véhicule une information contradictoire (‘fenêtre fixe et mobile en même temps’). 5 « La mise en oeuvre des éléments s'est faite par juxtaposition des menuiseries sur la structure porteuse en acier » http://www.smabtp.fr/WebFondation/WEB_P_SERVICES_FONDATION.NSF/SPI/ 910892E63 D468605 C1256F08004B4A87?OpenDocument, 15.12.2010). 192 BARBARA WAŁKIEWICZ Une traduction fondée uniquement sur la norme linguistique, sans connaître les normes discursives propres aux genres relevant du type de discours correspondant, a abouti à des erreurs de traduction aussi dans le cas des autres termes susmentionnés : « inwestor » et « numer uprawnień », rendus respectivement comme « investisseur » et « numéro de licence ». Or, le terme d’« investisseur » peut se référer aussi à une personne ayant investi dans une construction, sans en être le maître d’ouvrage. Par conséquent, la traduction ne reflète pas le vrai statut de l’un des partenaires discursifs du projet en question. S’agissant de l’expression « numéro de licence » – version française de la séquence « numer uprawnień », elle renvoie au numéro du diplôme de fin d’études supérieures du premier cycle qui n’est pas suffisant pour pouvoir exercer la profession d’architecte de façon autonome. Les règlements français en vigueur ne requièrent pas ce type d’information dans les projets architecturaux, mais il y a d’autres genres relatifs à la procédure administrative que les diplômés d’état en architecture sont tenus de passer pour se voir octroyer l’habilitation à exercer la maîtrise d’oeuvre en son nom propre où on peut trouver l’expression équivalente. De la sorte, le traducteur peut retrouver des stratégies rédactionnelles adéquates à la microstructure de départ, en puisant dans des textes discursivement apparentés par des relations de ressemblance familiale. On peut ainsi transférer toutes les informations contenues dans le texte de départ sans rien changer dans la superstructure de l’original, en les reverbalisant conformément aux normes discursives propres à la langue et à la culture d’arrivée, ce qui permettra de déclencher le mécanisme d’interprétation correspondant à l’intention communicationnelle de l’auteur du TD (Walkiewicz, 2009b:222-224; 2010, sous presse). 7. En guise de conclusion Les textes de spécialité sont strictement codifiés discursivement ce qui se manifeste dans des formes rédactionnelles hautement conventionnalisées, conçues en vue de réduire le parcours interprétatif au minimum et de rentabiliser l’efficacité informationnelle des textes. La traduction d’un texte de spécialité consiste à trouver la formule algorythmique qui serve de code d’accès à l’interprétation en LA, ce qui paraît impossible sans connaître la distribution des genres desservant le type de discours dont relève le texte à traduire. L’élaboration d’une cartographie générique relative aux discours de spécialité dans une perspective contrastive constituerait une aide inappréciable, surtout qu’il est rare qu’un traducteur soit en même temps expert en matière des textes qu’il traduit. Une telle étude génologique indiquerait des ressemblances entre deux polysystèmes discursifs différents, ainsi que des analogies interculturelles au niveau des critères de choix de stratégies verbales en fonction des intentions communicationnelles. Montrant des réseaux d’interférences génériques intralinguales et interlinguales, la cartographie générique contrastive permettrait d’éviter de traduire des textes de spécialité par imitation de leurs traits rédactionnels et d’en rendre les messages conformément aux normes discursives en LA. La connaissance en semble indispensable pour améliorer la compétence discursive non seulement des traducteurs et interprètes mais de qui apprend une langue donnée. Bibliographie Adam, J.-M. (1999), Linguistique textuelle. Des genres de discours aux textes. Paris : Nathan Universités. Beacco, J.-C. (2004), « Trois perspectives linguistiques sur la notion de genre discursif », in: Langages 153, pp. 109-119. Calvat, G. (2009), Les plans d’architecte et d’exécution. Athènes : G.R. Presse. Fèvre, B., Chailloux, J.-M. (2008), Réussir sa demande de permis de construction. Paris : Eyrolles. Fèvre, B., Fourage, S. (2010), Le mémento du conducteur de travaux. Paris : Eyrolles. Grzmil-Tylutki, H. (2007), Gatunek w świetle francuskiej teorii dyskursu. Kraków : Universitas. Lavault-Olléon, E. (2007), Traduction spécialisée : Pratiques, théories, formations. Bern, Berlin, etc. : Peter Lang Publishing. Reiss, K. (1995 / 2009), Problématiques de la traduction. Paris : Ed. Economica. Rastier, F. (2007), « La traduction: interprétation et genèse du sens », www.revue-texto.net/Lettre/Rastier_Traduction.pdf, 20.03.2010. TRANSLATION AND MEANING, PART 10, 2013, 181 - 193 193 Walkiewicz, B. (2009a), « Entre la bidimension et la tridimension ou de l’architecture d’un texte sur l’architecture », in : Synérgies Pologne, no 6, vol.II, Kraków, pp. 167-174. Walkiewicz, B. (2009b), « O barierach kulturowych w przekładzie tekstów specjalistycznych », in : Rocznik Przekładoznawczy 5, Toruń, pp. 213-225. Walkiewicz, B. (2009c) (sous presse), « Między formą a normą czyli o identyfikacji dyskursywnej w tłumaczeniu tekstów specjalistycznych ». Walkiewicz, B. (2010a) (sous presse), « Entre la sémantique et la pragmatique ou de la traduction des textes de spécialité ». Walkiewicz, B. (2010b) (sous presse), « Interdyskursywność w przekładzie tekstów specjalistycznych ». Wittgenstein, L. (1953 / 2000), Dociekania filozoficzne, trad. par B. Wolniewicz, Warszawa : PWN. Wojtak, M. (2009), « Wzorce gatunkowe wypowiedzi a realizacje tekstowe », in : Polska genologia lingwistyczna, D. Ostaszewska (éd.), Warszawa : PWN, pp. 353-361. 194 195 Section III: The Translator & Translator Education 196 TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 197 BEING A TRANSLATOR: HOW DOES IT FEEL? Joanna Albin Pedagogical University of Kraków, Poland Abstract: The aim of this paper is to present the results of a study on the affective aspects of the translator profession. Their description revolves around concepts developed within the field of psychology: self-efficacy, attributions and motivations. The results obtained reveal links between attitudes and certain skill patterns. Namely, high self-efficacy and a favourable attribution style correlate with high level management and CAT skills. Also, criteria applied for selfevaluation prove meaningful. Translators with high self-efficacy evaluate themselves against money and prestige criteria, while professionals with deficient self-efficacy are mostly selfreferenced. This proves the importance of the social components in the translation profession. 1. Introduction This paper presents some of the results of the study on translators’ activity, undertaken in a specific situation of Poland, where institutional training of translators was non-existent until some years ago. Even now, it is carried out mostly within philological studies. The author herself, a self- educated translator and translators’ trainer, belongs to the generation in whose training universities have borne quite little, if any, merit. Despite this fact, the integration of the experiences and intuitions of the author and of her colleagues, allowed for designing a coherent research study, whose main aim is to make translators’ self-learning explicit. Self-learning, as an universal human capacity of reacting to the environment, offered an integrating perspective for all kind of personal, behavioural and social factors intervening in the translators’ activity. All the same, we approached translation as a self-didactic process. We believe that the results may not only cast some light on learning processes outside educational institutions, but also contribute to the well-being, prestige and support to professional translators. Although abundant data were collected, herein we will focus on the affective aspects of the translator profession. They were approached by the exploration of three concepts: self-efficacy beliefs, motivations and failure attribution. 2. Universe, sample and methodology We examine these aspects of translation activity, which have remained largely ignored so far. None of them were studied systematically within Translation Studies, what obliged us to take advantage of copious research results by psychologists and pedagogues. Consequently, in many aspects it constitutes an exploration study, which will hopefully serve as orientation to future research. Regarding conceptual design, it is enlightened by conceptions developed within the fields of psychology and pedagogy, especially the Social Cognitive Theory (Bandura, 1977a, 1977b, 1999, 2006). The universe of the study is constituted by Spanish translators, with a minimum of five years of professional experience, educated and based in Poland. The population was estimated at the level of approximately 400 professionals. Spanish is a minority language in Poland, however not extremely rare like Asiatic or African languages. Moreover, the population of potential users of Spanish has grown dynamically over the last 15 years as a consequence of the broadening of the educational offer of secondary and higher schools. However, as English has almost monopolised the public discourse and economic reality, studies of this professional group do not exist at all, resulting in certain inferiority complexes and lower self-esteem of individuals specialised in Spanish. Thus, we consider that the more attention placed on this group may help to discover interesting relations between performance and self-concept. 198 JOANNA ALBIN Data elicitation was accomplished by means of a questionnaire based on retrospection and introspection. It consisted of questions regarding translators’ working habits and methods, their motivations and feelings related with their professional activity, their self-image and the initiatives they take to improve or change it. According to the intersubjective approach, the reflection that takes place in the translator's mind, prompted by the questionnaire, is an act of negotiation between the observer and the observed, based on the shared ground of language, translation profession, and training (Vrasidas, 2000). In the very moment of completing the questionnaire, mechanisms of introspective and retrospective reflection are triggered, offering data on the whole translators' working environment, including aspects such as rates, work organisation, type of translation performed mostly (which let us conclude on market demands), emotions and attitudes in negotiating and management situations, etc. What we obtain, thus, is a broad panorama as perceived and experienced by the translator. In change, we resign from triangulation data by means of results obtained with experimental, objectivist methods. 3. Demography Feminisation of the profession is a striking fact. Only 10% of the sample are men. Given the relatively small number of subjects included in the study, we resign from analysing our data from a gender perspective. Nevertheless, the overwhelming domination of women within the translation profession cannot be accidental and should be studied with techniques broadly known and used within the sociology and psychology fields. The Master’s Degree indicated by a large majority of translators was Spanish Philology (80%). Another large group are Romanic philology graduates (10%). More than a half of the sample (54%) stated having studied at the Jagiellonian University of Cracow. Given the reduced dimensions of the sample, it would be ill-considered to conclude on the dominant role of this university in Spanish translators’ education, however old and prestigious it is. It surely offered Spanish studies earlier than many others, which started to produce Spanish graduates as late as in the mid-Nineties. Nowadays, there are nearly 20 universities in Poland that offer studies in Spanish language. Thus, our data confirm that the present educational panorama in Poland is a result of a dynamical educational progress over the last 20 years. Translators included in our population had from 5 to 30 years of experience. However, experience cannot be simply calculated as the-number-of-years-since-graduation. Translators evaluated their experience subjectively, taking into consideration pauses in professional practice and, on the other hand, having started it before graduation. There is quite a large number of young translators (28%), with a minimum number of years of experience required, graduated even as late as 2008. 4. Affective factors 4.1 Self-efficacy The construct of self-efficacy was proposed in 1977, in the paper titled "Self-efficacy: Toward a Unifying Theory of Behavioral Change", and later elaborated in other Bandura’s works. "Perceived self-efficacy is concerned with people’s beliefs in their capabilities to produce given attainments. One cannot be all things, which would require mastery of every realm of human life. People differ in the areas in which they cultivate their efficacy and in the levels to which they develop it even within their given pursuits. For example, a business executive may have a high sense of organizational efficacy but low parenting efficacy. Thus, the efficacy belief system is not a global trait but a differentiated set of self-beliefs linked to distinct realms of functioning. Multidomain measures reveal the patterning and degree of generality of people’s sense of personal efficacy." (Bandura, 2006: 307). Why is self-efficacy so important? One of the most important characteristics of self-efficacy beliefs is their projection into the future. When faced with a task, an individual analyses its properties and TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 199 context and determines steps to take in order to accomplish it. Self-efficacy intervenes at this stage to organise and guide the course of action. Depending on its level in a particular context, it may lead to undertaking or avoiding the task, thus it constitutes a task selection factor. Also in the later stage, "efficacy expectations determine how much effort people will expend and how long they will persist in the face of obstacles and aversive experiences" (Bandura 1977a:194). 4.1.1 Self-efficacy on graduation Regarding their master’s degree studies, translators answered the question: Do you think you were properly prepared for the profession on graduation? It was intended to obtain data on translators' evaluation of curricula and teaching methods from the perspective of their usefulness in professional exercise. The scarce proportion of 18% responded positively (Fig. 1). Fig. (1): After graduation prepared for profession Translators graded different reasons for their poor preparation on graduation. All agreed on three reasons, giving them high ratings: no translation teaching, no specialisation and literary/linguistic specialisation only. All considered lack of self-motivation not relevant. The evaluation of the received education is thus radically negative. 4.1.2 Overall self-efficacy Do you consider yourself prepared for the profession currently? was the question intended to provide information on translators' current both competence and self-efficacy (Fig. 2). 20% of the sample responded "excellent" and 55% "satisfactorily". Thus, most translators evaluate positively their competence, which we consider the expression of their high self-efficacy. This is another proof of self-learning being so common in this sector. In spite of a rather poor preparation for the profession on graduation, self-learning becomes an instrument of active and dynamic competence formation, opening the path of professional success. Still, 20% of the sample encounters occasional problems and 5% often feel not properly prepared. This is quite surprising, given the fact that all translators included in the study were supposed experienced professionals. We hope that focusing on this group can bring about interesting findings regarding the link between competence, self-efficacy and the actual level of performance. 200 JOANNA ALBIN Fig. (2): Currently prepared for profession 4.1.3 Skills The subsequent item allowed translators to evaluate their detailed set of skills. The whole sample agreed on 6 competences, rating them very highly: research skills, native language fluency, foreign language fluency, self-management, computer skills and general knowledge. Fig.( 3): Skills TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 201 3,87 4,07 4,47 4,33 3,33 2,00 3,20 3,67 3,07 N 15 15 15 15 15 15 15 15 15 15 15 2 Mean 4,67 managers N 9 4,89 4,22 4,00 4,67 5,00 4,78 1,67 4,00 4,00 3,67 3,56 9 9 9 9 9 9 9 9 9 9 9 3 jacks of Mean 4,12 all trades N 16 4,00 3,62 2,94 3,75 4,19 3,69 1,88 1,50 2,19 2,00 1,69 16 16 16 16 16 16 16 16 16 16 16 Total 4,48 4,02 3,53 4,07 4,48 4,17 2,38 2,25 2,97 3,00 2,62 15 Mean 4,45 Team Work Negotiation CAT Research Self-promotion Work organisation 4,33 Accountancy Specific field knowledge 4,73 1 CATpositive Computer General knowledge Mean 4,67 Foreign language Native language The rest of the skills received differentiated ratings, so cluster analysis was carried out, showing 3 clusters (Fig. 4). Cluster 1 and 2 give similar ratings to all the skills except for two: cluster 1 "CATpositive" has a high level of CAT skills, while cluster 2 "managers" has good control of management skills such as promotion, negotiation and accountancy, but an extremely poor knowledge of CAT programs. Cluster 3 "jacks-of-all-trades" is the most numerous (15 cases), but they do not have any skill at a high level to be proud of but the six general ones. They have no knowledge of CAT programs, cannot negotiate or promote themselves efficiently, are not good at team work. They are not good at any speciality either. Fig.(4): Skills cluster analysis Even a rough comparison of skills level clusters with overall self-efficacy scores shows that CATpositiveness correlates with translators' high level of self-efficacy. Still, the highest self-efficacy is observed among “managers”. They also have the highest level of specific field knowledge. On the other hand, "jack-of-all-trades" are translators most dissatisfied with their preparation. 4.1.4 Subcompetence missed most Having evaluated their set of competences, translators were asked to point at one single skill which they wish they had most in their actual working conditions. The answer was expected to be partly redundant with the results of skills evaluation, although not completely. We assumed that translators may not be competent at some skills and not miss them at all. Thus, once again we allowed translators to evaluate their working conditions and specialisation and choose a skill they really need. This approach is consistent with our view of translators as, in the first place, specialising professionals, and in the second place, as individuals diagnosing actively their learning needs. The higher score was registered for specific field knowledge, followed by CAT skills and self-promotion. These three skills prove central for 3 clusters we identified. We will call them “knowledge missing”, “CAT missing” and “management skills missing” clusters, the latter including skills in negotiation, self-promotion and team work. The crosstabulation of missing skill clustering with skills evaluation, reveals that the skill missed most receive only slightly lower ratings than the "strong" skills. For example, "knowledge missing" cluster has a set of "strong" skills such as bilingual fluency, research skills and computer skills. Their CAT and market skills are really poor (below 2,5), however they do not miss them at all. The desired skills are knowledge related, but they receive relatively high scores (2,94 and 3,75). Similarly, cluster 2 "CAT missing" do not bother about their poorest skills: negotiation and promotion, but in change 202 JOANNA ALBIN miss CAT skills and accountancy skills. Cluster 3 "management skills missing" do not use and consider they do not need CAT programs (evaluated at the lowest level) nor accountancy, but wish they had a better grasp of negotiation, self-promotion and teamwork skills. We conclude that a decent level of a skill enables translators to be aware of the need of their further development. It also proves that these are "active" skills, i.e., those needed in the translators' everyday activity within their particular working environment, market sector or specialisation. Some of the skills are discarded as useless in a given environment. 4.1.5 Criteria of evaluation Since we were asking translators to evaluate their competence and skills, we were also interested what are the criteria used by them to evaluate themselves. That is, translators informed us on the values and standards they refer to when monitoring their performance and success. From five possible criteria proposed, one was internal (self-evaluation) and the rest social. Cluster analysis showed three clusters, which were named "self-referenced", "direct feedback" and "money and prestige" (Fig. 5). Self-analysis is considered important by virtually all translators. Still, the group of seven translators who gave the highest scores to self-analysis, also gave radically low ratings to the other factors. This suggests that they do not ever take into consideration any social factor to evaluate their performance. Cluster 1 "direct feedback" refers mostly to clients and fellow translators to monitor their performance. Self-analysis is also quite important for them, but the money they earn or official diplomas are quite irrelevant. This cluster is the most numerous (24 cases). For cluster 3 "money and prestige", all factors are relatively important (the lowest score for colleagues feedback), but they are the only eight professionals who take into consideration negotiated rates and professional recognition. Selfevaluation direct feedback self-referenced Client’s feedback Colleagues’ feedback Negotiated rates Professional recognition Mean 3,83 4,26 4,13 2,35 2,91 N 23 23 23 23 Mean 5,00 2,86 1,57 1,14 1,43 N 7 7 7 7 4,75 3,25 4,62 4,38 8 8 8 8 Mean 4,18 4,11 3,47 2,61 2,95 N 38 38 38 38 23 7 money and prestige Mean 4,50 N Total 8 38 Fig. (5): Criteria of evaluation clusters Crosstabulation of criteria of self-evaluation with overall self-efficacy shows interesting results. The most self-efficient group is the "money and prestige" cluster. They all feel satisfactorily or excellent prepared for professional activity. We cannot confirm whether their rates are high or low, but their optimism may suggest economic success. On the other hand, the less competent are, to their own view, the self-referenced translators. This is another confirmation that isolation correlates with disfavourable results such as low self-efficacy. Crosstabulation with skills confirms that "money and prestige" group gives themselves the highest ratings in the set of CAT and market skills. On the other hand, quite foreseeably, direct-feedback cluster has a high level of teamwork skills. TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 203 4.2 Motivations Studies on human motivation emerged from humanistic psychology, which became an influent perspective in psychology in the 1950s. One of the early contributions was that of the Maslow’s pyramid of needs (1943). Needs are defined as factors stimulating actions aimed at satisfying them. Different needs that an organism can experience, do not ever manifest themselves at the same time. They form a hierarchical system, usually represented as a pyramid. High needs are not perceived unless the needs of a lower order are satisfied. For example, if a basic physiological need, such as hunger, is not satisfied, all the other needs are pushed to the background. Morality Creativity, Spontaneity Problem solving Lack of prejudice Self-actualisation Acceptance of facts Confidence, achievement Self-esteem Respect of others, respect by others Esteem Friendship, family, sexual intimacy Security of the body, of employment, of resources Safety Physiological Of morality, of the family, of health, of property Breathing, food, water, sex, sleep, homeostasis, excretion Fig. (6): Pyramid of needs (Maslow, 1943) Needs of higher order involve mental phenomena, both personal and social. The highest need of selfactualisation refers to “peoples’ desire to become more and more what one idiosyncratically is, to become everything that one is capable of becoming” (1943:22). At this level, the way of satisfying needs differs greatly from one individual to another. What is worth underlining is the “self-reference” component, common to self-efficacy concept of Bandura and Maslow’s Theory of Motivation. Also social components are present in both: belonging and esteem needs are satisfied by inner emotional states in social interaction. This is also the case of self-efficacy. To learn about translators’ needs, we asked about their other occupations. Only 35% of translators limit their activity to translation. When enquired on the reasons for maintaining other occupations, most translators pointed at avoiding isolation, which typically afflicts the translators’ population (26%). The second and the third position occupy seeking intellectual challenge (22%) and seeking regular income. The less relevant were seeking social benefits (16%) and seeking social prestige (15%). 204 JOANNA ALBIN Fig. (7): Translation as main occupation Fig. (8): Causes of moonlighting TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 205 Cluster analysis was carried out to unravel the differences within the sample (Fig. 9) Ratings of the answers isolation and social benefits are comparable in all clusters. In contrast, ratings of the remaining answers are differentiated, revealing 3 clusters. Cluster 1 (“prestige") is the only one to justify their additional occupations by the need of social prestige. Cluster 2 ("intellectual"), gave a high score, along with isolation, to seeking intellectual challenge. In comparison with the cluster 1, who gave relatively generous ratings to all items, cluster 2 is more radical, clearly emphasising intellectual challenge and isolation. In cluster 3 ("income") the dominating reason of moonlighting is lack of regular income. Analysis of this data provides important hints on translators' motivations and affective, personal and economic needs. No Regular Income 1 prestige Mean N 2 intellectual Mean N 3 income Mean N Total Mean N No Social Benefits No intellectual Challenge No social Prestige Isolation 3,50 3,12 4,62 4,50 4,62 8 8 8 8 8 1,57 1,57 4,14 1,57 4,00 7 7 7 7 7 4,45 2,55 1,91 1,36 3,45 11 11 11 11 11 3,38 2,46 3,35 2,38 3,96 26 26 26 26 26 Fig. (9): Not main occupation causes - cluster analysis The causes of moonlighting suggested to translators and evaluated by them are, as we believe, the expression of translators’ needs. Among all motivations, we clearly observe the domination of high needs. This is consistent with Maslow’s claim that in a developed society basic needs are rarely unsatisfied. According to this statement, we should interpret that only translators who have their basic needs covered (including the sense of belonging (i.e. not being isolated) are able to perceive needs of a higher level (self-actualisation expressed by intellectual challenge need). Still, the highest ratings were conceded to isolation. Moreover, crosstabulation with criteria of evaluation reveals that a considerable group of isolated translators are not able to deal with this situation within the translation profession and do not take advantage of feedback offered by colleagues. We believe that these results should draw the attention of translation professionals and scholars to social aspects of translation activity. 206 JOANNA ALBIN Fig. (10): Moonlighting / criteria of evaluation clusters Colleagues Feedback not important at all quite important very important most important Total Isolation not important at all 1 0 1 0 2 not very important 0 0 0 1 1 quite important 0 1 3 0 4 very important 1 1 2 3 7 most important 3 2 3 3 11 5 4 9 7 25 Total Fig. (11): Isolation / colleagues feedback crosstabulation 4.3 Failure attribution Attributions are one of the constructs used to describe motivations and actions within personality psychology. They reflect people's struggle to discover causes of events that affect their lives. The term was introduced by Heider (1958), who observed that some individuals are convinced that they have control of events. They believe that their success is due to their own skills and efforts, whereas, in case of failure, the lack of capability or effort is to blame (internal attribution). On the contrary, others believe that events are independent of their control and they attribute their causes to external factors, such as luck, task difficulty, other people's actions, weather, etc. (external attribution). This TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 207 conception inspired many further studies. Social cognitive theorist Rotter (1966) designed an I-E (internal-external) scale to measure individuals' attribution style, which he called locus of control. Weiner (1974) defined two more criteria of attribution: stability and controlability. The combination of attribution characteristics may have favourable or unfavourable effects on performance. Given their future-projectedness, attribution style has a particular impact on self-efficacy beliefs. For example, attributing success to an internal and stable factors will have favourable effects on motivation and performance. On the contrary, attributing failure to internal stable factors, such as lack of ability, is unfavourable and can undermine one's self-efficacy. When failure is attributed to internal unstable factors, such as temporal lack of attention or effort, a positive base for self-improvement may arise, without damaging the individual's self-efficacy (Bandura, 1999). Therefore, we tried to find out what are the translators' attribution styles. In order to discover possible differences within the sample we used cluster analysis. Answering the question In situations you were dissatisfied with your performance, what was normally the reason?, translators graded four options responding to internal or external and stable or unstable forces, as shown in the table below (Fig. 12). favourable Attribution style external + stable internal + unstable internal + stable external +unstable Did not do everything necessary Difficulty of assignment 1 favourable 2 unfavourable 3 mixed Total unfavourable Did not have appropriate skills Did not have luck Mean 3,20 1,40 1,75 1,10 N 20 20 20 Mean 1,57 1,43 4,71 1,43 N 7 7 7 Mean 4,23 2,77 4,00 1,08 N 13 13 13 Mean 3,25 1,85 3,00 1,15 N 40 40 40 20 7 13 40 Fig. (12): Failure attribution cluster analysis Two answers clearly got the highest scores: assignment difficulty and lack of skills. At the first sight they may seem synonyms, but they differ from the point of view of an internal or external perspective. Cluster analysis confirms that cluster “favourable” chooses assignment difficulty (3,20) and cluster “unfavourable” gives extremely high rating to lack of skills (4,71). They also give ratings higher than others to did not have luck, which proves that they feel they cannot control anything and are completely helpless in the face of failure. Thus, cluster “favourable” has the most favourable and the cluster “unfavourable” the most unfavourable failure attribution. Cluster “mixed” gives equally high ratings to assignment difficulty and lack of skills and also relatively high ratings to did not do everything necessary (2,77). This factor is internal but unstable, which means that the lack of effort was temporal and can be corrected. It may have, thus, either a demotivating effect if an individual blames herself for failure to insistently, or may trigger efforts leading to improvement without undermining self-efficacy. This group reveals a mixed attribution style, however the emphasis on did not do everything necessary suggests they could be self-demanding but willing to improve. 208 JOANNA ALBIN Crosstabulation of failure attribution clusters with other variables prove that translators with favourable failure attribution also tend to have high level of self-efficacy. They also evaluate themselves against money and prestige criteria and, a large proportion of them, against direct feedback. On the contrary, all translators who often feel not properly prepared, have unfavourable failure attribution and tend to be self-referenced. 5. Conclusions According to Bandura, the importance of self-efficacy and other affective factors consists in the possibility of influence of the subjective perceptions on the actual level of performance. Therefore, we believe, they should be thoroughly studied as basic factors in translation activity. Moreover, they allow for linking phenomena such as persistence or abandonment of the profession, quality deficiencies, or lack of balance among different motivations (dominance of economic or intellectual motivations). The review of our exploration data on the affective and behavioural aspects of the translator profession reveal the existence of certain patterns. At this stage, we cannot explain the causal link between them. Still, it would certainly be interesting to discover the mechanisms linking favourable failure attribution and high level of self-efficacy with high level of management and CAT skills, as well as with prestige and economic criteria of evaluation. Self-efficacy: HIGH failure attribution: FAVOURABLE Evaluation criteria: MONEY AND PRESTIGE Strong skills: management, CAT tools Self-efficacy: LOW failure attribution: UNFAVOURABLE Evaluation criteria: SELF-REFERENCE Strong skills: -- Consistently with our declared perspective, translators’ self-evaluation provides a range of significant data. On one hand, it informs on the social dimension of the profession, including working environment or economic conditions. On the other hand, they reflect the individual dimension, including personal experiences and self-reflection of translators, their beliefs and motivations, their persistence or abandonment of profession, and their negotiation, self-management and evaluation practice. The subjective nature of data should not be seen as a methodological defect. If we pretend a complete description of cognitive and social phenomena of similar complexity it is compulsory to resort to those who constitute it, and thus are the most competent to describe it: the translators. References: Bandura, A. (1977a), "Self-efficacy: Toward a unifying theory of behavioral change", Psychological Review, 84, 19-215 Bandura, A. (1977b), Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1999), Social cognitive theory: An agentic perspective, Asian Journal of Social Psychology, 2, 21-41. Bandura A. (2006), “Guide for constructing self-efficacy scales”. In: Pajares, F., & Urdan, T. (eds.), Adolescence and education, Vol. 5: Self-Efficacy Beliefs of Adolescents. Greenwich, CT: Information Age Publishing. pp. 307–337. Centrum Badania Opinii Społecznej, (2009), Prestiż zawodów, accessed online http://www.cbos.pl/SPISKOM.POL/1999/K_032_99.PDF on 2010-10-10. Heider, F. (1958), The Psychology of Interpersonal Relations, New York: Wiley. Maslow, A. (1943), A Theory of Human Motivation, Psychological Review, (1943), Vol. 50 :4, pp. 370–396), accessed online on 2010-10-10 at http://psychclassics.yorku.ca/Maslow/motivation.htm. Vrasidas, C. (2000), "Constructivism versus objectivism: Implications for interaction, course design, and evaluation in distance education", International Journal of Educational Telecommunications, 6(4), 339362. Weiner, B., (1974), Achievement Motivation and Attribution Theory. NY: General Learning Press. TRANSLATION AND MEANING, PART 10, 2013, 197 - 209 209 Rotter, J. (1966), “Generalized expectancies for internal versus external control of reinforcements”. Psychological Monographs, 80, Whole No. 609. 210 TRANSLATION AND MEANING, PART 10, 2013, 211 - 220 211 NON-FORMAL ELEMENTS IN ACADEMIC TRANSLATOR EDUCATION: WHY, HOW AND WHAT FOR? Konrad Klimkowski John Paul II Catholic University of Lublin, Lublin, Poland Abstract: The author proposes that the formal curriculum of academic translator education should recognize the benefits of non-formal educational initiatives in translator education. As it stands, this thesis may seem obvious for a large number of translator trainers: translation projects for students and introducing real business projects to the translation classroom have become standard practice. However, the author argues for a closer interrelation between the formal curriculum, the non-formal initiatives and the life-long educational perspective in academic translator education. 1. Introduction Until the year 2000, the debate on translator education was predominated by the attempts to define the concepts of translation competence and to develop models of its acquisition and evaluation (cf. Schaeffner and Adab, 2000). Donald Kiraly, in 2000, introduced yet another perspective on translator education: the student-teacher interaction. Kiraly’s views are difficult to summarize in brief. Nevertheless, the author pinpoints four main observations made by Kiraly (2000), since they constitute the basis for the arguments and proposals made in the main part of this paper. The first two observations express Kiraly’s criticism of the “traditional classroom.” Drawing upon Nord’s (1996) description of the translation classroom methodology named “who takes the next sentence,” Kiraly exposes all the grave consequences of the so-called transmissionist approach to (translator) education. Firstly, it relies on the central role of the teacher: they choose the content, the methods and “the best/true” answers to translation problems. In consequence, the teacher is completely in control of the translation process and product. Students’ performance is thwarted, since the central position of the teacher renders it utterly impossible for students to realize the translation task. The consequences of this approach for the development of students’ translation competence are, according to Kiraly, more than obvious: cognitive skills cannot develop well in the transmissionist context, let alone students’ self-esteem, motivation and interests. The main, strategic didactic problem that Kiraly (2000) points out in his criticism of the transmissionist approach is that teachers believe too much in the false metaphor of teaching,1 rather than concentrating on helping students to learn. There are two remedies that Kiraly (2000) proposes. Firstly, to overcome the drawbacks of the transmissionist approach, teachers should realize the potential of the social constructivist perspective on education. It has it that knowledge or skills are not transferred or taught. They can only be learned or developed by each person individually. Nevertheless, the growth of an individual is fostered when people learn in groups and teams. This is why Kiraly favours an educational model where learners collaborate in a shared environment in the classroom. In this way, Kiraly (2000) makes the point that the student-teacher relationship is pivotal for the success of any translation curriculum. When reflected upon, this relationship provides grounds for the development of students’ declarative and procedural knowledge as well as their educational and professional attitudes. Kiraly’s (2000) empowerment approach results in a holistic curriculum, where the student is seen as a person and not just a client of an educational institution. 1 A belief that one person can pass knowledge on another person directly. 212 KONRAD KLIMKOWSKI In fact, the holistic nature of Kiraly’s educational empowerment is emphasised by yet another aspect of his approach, namely the transformative view of education.2 This view is inspired by an educational theory known as the Transformative Theory of Learning. This theory has it that learning is not about acquiring or accumulating knowledge, but that the main sense and result of learning is the change, the transformation of the learner into someone ‘new’. Most researchers accept the holistic understanding of the transformation: the change concerns the emotional and the axiological facets of human functioning, along with the cognitive domain. Miller and Seller (1985), referred to by Kiraly (2000), claim that the transformation also takes place in the classroom when a constructivist teacher creates an exploratory environment for their students. In this case, the student-teacher interaction enables the transformation of all the protagonists of the educational process: the students and the teacher. Donald Kiraly makes a strong appeal to teachers and curriculum designers to introduce empowerment into the formal academic curriculum. The author of this paper fully supports this view, and admits that Kiraly’s (2000) model has been inspirational for his own educational practice and research. Nonetheless, the author ventures a claim that comprehensive translator education must not rest on empowering the formal curriculum exclusively. This paper discusses reasons for expanding the formal educational framework. Apart from explaining why, the author demonstrates how to expand the formal framework; in other words how to implement empowerment beyond the formal classroom. In an attempt to provide the reason for expanding the formal framework of translator education, the author refers to three concepts that recur in the debate in the field of adult education: autonomous learning, heutagogy and non-formal learning. 2. Autonomous learning González Davies (2004: 14) enumerates “three main approaches to the learning process”: the transmissionist, transactional and transformational educational styles. Since the first and the last approach have been discussed above, the author confines himself to a short presentation of the transactional model, which is defined by González Davies (ibid.) as “based on cooperative learning, there is group work and interaction, but the teacher still has the final answer to the problems set in the activities.” In González Davies’ view, this approach marks a transition stage between the transmissionist and the transformative position. Grow (1991: 129ff.) also maintains that learning cannot be governed by only one educational strategy and that learning styles change in time. The stages that a learner can go through are presented below: Stage Dependence Interest Engagament Autonomy Student Teacher Relies on teacher for the choice of Chooses contect, elicits results, evaluates content and methods Learns through own work Motivates, inspires, leads to enthusiastic excitement with learning Learns mostly through organizing Encourages and supports students’work, and realizing objectives that they they both seek knowledge in cooperation planned for themselves Consults, shows direction, gives advice Table (1): Four stages of learning by Grow (1991) 2 Kiraly’s makes a direct reference to Miller and Seller (1985), who belong to a large group of advocates of the Transformative Theory of Learning. Jack Mezirow is widely recognized as the theory’s proponent, however other researchers have also contributed to the definition of educational transformation, including Robert Boyd, Gordon Myers or Edmund O’Sullivan. See e.g. Boyd and Myers (1988), Mezirow (2000) or O’Sullivan, Morrell and O’Connor (2002). TRANSLATION AND MEANING, PART 10, 2013, 211 - 220 213 Although González Davies (2004) presents the three approaches as “teaching styles”, and Grow (1991) talks about “learning styles”, one could observe obvious parallels between these two perspectives. Firstly, they accept the fact that learning and teaching must adapt to a given stage in the learner’s development. Secondly, both authors stress the transition that the learner and the teacher can experience in the classroom (and beyond). It is true that the transactional approach in González Davies (2004) and the interest stage in Grow (1991) do not necessarily match in a one-to-one fashion. At the same time, the assumption that there can be a reasonable degree of overlap between the two is viable. However, the main point that makes González Davies’ continuum distinct from Grow’s is the stage of autonomy, which is proposed only by the latter author. This difference is obvious in view of the fact that autonomy concerns learning rather than teaching. In fact, autonomy as a learning style is a necessary prerequisite for the engaged or transformative stage in learning or teaching. However, Grow’s proposal may have a more radical reading, under which learning with the help of a teacher is only part of learning understood as a holistic, life-long process. In other words, at some stage learners abandon the formal framework and develop their own self-directed learning skills. In this paper, the author discusses the concept of heutagogy, as proposed and advocated by Hase and Kenyon (2000), since it illustrates well a radical approach to deschooling education. 3. Heutagogy The concept of heutagogy is easier to explain when presented in juxtaposition to the notion of andragogy. The latter was proposed by one of the most prominent American researchers in the field of education, Malcolm Knowles. The concept of andragogy, that is “pedagogy of the adult” or adult education, relies on Knowles’ claim that adults and young adults require educational strategies that are substantially different from those applicable for children or teenagers.3 In fact, the transition stages suggested by Grow (1991) and González Davies (2004) – as discussed above – go hand in hand with the idea of change from the pedagogical towards the andragogical approach to learning and teaching. On the one hand, Hase and Kenyon (2000) accept the fact that human learning changes with age. On the other hand, they seem far more critical of the formal, teacher-dependent educational formats than Knowles, Grow or González Davies are, as evidenced by the following quote (Hase and Kenyon 2000: 2): “While andragogy […] provided many useful approaches for improving educational methodology, and indeed has been accepted almost universally, it still has connotations of a teacher-learner relationship. It may be argued that the rapid rate of change in society, and the so-called information explosion, suggest that we should now be looking at an educational approach where it is the learner himself who determines what and how learning should take place”. The above quote from Hase and Kenyon betrays their distrust of present-day educational practice, and – in the author’s opinion – is to be understood as a call for a review of the formal curriculum as such. What Hase and Kenyon propose instead is a radical vision of the learner whose autonomy reaches far beyond organizing their own deliberate practice with the use of material assigned by the tutor. In Hase and Kenyon’s conception of heutagogy, the learner becomes self-directed: they choose 3 Although Knowles presented the concept of andragogy already in the 1970s, the author refers the reader to one of the later works by Knowles and his collaborators, that is (Knowles, Holton and Swanson 2005). The notion of andragogy is most extensively discussed in chapter 4 of the book (p.35 ff.) 214 KONRAD KLIMKOWSKI their own learning goals and methods.4 Thus, in the work of Hase and Kenyon (2000), learners’ autonomy reaches its extreme meaning. The authors formulate explicitly the claim that at some stage the learner must transcend the formal curriculum if they intend to successfully meet their individual and social educational needs. Hase and Kenyon’s critique is rooted in their recognition of the changing context in which the educational process takes place (Hase and Kenyon 2000: 2): “There is, however, another revolution taking place in educational circles that appears to go one step beyond andragogy, to a new set of principles and practices that may have application across the whole spectrum of the education and learning lifespan. This revolution recognises the changed world in which we live. A world in which: information is readily and easily accessible; where change is so rapid that traditional methods of training and education are totally inadequate; discipline based knowledge is inappropriate to prepare for living in modern communities and workplaces; learning is increasingly aligned with what we do; modern organisational structures require flexible learning practices; and there is a need for immediacy of learning. In response to this environment there have emerged some innovative approaches that address the deficiencies of the pedagogical and andragogical methods”. In other words, if education is to help the learner meet the demands of the contemporary world, it must seek paths towards heutagogy. There is yet another distinction between the andragogical and the heutagogical approach that Hase and Kenyon (2000: 5) reveal: “Knowles’ definition [of self-directed learning – KK] provides a linear approach to learning and sounds a little like the chapters of a train the trainer guide. Heutagogy takes account of intuition and concepts such as ‘double loop learning’ that are not linear and not necessarily planned. It may well be that a person does not identify a learning need at all but identifies the potential to learn from a novel experience as a matter of course and recognises that opportunity to reflect on what has happened and see how it challenges, disconfirms or supports existing values and assumptions”. Hence, learning cannot be limited to the formal educational context, as a lot of learning happens out of that context, even if it is inspired by what goes on in the formal curriculum. In this way, our debate on students’ autonomy and ways of understanding it leads us to the distinction between the formal and non-formal education and styles of learning. 4. Informal and non-formal education The concepts of formal, informal and non-formal education require some preliminary comments. Coombs, Prosser and Ahmed (1973) defined the three concepts as distinct but related. The relation may be presented graphically in the following way: 4 The concept of self-directed learning is another important notion discussed in the literature of the field. For reasons of clarity, the author abstains from further defining the notion. More details to be found in e.g. Knowles, Holton and Swanson (2005), Hase and Kenyon (2000) or Song and Hill (2007). TRANSLATION AND MEANING, PART 10, 2013, 211 - 220 215 FORMAL LEARNING institutionalized hierarchical chronologically ordered NON-FORMAL LEARNING INFORMAL LEARNING organized educational activity outside the formal curriculum attitudes, skills and knowledge acquired through daily experiences throughout life Figure (1): Formal, informal and non-formal learning (on the basis of Coombs, Prosser and Ahmed 1973) As may be inferred from Figure 3, formal and non-formal learning share the characteristics of being organized and planned, although with a different extent of institutionalization. Informal learning, in contrast, is devoid of direct institutional influences, although they can also be used to facilitate informal learning, listed among a cornucopia of experiences that make people construct knowledge, skills and attitudes. While the adjectives formal and non-formal are used to name forms of education, informal learning is perhaps best understood as a learning style. This latter concept concentrates on the learning individual, rather than describing the educational process. This is why, for the purposes of this paper, the author reduces the tripartite division presented above to the dichotomy between the formal and the non-formal educational frameworks, assuming that the informal aspect will always be present within the two approaches to education. An analysis of research in the field5 made the author adopt the following criteria to distinguish between the formal and non-formal education: formal education organized content planned and evaluated results taught informal/non-formal education incidental learning open-ended self-taught Table (2): Principles of formal and informal/non-formal education The criteria in the table above represent the extremes on the scale of contrasts between the formal and the non-formal educational frameworks. One of the most prominent researchers to adopt this contrastive view on formal and non-formal learning and education is Eraut (2000). In his opinion, learning takes place almost exclusively in a non-formal context. In this way, Eraut (2000) falls within the group of researchers of education who find non-formal education superior to the formal one, especially as regards adults and their professional development. It stands to reason to claim that his views would also be compatible with the stance of Hase and Kenyon (2000), as discussed above, or with Leadbeater (2000), who claims that (ibid.: 112) “more learning needs to be done at home, in offices and kitchens, in contexts where knowledge is deployed to solve problems and add value to people’s lives.” 5 The author mostly relied on Livingstone (2001), Livingstone et al. (2006) and also on Eraut (2000). 216 KONRAD KLIMKOWSKI However, the formal vs. non-formal dichotomy can be approached in a less radical way. While formal education always relies on organized content, it is also capable of recognizing the importance of those aspects of knowledge that are difficult to include within the curriculum. Internships and student practice are a way of admitting that the formal curriculum needs support from extra-curricular educational initiatives. Alternatively, not all sorts of non-formal education need to depend on unplanned, incidental learning exclusively. Garrison (1997) observes that self-directed learning, which the nonformal framework promotes, depends on conscious effort on the part of the learners in providing themselves with an adequate educational environment. Finally, there comes the problem of results and evaluation. Formal education is close-ended in that it expects measurable effects from the educational process. Non-formal education is said to be open-ended in that it does not rely on planned results. Again, these extreme positions can be seen as an abstraction, while educational practice calls a compromise between the two. The author finds questionable a claim that the formal curriculum can develop tools for precise measuring of all the effects of the educational process. Unlike e.g. Shreve (1995: xiv), who stated that “to build the competence we want in our students we have to design precise pedagogical tools – tools for particular purposes that will yield specific desired effects,” is not an advocate of a radically linear, effect-driven, behavioural concept of translator education. Take the above-mentioned idea of transformative learning for example: is it feasible to assume that the effects of such a transformation can be measured? Consequently, the formal curriculum must allow for “unplanned” results.6 At the same time, the non-formal approach to education is flexible enough to employ planned problem-based learning, tuition or facilitation and project work leading to material results (publication). Also, the non-formal framework can also make a good use of collaborative learning. The above arguments are meant to demonstrate that, apart from the radical, contrastive stance that sees non-formal education as superior to the formal one, an integrated view is possible under which the formal and the non-formal approaches merge. This view is represented in the literature of the subject e.g. by Billett (2001), who claims that the distinction between formal and non-formal learning is, in fact, false. On the one hand, Billett confirms Eraut’s claim that learning takes place predominantly in a non-formal environment. On the other hand, Billett asserts that learning always takes places in some form of formal (social) context: school, work, home or church. Thus, the formal and non-formal coincide and they must be seen as the two faces of one coin. Concluding this section, the author states that a number of researchers express their awareness that the formal curriculum on its own does not do the justice to the educational needs of the present-day adult learner. Some of the researchers argue that the sooner the curriculum becomes less formal, nonformal or self-directed (informal), the better for the learners, educators and the society at large (Eraut 2000; Hase and Kenyon 2000). Others opt for merging the formal and the non-formal aspects of learning and education (Billett, 2001). Both these approaches serve as the basis for the proposals made by the author in what follows. 5. Non-formal elements in support of the formal curriculum With all the observations made so far in this paper, the author wishes to proceed to his own theoretical and practical proposals concerning the introduction of non-formal elements into the formal framework of translator education. Drawing upon the research reported above, the author proposes that curriculum designers should plan a non-formal ‘programme’ of initiatives running parallel to the formal academic curriculum. These initiatives are not only to accompany the curriculum, but also to be an element of it. The proposal put here is that they should be integrated with the formal curriculum more closely and more purposefully. They can influence the way the formal curriculum is planned and realized (some formal content may be moved to the non-formal area), or may serve as a 6 See the concept of the ‘double loop learning’, as quoted above from Hase and Kenyon (2000:5). TRANSLATION AND MEANING, PART 10, 2013, 211 - 220 217 ‘testing ground’ for certain educational solutions that the teachers may need to observe in a nonformal context before implementing them in the classroom. Table (3) below sums up the major areas in which the formal curriculum can gain support from the non-formal programme. Formal curriculum Limited flexibility of curriculum design Limited degree of professional simulation limited degree of teamwork and collaboration Limited functionality of the grade system Limited levels of intrinsic motivation/self-determination Constrained involvement of students and teachers Non-formal programme Flexibility of choice (topics, methods, etc.) Greater degree of professional simulation Increased teamwork and collaboration Increased use of evaluation as a source of feedback Increased boost to intrinsic motivation/self-determination Greater involvement of students and teachers Table (3): Formal and non-formal translator education in contrast As with Table (2), contrasts that are shown in Table (3) represent extreme opposites. And also in this case the author reads these contrasts in a less radical way. Hence, the author does not claim that no flexibility in the formal curriculum is possible at all, but that it is reduced in contrast to the nonformal context. One reason for the reduced flexibility of formal education is that it serves the goal of certification: to be able to award certificates of education, universities must rely on a stable and transparent curriculum. Unfortunately, this often leads to the inadequacies remarked above in the quote from Hase and Kenyon (2000: 2). Non-formal initiatives, on the other hand, can cater for redefinitions of the scope of interest in accordance with the needs of the students or the market. It is also easier to extend the educational arena in the non-formal context by inviting experts or practicing translators. This invitation of real-life players enhances the simulation of the daily translator’s routine. And so, it has its projection onto the real engagement of students and teachers in teamwork over a translation project like the one described below. A key problem worth highlighting here is assessment. It is the author’s strong conviction that the formal model of education will always express evaluation by grades, points or other statistic means that are expected to provide objective information about the results of students’ learning efforts, but which often tend to be devoid of feedback information given to the student on the process of learning, their progress or regress. When used in this way, the assessment system is degraded to the function of a reward or punishment, understood in the terms of the classical conditioning theory. This reduction also suggests that students and teachers get trapped in the game of give-and-take concerning grades, which renders the development of skills or competences completely irrelevant. Grades without feedback put the educational process to an end, like settling the account ends a transaction. Whereas, in the author’s opinion, assessment (whether formative or summative, positive or negative) should always foster the existing or inspire new educational processes. The non-formal context allows students and teachers to free themselves from the game. The purpose of the non-formal initiatives never focuses on the academic ‘passmarks’. It opens a chance of creating a working and learning environment where learners want to do their best in realizing translation tasks. This view of assessment should help students realize that attitudes expected of them when they become 218 KONRAD KLIMKOWSKI professionals differ from those of students.7 Finally, the change in the model of assessment should have its bearing on students’ educational and professional intrinsic motivation. Real-life objectives, quality-oriented assessment of students’ work, as well as the fact that the results of their efforts are tangible (published), are likely to make a lot of students overcome their fear of failure as translators. 6. Case study As a way of demonstrating how the theses put forward in this paper can be put into practice, the author wants to present a report on a student translation project, in which students of Applied Linguistics at the UMCS (Maria Curie Sklodowska University) translated selected components of the Lublin City Office’s website into English. The project was realized in the years 2008-2010, with over a hundred students involved in three yearly cycles. The team members and some team managers were recruited from the second year students of the BA course in translation. There were also teams managed by MA course students. Students volunteered for the 2008 edition, while the 2009 and the 2010 editions were realized within the framework of regular students’ practice. In the 2010 edition, a number of students who were not obliged to participate in the practice joined in by forming two teams of eight members. MA students who were team managers were all volunteers. Each team, consisting of about seven members plus two managers, worked for a week on a commission from the Lublin City Office. There were two ‘working weeks’ per team each semester (4 texts per academic year). The ‘working week’ started on Friday, when the team manager contacted the City Office representative and was emailed the commissioned texts. Texts were in translation until late Monday, then they were peer-proofread.8 On Tuesday, the texts were finally checked through by the managers, whose role was either to send texts to the client, or launch the second cycle in the process in order to improve either the translators’ or the proof-readers’ work. On Wednesday or Thursday, the texts were sent to the reviewer in the City Office, and on Friday they were submitted to the City Office representative for publication. In most cases the City Office representatives were able to prepare texts in advance, so as to allow a week for the translation process and to avoid delay in the publication of the English language version of the service. When a given week was over, a feedback meeting was held for students, at which the City Office reviewer discussed the project with the students.9 The main reason for the author’s referring to the project in question is that it was originally devised as a test of how the reduced formality of the context in which a translation project was realized could influence the students’ performance. One of the crucial elements that the author wanted to achieve was to redirect students’ motivation from focus on grades towards focus on quality. To achieve this goal, the author needed to reframe the formal grading system into the system of feedback information on performance quality. For this purpose, the author decided to employ a double system of assessment. The first type of assessment was product-oriented: the text was accepted for publication by the City Office, or rejected. The other type of assessment concerned the particular team members and their individual performance. A special evaluation sheet was devised for that purpose, employing a number of criteria, such as original text analysis, language correctness, textual coherence and the communicative effectiveness of the target text. It also included points on terminology management and proof-reading done by the translators themselves. Although this latter assessment system used points, 7 See e.g. Fraser (1996), Jensen and Lykke Jakobsen (2000), Jonasson (1998), Künzli (2004) or TirkkonenCondit (1989). 8 By peer-proofreading the author means a system in which translator no. 1 in a team sent their text for proofreading to be done by translator no. 2 in the same team, translator no. 2 submitted their text for proof-reading to translator no. 3, and so on. 9 We (the author and the reviewer) were able to trace the development at the particular stage of project realization because we used two separate channels for file exchange. The translator was to produce their target file in two versions: a “clean” one, and a “trace-changes” one. The latter was sent to the reviewer directly. Similarly, the proofreaders were to sent their two files, and so were the managers. TRANSLATION AND MEANING, PART 10, 2013, 211 - 220 219 it was mostly employed to monitor progress in performance. So it happened that the same statistic score of e.g. 145 points (max. 200) could mean something different for a translator who showed progress than for the one whose performance worsened. Students were informed of their points and were given feedback on their progress. They were asked to comment on the data, the potential reasons behind the status quo, and also asked what they planned to do next as regards skill development. 7. Extending the curriculum The conclusion that the author wishes to draw here is that it is perhaps no longer enough to provide students of translation with a considerable amount of student practice as a form of completion of the formal curriculum. The point here is that our conception of the curriculum should evolve towards one in which the two areas are inseparably intertwined: the formal and the non-formal. The relationship between the two is illustrated below. Figure (2): A proposal of an extended curriculum From this viewpoint, the purpose of the non-formal component is not only to provide students with professional practice, but also to inform the curriculum of the changes it can undergo in order to empower translator education. This supportive function of the non-formal component concerns both content and methodology. One of the main methodological problems it can help to solve is how to adapt academic assessment to the requirements of professional quality management. High quality in a professional’s work depends predominantly on their motivation and accountability, which are hard to develop and monitor by means of standard academic grading procedures. The final point to make here brings us back to Donald Kiraly’s seminal work. It is the author’s strong conviction, based on his educational experience, that the non-formal component of the extended curriculum can help students and teachers develop new ways of defining their in-class interaction as a method in their collaborative search for knowledge and personal development. References Billett, S. (2001), “Learning through working life: interdependencies at work.” Studies in Continuing Education 23 (1). 19-35. Boyd, R.D. and Myers, G. (1998), “Transformative Education.” International Journal of Lifelong Education 7, 4. 261-284. Coombs, P.H., Prosser, R.C. and Ahmed, M. (1973), New Paths to Learning for Rural Children and Youth, New York: International Council for Educational Development. Eraut, M. (2000), “Non-formal learning, implicit learning and tacit knowledge.” Coffield F. (ed.) The Necessity of Informal Learning. Bristol: Policy Press. 12-31. 220 KONRAD KLIMKOWSKI Fraser, J. (1996), “Professional versus student behaviour.” Dollerup, Cay. and Appel, Vibeke. (eds.) Teaching Translation and Interpreting 3. New Horizons. Amsterdam / Philadelphia: John Benjamins. 243-250. Garrison, M. (1997), “Self-directed Learning: Towards a Comprehensive Model.” Adult Education Quarterly 48. 18-33. González Davies, M. (2004), Multiple Voices in the Translation Classroom: Activities, Tasks and Projects. Amsterdam / Philadelphia: John Benjamins. Grow, G. (1991), “Teaching Learners to Be Self-directed.” Adult Education Quarterly 41. 125-149. Also available at: http://www.longleaf.net/ggrow/SSDL/SSDLIndex.html (last retrieved on 21 December 2010). Hase, S. and Kenyon, C. (2000), “From Andragogy to Heutagogy.” UltiBASE Articles December 2000 Edition. Available at: http://ultibase.rmit.edu.au/Articles/dec00/hase2.htm (last retrieved on 21 December 2010). Jensen, A. and Lykke Jakobsen, A. (2000), “Translating under Time Pressure: An Empirical Investigation of Problem-solving Activity and Translation Strategies by Nonprofessional and Professional Translators.” Chesterman Andrew; Gallardo San Salvador, Natividad and Gambier, Yves (eds.) Translation in Context. Amsterdam / Philadelphia: John Benjamins. 105-116. Jonasson, K. (1998), “Degree of Text Awareness in Professional vs. Non-Professional Translators.” BeylardOzeroff, Ann; Králová, Jana and Moser-Mercer, Barbara (eds.) Translators' Strategies and Creativity. Amsterdam / Philadelphia: John Benjamins). 189-200. Kiraly, D. (2000), A Social Constructivist Approach to Translator Education: Empowerment from Theory to Practice. Manchester: St. Jerome. Knowles, M.; Holton, E.F. III and Swanson, R.A. (2005),The adult learner: The definitive classic in adult education and human resource development. Burlington, MA: Elsevier. Künzli, A. (2004), “Risk taking: trainee translators vs. professional translators. A case study.” The Journal of Specialised Translation 2. 34-49. Available at http://www.jostrans.org/issue02/art_kunzli.php (last retrieved on 21 December 2010). Leadbeater, C. (2000), Living on Thin Air. The new economy. London: Penguin. Livingstone, D.W. (2001), Adults’ Informal Learning: Definitions, Findings, Gaps and Future Research. Toronto: OISE/UT (NALL Working Paper No.21) at http://nall.oise.utoronto.ca/res/21adultsifnormallearning.htm (last retrieved on 21 December 2010). Livingstone, D.W., Raykov, M., Pollock, K. and Antonelli, F. (2006), Work and Lifelong Learning Resource Base: Materials for Teaching, Research and Policy Making. Toronto: Centre for the Study of Education and Work. Mezirow, J. (2000), Learning As Transformation: Critical Perspectives on A Theory in Progress. San Francisco: Jossey Bass. Miller, J.P. and Seller, W. (1985), “Transmission Position: Educational Practice” Miller, John, P. and Seller, Wayne (eds.) Curriculum Perspectives and Practice. New York: Longman. 37-61. Nord, C. (1996), “Wer nimmt denn mal den ersten Satz? Überlegungen zu neuen Arbeitsformen im Übersetzungsunterricht." Lauer, Angelika et al. (eds.) Translationwissenschaft im Umbruch. Festschrift für Wolfram Wilss. Tübingen: Gunter Narr. 313-327. O’Sullivan, E., Morell and M.-A. O’Connor (eds.) (2002), Expanding the Boundaries of Transformative Learning. New York: Palgrave. Schaeffner, C. and B. Adab (eds.) (2000), Developing Translation Competence. Amsterdam / Philadelphia: John Benjamins. Shreve, G.M. (1995), “Prescription and Description in Translation Teaching”. Foreword to Kiraly, Donald. 1995. Pathways to Translation. Pedagogy and Process. Kent, Ohio / London: The Kent State University Press. ix–xiv. Song and Hill (2007), “A Conceptual Model for Understanding Self-Directed Learning in Online Environments.” Journal of Interactive Online Learning 6, 1. 27-42. Tirkkonen-Condit, S. (1989), ‘Professional versus Non-professional Translation: A Think aloud Protocol Study.’Séguinot, Candace (ed.) The Translation Process (Toronto: HG Publications). 73-85. TRANSLATION AND MEANING, PART 10, 2013, 221 - 230 221 KOGNITIVE PROZESSE IN DER TRANSLATORISCH GEMITTELTEN INTERKULTURELLEN KOMMUNIKATION Anna Małgorzewicz University of Breslau, Poland Abstract: The present article considers translation as a form of intercultural communication. Besides, it takes into account aspects of cognitive processes which take place in the translator’s mental system in the receptive and productive phases. In order to explicate the cognitive operations, cognitivist developments within the area of text comprehension and production as well as Fillmore’s scene-and-frame semantics model and Gruczas translation model have been used.The paper comments on the interaction between semantics and the translator’s mental system in realizing an appropriate target translation equivalence which takes into account cultural background. 1. Modell des Translationsgefüges Im vorliegenden Beitrag wird von der Annahme ausgegangen, dass die Translation eine auf den Verstehensprozessen aufbauende Handlung im Rahmen der interkulturellen Kommunikation ist. Die im Titel des Artikels annoncierten Aspekte werden vor dem Hintergrund der im Translationsgefüge verorteten Translation erörtert. Unseren Überlegungen liegt das von Franciszek Grucza entworfene Modell des Translationsgefüges zugrunde, das sich schematisch wie folgt darstellen lässt: TRANSLATION initialer Sender AText mittelbarer Empfänger mittelbarer Sender BText finaler Empfänger Abb.(1): Translationsgefüge (vgl. Grucza, 1981, 1998, Kielar 2003) Der Translator – mittelbarer Empfänger und mittelbarer Sender zugleich – fungiert im illustrierten Gefüge als Vermittler zwischen dem initialen Sender des A-Textes und dem finalen Empfänger des B-Textes. Mittel der Translation sind der in der Ausgangkultur und deren Sprache verankerte A-Text sowie der in der Zielkultur und deren Sprache verankerte B-Text. Das Ziel des Translators ist es, eine funktionale – in manchen Fällen auch eine formale – Äquivalenz zwischen dem A-Text und dem BText zu erreichen. Die Voraussetzung für die Ausführung dieser Aufgabe sind die translatorischen Eigenschaften des Translators, die nach Grucza (1985:35) im Vergleich zu den Eigenschaften anderer bilingualer Personen einen gewissen Überschuss darstellen. Diesen spezifischen Eigenschaften gilt unser Augenmerk in weiteren Teilen des Beitrags. Im Einzelnen handelt es sich um kognitive Prozeduren, die bei der Aufnahme des ausgangssprachlichen Textes aktiviert werden, sowie um das metakognitive Wissen des Translators, das es erlaubt, den vom Ausgangstextproduzenten intendierten Sinn im mentalen System des Zieltextempfängers entstehen zu lassen. Zur Diskussion werden also zwei Fragen gestellt: Was geschieht im Kopf des Translators in der Phase der A-Text-Verarbeitung? 222 ANNA MAŁGORZEWICZ und Welche Strategien setzt der Translator bei der Umsetzung der in seinem Kopf hergestellten mentalen Repräsentation des A-Textes in die mentale Repräsentation des B-Textes ein? 2. Kognitive Prozesse des Translators 2.1 Professionelle Translatoren versus angehende Translatoren Aufschlussreich für die Klärung der gestellten Fragen sind Ergebnisse von Studien, die die mentalen Prozesse professioneller Translatoren mit denen von Anfängern vergleichen. Im Rahmen der Translationsforschung wurden auf diesem Gebiet zahlreiche Untersuchungen in unterschiedlichen translationswissenschaftlichen Zentren durchgeführt. Auf ihre Ergebnisse geht Paul Kußmaul (2009) in einem seiner Beiträge ausführlich ein. Relevant für die oben formulierten Fragestellungen sind folgende von Kußmaul (2009: passim) angeführte Ermittlungen im Bereich der Erforschung der rezeptiven Phase im Translationsprozess: - - - - - Professionelle Translatoren organisieren ihre Prozesse effektiver, indem sie das Sachgebiet des Textes im Auge behalten und sich von ihrem Verständnis von dessen kommunikativer Funktion leiten lassen (vgl. Jakobsen, 2005: 115, Krings, 1987: passim). Professionelle Translatoren setzen globale, den gesamten Text erfassende Strategien ein, angehende Translatoren dagegen bedienen sich eher linearer, d.h. die einzelnen Textabschnitte berücksichtigender Strategien (vgl. Krings, 1987: passim). Professionelle Translatoren setzen Strategien von ausgeprägterer Rekursivität ein. Sie verwenden globale Textverarbeitungsstrategien und bewegen sich souverän im Text vor und zurück (vgl. Opitz, 2004: 668f.). Professionelle Translatoren berücksichtigen bei der Ausgangtextverarbeitung den größeren Kontext, den Texttyp, die Textsorte und rhetorische Muster. Sie führen eine Makroanalyse durch (vgl. Barbosa/Neiva, 2003: 150). Professionelle Translatoren greifen bei der Textverarbeitung auf ihr Weltwissen zurück (vgl. Jääskeläinen, 1989: passim, 1996: passim, 1999: 120; Jensen, 2001: 167). Auf Grund der angeführten Untersuchungsresultate kann festgestellt werden, dass professionelle Translatoren im Vergleich zu angehenden Übersetzern und Dolmetschern in ihrer Ausgangstextverarbeitung autonomer und wissensbasierter verfahren. Sie unterstützen also ihre Verstehensprozesse mit der Analyse von Faktoren, die außerhalb der einzelnen Textsegmente liegen. Sie haften nicht am Wortlaut des Ausgangstextes, sondern versuchen diesen auf der Grundlage aller ihn konstituierenden Aspekte zu interpretieren. 2.2 Translationsrelevante Verstehensprozesse vor dem Hintergrund eines psycholinguistischen Verstehensmodells Aufschlussreiche Einsichten in das Zusammenspiel von außersprachlichen Faktoren und Textsegmenten bei den Verstehensprozessen liefert ein psycholinguistisches Modell, das, zwei Prozesse umfasst – auf der Konferenz der Polnischen Gesellschaft für Angewandte Linguistik im März 2010 bottom-up-Prozesse und top-down-Prozesse. Mit bottom-up-Prozessen ist das gemeint, "was als Input von draußen reinkommt", und mit top-down-Prozessen, alles, was "dem Einlaufenden formend, aufnehmend oder auch ablehnend entgegentritt“ (Hörmann, 1980: 18). Die beiden Prozesse wirken zusammen, ergänzen und bedingen einander. Erst in ihrer strategischen Interaktion kann Verständnis entstehen. In diesem Sinne postuliert Danica Seleskovitch (1984: 40) für den Verstehensprozess unter dolmetschspezifischen Gesichtspunkten den „Brückenschlag zwischen bereits vorhandenem Wissen und dem Wortlaut der Aussage“. Dieses Postulat machen wir auch für das Übersetzen geltend. In Anlehnung an Seleskovitch (1980: 155) lässt sich der Verstehensvorgang im Zusammenspiel von Sprachsignal und kognitiver Ergänzung wie folgt darstellen: - richtiges Sprachsignal vom Sprecher emittiert + richtige kognitive Ergänzung = richtiges Verständnis TRANSLATION AND MEANING, PART 10, 2013, 221 - 230 - 223 richtiges Sprachsignal + falsche kognitive Ergänzung = falsches Verständnis falsches Sprachsignal + richtige kognitive Ergänzung = richtiges Verständnis. Es wird ersichtlich, dass für das erfolgreiche Verstehen allein die richtige kognitive Ergänzung, also die Verbindung mit den richtigen kognitiven Strukturen im mentalen System des Translators, von entscheidendem Belang ist. Im Sinne des Festgestellten kann für die Auffassung der rezeptiven translationsrelevanten Prozesse angenommen werden, dass das Verstehen des Translators ein durch vielerlei „Anweisungen” determinierter Vektor seines kognitiven Systems ist. Dieses System weist eine mehrschichtige mentale Struktur auf, die solche Verarbeitungsebenen umfasst wie die Ebene der Sensomotorik, der Syntax, der Semantik und der Pragmatik (Strohner, 2005: 191). Ausgehend von den genannten Wissensebenen des kognitiven Systems können nach Hans Strohner (2005: 193f.) folgende Teilprozesse unterschieden werden: - sensomotorische Prozesse – Prozesse der graphemischen und phonologischen Informationsverarbeitung sowie Verarbeitung nonverbaler und graphischer Informationen, syntaktische Prozesse – Prozesse, denen primär Ordnungs- und Vermittlungsfunktionen innerhalb und zwischen den anderen drei Prozessbereichen zufallen, semantische Prozesse – Prozesse, deren Ziel in der Zuordnung von Bedeutung zum übermittelten Text liegt, pragmatische Prozesse – Prozesse, die die Brücke zum Kommunikationssystem und damit den funktionalen Bezugsrahmen für alle anderen Prozesse des Textverstehens bilden. Die bestimmende Rolle spielen im Textverstehen die semantischen Prozesse, darunter Teilprozesse der Referenz, der Kohärenz und der Inferenz. 2.2.1 Referenzprozesse Die Funktion der Referenzprozesse besteht in der Rekonstruktion des vom Textproduzenten intendierten Referenzbereichs. Dies kommt infolge der Aktivierung mentaler Konzepte in Abhängigkeit von den semantischen Verarbeitungsebenen und dem pragmatischen Wissen zustande (Strohner, 2005: 194). Die Referenz als Konstrukt der kognitiven Operationen ist somit keine statische, im vornherein festgelegte Größe, sondern „eine aktuell-spontane Relation zwischen Sprache und Situation“ (Kupsch-Losereit, 1995: 186). Im Lichte des Erörterten soll für das Verstehen des Translators festgehalten werden, dass sein Gegenstand nicht allein der A-Text bildet, sondern der von dem initialen Sender mit dem A-Text angestrebte Referenzbereich. Verarbeitet wird also vom Translator nicht nur das Gesagte, sondern auch das Gemeinte. Bei der Rekonstruktion des vom initialen Sender intendierten Referenzbereiches muss der Translator alle das Translationsgefüge konstituierenden Faktoren berücksichtigen, wie z.B. den soziokulturellen Situationshintergrund im A-Text, Vorwissen und Erwartungen seines Produzenten, das kommunikative Ziel, das der initiale Sender erreichen will. Beim Dolmetschen ist darüber hinaus die richtige Interpretation der Intonation, Gestik und Mimik des Sprechenden wichtig sowie das Verhalten der übrigen Teilnehmer. 2.2.2 Kohärenzprozesse Das Ziel der Kohärenzprozesse ist die Integration der einzelnen Konzepte eines Textes zu einer semantischen Gesamtrepräsentation (Strohner, 2005: 195). An diesen Prozessen sind sowohl das Langzeitgedächtnis als auch das Arbeitsgedächtnis beteiligt. Die Kohärenzherstellung kommt, wie Walter Kintsch und Teun van Dijk (1978) behaupten, infolge der zyklischen Textverarbeitung auf mehrerer Ebenen zustande. In ihrem Modell postulieren Kintsch und van Dijk folgende Ebenen der Textverarbeitung: Ebene der atomaren Propositionen: semantische Grundeinheiten, Wörter; Ebene der komplexen Propositionen: Teilsätze; Ebene der lokalen Kohärenz: Satzverbindungen; 224 ANNA MAŁGORZEWICZ Ebene der Makrostruktur: aus den komplexen Propositionen inferierte allgemeine Aussagen, die zusammen die Makrostruktur des Textes bilden; Ebene der Superstruktur: konventionalisierte Formen von Textsorten (Rickheit/Strohner, 1993: 77). Die Propositionen als elementare Bedeutungseinheiten verdeutlichen die Relationen zwischen den Konzepten und bilden die sog. Textbasis, die von van Dijk und Kintsch (1983: 11) als die semantische Repräsentation des Inputs im episodischen Gedächtnis definiert wird. Gleichzeitig mit dem Aufbau der Textbasis wird ein durch Wissen, Präsuppositionen und Motive beeinflusstes Situationsmodell (die kognitive Repräsentation der Situation, die der Text beschreibt) im episodischen Gedächtnis aktiviert, mit dem die Textbasis laufend verglichen und aktualisiert wird. Es ist somit anzunehmen, dass der Translator nicht wartet, bis er alle Daten aufgenommen hat, sondern dass er schon beim ersten rezipierten Wort mit der Kohärenzherstellung beginnt. Bei jedem wieteren Input ergänzt er die semantische Repräsentation des Ausgangstextes, die Textbasis also, die im Verarbeitungsprozess nicht konstant bleibt. Sie wird ständig modifiziert und an die aus dem rezipierten Text einlaufenden Informationen aktuell angepasst. 2.2.3 Inferenzprozesse Referenz und Kohärenz werden mithilfe von Inferenzprozessen hergestellt. Ihr Ziel ist die Verknüpfung der durch den Text explizit angesprochenen Konzepte mit dem Weltwissen (Strohner, 2005: 197). Strohner (2005: 198) unterscheidet zwei Arten von semantischen Inferenzprozessen: obligatorische Inferenzen, die zur Lösung der Kohärenz- und Referenzprobleme dienen, und optionale Inferenzen, die auf Grund von weiter gehenden Weltwissensvorräten zu Stande kommen. Es handelt sich dabei um Informationen hinsichtlich von Raum, Zeit, Ursachen, Wirkungen, beteiligten Personen und Objekten der dargestellten Ereignisse. Während die obligatorischen Inferenzen beinahe automatisch gezogen werden, werden die optionalen Inferenzen vor allem dann gebildet, wenn „besondere Verstehensinteressen dies verlangen“ (Strohner, 2005: 198). Die besonderen Verstehensinteressen treten im Rahmen der interkulturellen Kommunikation ohne jeden Zweifel auf, demgemäß sollte dieser Inferenzart eine ausschlaggebende Bedeutung bei Verstehensvorgängen beigemessen und ihnen auch in der translatorischen Bildung genügend Raum eingeräumt werden. In Anlehnung an das von Strohner dargestellte Modell des Textverstehens lassen sich die Verstehensprozesse des Translators wie folgt illustrieren: TRANSLATION AND MEANING, PART 10, 2013, 221 - 230 225 Translationssituation Modell des initialen Senders Pragmatik A-Kultur Semantik ? Referenzbereich Konzept 1 A-Text Konzept 2 Syntax Sensomotorik Abb.(2): Verstehensprozesse des Translators Die dargestellten Verstehensprozesse des Translators – des mittelbaren Empfängers – sind zweigegliedert: in die Konstruktion der Textbasis und die Aktualisierung eines Situationsmodells als Referenzbereich. Der Translator verarbeitet in seinem kognitiven System die in der A-Kultur verankerten A-Textinformationen. Diese Verarbeitung geschieht auf der Ebene der Sensomotorik, Syntax, Semantik und Pragmatik. Auf den Ebenen des Wortes, des Satzes und des Textes werden vom Translator wissensbasierte Inferenzen eingesetzt. Bei Inferenzen wirken die aktivierten Ebenen verschiedentlich aufeinander ein. Die Verstehensprozesse des mittelbaren Empfängers zielen in der Translation auf die Herstellung des vom initialen Sender intendierten Referenzbereichs ab. Der A-Text stellt für den Translator lediglich einen Ausgangspunkt für seine Schlussprozesse dar. Der aufgebaute Referenzbereich ist im Vergleich zum Text viel komplexer. Die ihn konstruierenden Informationen lösen mehr Konnotationen und Assoziationen aus, als im Text verarbeitet werden können. Wenn wir das Festgestellte auf die Verstehensphase des finalen Empfängers beziehen, ergeben sich für den mittelbaren Sender weit reichende Konsequenzen. Er wird vor die Aufgabe gestellt, einen Text zu produzieren, der den von ihm infolge seiner Verstehensprozesse hergestellten Referenzbereich im mentalen System des finalen Empfängers evoziert. Gegenstand der sich im Kopf des mittelbaren Senders vollziehenden mentalen Prozesse sind demzufolge einerseits der in der A-Kultur aufgebaute Referenzbereich und andererseits die Strategien zum Aktivieren desselben unter Berücksichtigung der B-Kultur. Es muss dabei angenommen werden, dass die A-Sprache und die A-Kultur im Vergleich zu der B-Sprache und B-Kultur divergente Zugangsweisen zu dem translationsrelevaten Referenzbereich aufweisen. Unterschiedlich sind auch die Wirklichkeitsmodelle sowie Erfahrungswelten des initialen Senders und des finalen Empfängers. Der Translator vermittelt somit zwischen zwei unterschiedlichen kultur- und sprachlich geprägten Referenzsystemen. Damit eine Kommunikation im Rahmen des Translationsgefüges zustande kommen kann, muss der Translator die kulturgeprägten Inhalte des vom initialen Sender intendierten Referenzbereichs den Transformationen unterziehen, die es ermöglichen, die zu übermittelnden Informationen mit der Zielkultur kompatibel zu machen. 226 ANNA MAŁGORZEWICZ Eine gravierende Rolle spielt bei diesen Prozessen das interkulturelle oder aber transkulturelle1 Bewusstsein des Translators. Operationen, die in Abhängigkeit vom transkulturellen Bewusstsein vollzogen werden, lassen sich anhand der von Fillmore entwickelten scene-and-frame-Semantik beschreiben. 2.3 Translationsrelevante kognitive Prozesse vor dem Hintergrund der scene-and-frameSemantik Das Fillmore-Modell (1977) wurde in die translationwissenschaftliche Diskussion durch Mia Vannerem und Mary Snell-Hornby (1986) eingeführt und anschließend auch von anderen Translatologen als Grundlage für Erklärungsversuche der Interpretationsprozesse innerhalb von Übersetzungsprozessen eingesetzt. Frame wird von Fillmore für jedes System der sprachlichen Möglichkeiten benutzt, das mit prototypischen Szenen in Verbindung gebracht werden kann, und scene für jede Art von zusammenhängenden Segmenten menschlicher Überzeugungen, Erfahrungen und Vorstellungen (Hönig, 1998: 330). Der Sprachbenutzer findet im Kommunikations- und Verstehensvorgang zu jeder linguistischen Form (frame) zunächst Zugang mittels eigener Erfahrung bzw. einer Situation, die für ihn persönlich von Bedeutung ist (scene). Dies ist ein dynamischer Prozess: „Scenes und frames aktivieren einander wechselseitig und in unterschiedlicher Komplexität, das heißt, eine bestimmte sprachliche Form ruft Assoziationen hervor, diese wiederum aktivieren andere linguistische Formen, bzw. erwecken weitere Assoziationen durch szenische Inferenzen.“ (Stolze, 2003: 98). Scenes, unsere Wissensbestände und Erfahrungen also, formieren sich als Erwartungsstrukturen in Bezug auf die Welt und Texte, die wir wahrnehmen. Einerseits ermöglichen sie die Identifizierung und Interpretation der Objekte und Ereignisse unserer Umwelt, und andererseits modellieren sie unsere Wahrnehmungen. Angesichts des Gesagten lässt sich für die Translation die Forderung formulieren, dass der die Rolle des mittelbaren Senders ausführende Translator entsprechende Methoden der adäquaten scenes-Überführung in die B-Kultur einsetzen muss, um das Ziel der interkulturellen Kommunikation erfolgreich zu erreichen. Der Translator baut auf der Basis der mittels des A-Textes wahrgenommenen ausgangssprachlichen frames die von ihnen intendierten scenes (mentalen Bilder) auf, die die Gesamtszene des A-Textes bilden. Er muss also nicht nur ausgangssprachliche Zeichen richtig identifizieren, sondern sie im Kontext der ausgangskulturellen Faktoren, auf Grund seines Wissens, seiner Erfahrungen, vor dem Hintergrund der Erwartungen und Intentionen des initialen Senders und unter Einbeziehung der anderen frames interpretieren. So kommen in dieser Verstehensphase Inferenzstrategien im Zusammenspiel mit allen Ebenen der Sprachverarbeitung zum Einsatz. Dabei muss der Translator darauf sensibilisiert sein, bei der Bildung der ausgangskulturellen scenes die vom Sender des A-Textes intendierten Konzepte zu aktivieren. Diese Übereinstimmung ist nur im Falle einer vollkommen ausgebildeten Ausgangssprache und ausgangskulturellen Kompetenz des Translators und dessen Vermögens möglich, die ausgangssprachlichen Konzeptualisierungsverfahren nachzuvollziehen. Eine weitere Etappe bilden die mentalen Prozesse im kognitiven System des mittelbaren Senders, die auf die Überführung der ausgangskulturellen scenes in die zielkulturellen scenes abzielen. Die vom sprachlichen Gewand gelösten Ausgangs-scenes werden in die von dem finalen Empfänger wahrnehmbaren Ziel-scenes transformiert. Ihr Resultat sind die zielsprachlichen frames, die 1 Frau Prof. Barbara Lewandowska-Tomaszczyk hat in ihrem Vortrag auf der Konferenz der Polnischen Gesellschaft für Angewandte Linguistik im April 2010 postuliert, das transkulturelle Bewusstsein des Translators in die translationswissenschaftliche Diskussion einzubeziehen. TRANSLATION AND MEANING, PART 10, 2013, 221 - 230 227 wiederum im mentalen System des finalen frames-Empfängers Konzepte aktiviert, welche die vom Translator angestrebten scenes hervorrufen sollen. Der Translator soll bei der Formulierung des BTextes solche linguistischen Formen auswählen, die Konzepte aktivieren, die der finale Empfänger in seine kognitiven Strukturen integrieren kann. Dies ist die Voraussetzung für den Aufbau äquivalenter Szenen, die in die mentale Weltrepräsentation des finalen Empfängers erfolgreich übertragen werden können, und somit die Bedingung für gelungene interkulturelle Kommunikation. Das um Aspekte der frames- und scenes-Bildung erweitere Modell des Translationsgefüges kann folgendermaßen dargestellt werden: A-Kultur B-Kultur Ascene initialer Sender Aframe A1scene Bscene mittelbarer Empfänger mittelbarer Sender B1scene Bframe finaler Empfänger Translator Translationssituation Abb.(3): Frames- und scenes-Verarbeitung im Translationsgefüge 3. Resümee und Ausblick Die beiden eingesetzten Modelle – die psycholinguistische Verstehenstheorie von Strohner und die scene-and-frame-Semantik von Fillmore – erlaubten es, die Komplexität der mentalen Vorgänge im Kopf des Translators zu schildern. Die Sprachverarbeitung beschränkt sich nicht nur auf verbale Informationen, sondern benötigt auch außersprachliches Situations- und Kulturwissen. Die genutzten Ansätze ließen darüber hinaus darauf schließen, dass bei der Translation nicht der A-Text in den B-Text transkodiert wird, sondern dass nach einer annähernden semantischen Ähnlichkeit dieser Texte (vgl. dazu Järventausta, 1998: 232) gestrebt wird. Zum Gegenstand der translationsrelevanten mentalen Prozesse wird die Herstellung eines vom initialen Sender intendierten Referenzbereichs, der im Einklang mit den im kognitiven System des finalen Empfängers hervorgerufenen Szenen steht. Der A-Text und der B-Text liefern Orientierungspunkte, die „in einem vordefinierten, meist schon vorhandenen epistemischen Raum situiert werden“ (Busse, 1994: 232). Auf Grund der bewussten Entscheidungen formuliert der Translator den B-Text, der Orientierungspunkte im kognitiven System des finalen Empfängers wachruft, die wiederum die Bedeutung der Äußerung durch Bezugsetzung zum Wissen des finalen Verstehenden konstruieren (vgl. Busse, ebd.). Ausschlaggebend für die Entscheidungen des Translators ist das von ihm definierte Ziel der Translation. Wie im Modell von Grucza postuliert wurde, ist dieses der funktionalen Äquivalenz untergeordnet. Dabei soll jedoch festgehalten werden, dass die Intentions- und Funktionskonstanz des A-Textes und B-Textes im Translationsprozess nicht angenommen werden kann. Wir schließen uns dem funktionalen Para- 228 ANNA MAŁGORZEWICZ digma der Translationsforschung an und vertreten dementsprechend die Ansicht, dass für die Ermittlung des Translationsziels neben der Analyse der funktionalen Einheiten im A-Text auch die Interpretation des „Übersetzungsauftrags“ durch den Translator eine gravierende Rolle spielt (vgl. dazu Nord, 2002: 17). Den nächsten Schritt stellt nach Christiane Nord die Entscheidung zwischen dem „Soll“ des Translationsauftrags und dem „Ist“ des vorhandenen Textes dar. Das Zusammenspiel textexterner und textinterner Faktoren wird von Nord (1991: 41, 2002: 49) durch die sogenannten WFragen auf eine griffige Formel gebracht: Wer übermittelt wozu wem über welches Medium wo wann warum einen Text mit welcher Funktion? Worüber sagt er was (was nicht) in welcher Reihenfolge unter Einsatz welcher nonverbalen Elemente in welchen Worten in was für Sätzen in welchem Ton mit welcher Wirkung? Im Lichte des Gesagten kann die Translation, Lew Zybatow (2004: 63) zustimmend, als ein offenes Verfahren definiert werden, in dem nicht der A-Text, sondern die Interpretation der Anhaltspunkt für die Einzelentscheidungen des translatorisch Handelnden ist. Für die Ausführung der Translationsaufgabe ist die entsprechend entwickelte metakognitive Kompetenz unabdingbar. Sie stellt den von Franciszek Grucza (1985: 34, 35) und Jerzy Żmudzki postulierten wesentlichen Mehrwert dar. […] ein jeder erfahrene Translator besitzt zum einen eine Menge von selbst erzeugten spezifischen Regeln, Fertigkeiten und Kenntnissen, die diesen bestimmten Mehrwert bzw. Überschuss seiner Kenntnisse darstellen. Zum anderen verfügt er über einen bestimmten Mehrwert bzw. Überschuss von Wissen als seine eigene individuelle Translationstheorie gegenüber einer jeden anderen monolingualen Kommunikationskompetenz mit prätranslatorisch-bedingendem Charakter (Żmudzki, 2010: 119). Die metakognitive Kompetenz findet ihren Ausdruck in der Problemlösungsfähigkeit. Vom Translator wird also erwartet, dass er Problemlösungsverfahren einsetzt, die nach Wolfram Wilss erkennen lassen, „dass er seinen Navigationspfad unter dem Gesichtspunkt einer akzeptablen Input/OutputRelation zu finden vermag und dass er entscheidungskompetent handeln kann (Wilss, 1991: 136).“ Die optimal entwickelten Problemlösungsstrategien, aber auch die sie unterstützende Kontroll- und Selbstkorrekturfähigkeit sind Voraussetzung für eine erfolgreiche Handlung im Translationsgefüge. Die metakognitiven Strategien des Translators erlauben es, den finalen Empfänger mit dem B-Text entsprechend zu beeinflussen, d.h. dem Translationsziel angemessen zu orientieren. In diesem Sinne TRANSLATION AND MEANING, PART 10, 2013, 221 - 230 229 trägt der Translator die Verantwortung für die erfolgreiche Orientierung2 des finalen Empfängers, die jedoch nur unter der Voraussetzung gewährleistet werden kann, dass die Kohärenzregel3 der Translation eingehalten wird, also wenn die Translation vom Zieltextrezipienten kohärent mit seiner Situation interpretiert werden kann. Diese prognostische Fähigkeit setzt das Vorhandensein eines umfassend und vielwertig strukturierten Sprach- und Kulturbewusstseins sowie Selbstbewusstseins, mithin also der metakognitiven Kompetenz voraus. Literatur Barbosa, H. G. / Neiva, A. M. S. (2003), “Using think-aloud protocols to investigate the translation process of foreign language learners and experienced translators”. In: Alves, F. (Hrsg.), Triangulating Translation. Perspektives in process oriented research. Amsterdam / Philadelphia: Benjamins, 137-155. Busse, D. (1994), „Kommunikationsmodelle und das Problem des Sprachverstehens. Zur Verwendung technischer Metaphern in der Sprachwissenschaft“. Hoberg, R. (Hrsg.), Technik in Sprache und Literatur. (Festschrift für Franz Hebel). Darmstadt: Verlag der TH Darmstadt, 207 - 234. Dijk, T., A., van / Kintsch, W. (1983), Strategies of Discourse Comprehension. New York: Academic Press. Fillmore, Ch. J. (1977), “Frames and Scenes Semantics”. In: Zampolli, A. (Hrsg.), Linguistic Structures Processing. Amsterdam: North-Holland, 55-81. Grucza, F. (1981), „Zagadnienia translatoryki”. In: Grucza, F. (Hrsg.), Glottodydaktyka a translatoryka. (Materiały z IV sympozjum zorganizowanego przez Instytut Lingwistyki Stosowanej UW. Jachranka 3-5 listopada 1976r.). Warszawa: Wydawnictwo Uniwersytetu Warszawskiego, 9-29. Grucza, F. (1985), „Lingwistyka, lingwistyka stosowana, glottodydaktyka, translatoryka”. In: Grucza, F. (Hrsg.): Lingwistyka, glottodydaktyka, translatoryka. (Materiały z VIII Sympozjum zorganizowanego przez Instytut Lingwistyki Stosowanej UW. Jadwisin, 5-7 listopada 1982r.). Warszawa: Wydawnictwo Uniwersytetu Warszawskiego, 19-44. Grucza, F. (1998), „Wyodrębnienie się, stan aktualny i perspektywy świata translacji oraz translatoryk”i. In: Lingua legis, Nr. 6, 2-12. Hönig, H. (1998), Sind Dolmetscher bessere Übersetzer? In: Jahrbuch Deutsch als Fremdsprache 24, 323343. Hörmann, H. (1980), „Der Vorgang des Verstehens“. In: Kühlwein, P. / Raasch, A. (Hrsg.), Sprache und Verstehen: Kongressberichte der 10. Jahrestagung der Gesellschaft für Angewandte Linguistik GAL e.V., Band 1. Tübingen: Narr, 17-29. Jakobsen, A. L. (2005), “Instances of Peak Performance in Translation”. In: Lebende Sprachen 50.3, 111-116. Jääskeläinen, R. (1989), “Translation Assignment in Professional vs. Non-Professional Translation: A ThinkAloud Protocol Study”. In: Seguinot, C. (Hrsg.), The Translation Process. Toronto: H.G. Publications, 8798. Jääskeläinen, R. (1996), “Hard work will bear beautiful fruit? A comparison of two think-aloud protocol studies”. In: Meta 41.1, 60-74. Jääskeläinen, R. (1999), Tapping the Process: An Explorative Study of the Cognitive and Affective Factors Involved in Translating. Joensuu: Joensuun yliopisto. Järventausta, M. (1998), „Scenes & Frames & Übersetzen“. In: Jahrbuch Deutsch als Fremdsprache 24, 217234. Jensen, A. (2001), The effects of time on cognitive processes and strategies in translation. Copenhagen Business School, Faculty of Modern Languages: Copenhagen Working Papers in LSP. Kielar, B. Z. (2003), Zarys translatoryki. Warszawa: Katedra Języków Specjalistycznych Uniwersytetu Warszawskiego. 2 Mit Aspekten der sozialen Interaktion im Verstehensprozess unter Berücksichtigung der Kognition von Kommunikationspartnern setzt sich der Ansatz der konstruktivistischen Theorie des Verstehens von Gebhard Rusch (1992) auseinander. Vor dem Hintergrund dieses Modells lassen sich die mentalen Prozesse sowie Entscheidungen des Translators in gegenseitiger Abhängigkeit beschreiben. Auf Grund des konstruktivistischen Ansatzes der Verstehenstheorie lässt sich auch ein didaktisches Modells der Translation entwickeln (vgl. dazu Małgorzewicz, 2010). 3 Es handelt sich um die Kohärenzregel im Sinne der funktionalen Skopostheorie. Kohärent bedeutet dabei ein „Glaube-verstanden-zu-haben“, ein „ist-für-Empfänger-interpretierbar“ (Vermeer, 1978:101). 230 ANNA MAŁGORZEWICZ Krings, H. P. (1987), „Der Übersetzungsprozess bei Berufsübersetzern – Eine Fallstudie“. In: Arntz, R. (Hrsg.), Textlinguistik und Fachsprache. Akten des Internationalen übersetzungswissenschaftlichen AILASymposiums Hildesheim 13.-14.4.1987. Hildesheim: Olms. Kupsch-Losereit, S. (1995), „Die Modellierung von Verstehensprozessen und die Konsequenzen für den Unterricht“. In: Text con Text 10, 179-196. Kußmaul, P. (2009), „Können wir besser übersetzen, wenn wir wissen, wie wir denken?“ In: Ahrens, B. / Černý, L. / Krein-Kähle, M. / Schreiber, M. (Hrsg.), Translationswissenschaftliches Kolloquium I. Beiträge zur Übersetzungs- und Dolmetschwissenschaft (Köln/Germersheim). Frankfurt am Main: Peter Lang, 515330. Lewandowska-Tomaszczyk, B. (2010), Nowe spojrzenie na przekład: interakcyjne modele kognitywne i korpusy językowe. Vortrag auf der Konferenz der Polnischen Gesellschaft für Angewandte Linguistik, Warszawa, den 9.04. 2010. Małgorzewicz, A. (2010), „Auf dem Weg zu einem Modell der Translationsdidaktik“. In: Glottodidactica XXXVI, 21-32. Nord, Ch. (1991), Textanalyse und Übersetzen: theoretische Grundlagen, Methode und didaktische Anwendung einer übersetzungsrelevanten Textanalyse, Heidelberg: Julius Groos. Nord, Ch. (2002), Fertigkeit Übersetzen. Alicante: Editorial Club Universitario. Opitz, C. (2004), „Zum Aufschlusswert prozesswissenschaftlicher Untersuchungen für Übersetzungsdidaktik“. In: Fleischmann, E. / Schmitt, P.A. / Wotjak, G. (Hrsg.), Translationskompetenz. Tagungsberichte der LICTRA (Leipzig Intenational Conference on Translation Studies 4. – 6.10.2002). Tübingen: Stauffenburg, 659-677. Rickheit, G. / Strohner, H. (1993), Grundlagen der kognitiven Sprachverarbeitung: Modelle, Methoden, Ergebnisse. Tübingen: Francke. Rusch, G. (1992), „Auffassen, Begreifen und Verstehen. Neue Überlegungen zu einer konstruktivistischen Theorie des Verstehens“. In: Schmidt, S. J. (Hrsg.), Kognition und Gesllschaft. Der Diskurs des Radikalen Konstruktivismus 2. Frankfurt am Main: Suhrkamp, 214-255. Seleskovitch, D. (1980), „Beitrag der Sprachmittlung zur Untersuchung des Verstehens“. In: Kühlwein, P. / Raasch, A. (Hrsg.), Sprache und Verstehen: Kongressberichte der 10. Jahrestagung der Gesellschaft für Angewandte Linguistik GAL e.V., Band 1. Tübingen: Narr, 155-160. Seleskovitch, D. (1984), „Zur Theorie des Dolmetschens“. In: Kapp, V. (Hrsg.), Übersetzer und Dolmetscher: theoretische Grundlagen, Ausbildung, Berufspraxis. München: Francke, 37-50. Stolze, R. (2003), Hermeneutik und Translation. Tübingen: Gunter Narr. Strohner, H. (2005), „Textverstehen aus psycholinguistischer Sicht“. In: Blüdorn, H. / Breindl, E. / Waßner, U. H. (Hrsg.), Text-Verstehen. Grammatik und darüber hinaus. Berlin und New York: Walter de Gruyter, 187-204. Vannerem, M. / Snell-Hornby, M. (1986), ”Die Szene hinter dem Text: ‘scenes-and-frames-semantics’ in der Übersetzung“. In: Snell-Hornby, M. (Hrsg.), Übersetzungswissenschaft - Eine Neuorientierung zur Integrierung von Theorie und Praxis. Tübingen: Francke Verlag, 194-205. Vermeer, H. J. (1978), „Ein Rahmen für allgemeine Translationstheorie“. In: Lebende Sprachen 23, 99-102. Wilss, W. (1991), „Kognitive Aspekte des Übersetzungsprozesses“. In: Schmitt, Ch. (Hrsg.), Neue Methoden der Sprachmittlung. Wilhelmsfeld: Egert, 121-148. Zybatow, L. (2004), „Methodologie der Translationswissenschaft und Methoden der Translation“. In: Hous, J. / Koller, W. / Schubert, K. (Hrsg.), Neue Perspektiven in der Übersetzungs- und Dolmetschwissenschaft. Festschrift für Heidrun Gerzymisch-Arbogast zum 60. Geburtstag. Bochum: AKS-Verlag, 47-72. Żmudzki, J. (2010), „Aktuelle Profile der germanistischen Translationsdidaktik in Polen“. In: Małgorzewicz, A. (Hrsg.), Translation: Theorie – Praxis – Didaktik. Dresden – Wrocław: Neisse Verlag / Oficyna Wydawnicza ATUT – Wrocławskie Wydawnictwo Oświatowe, 117-136. TRANSLATION AND MEANING, PART 10, 2013, 231 - 238 231 ZUR WAHL DER TEXTE FÜR DIE BILDUNG VON ÜBERSETZERN Marcin Michoń University of Łódź, Poland Abstract: Eins der Prinzipien für die Wahl der Texte im Bildungsprozess von Übersetzern und Dolmetschern heißt, dass der Textfächer die Berufspraxis des jeweiligen Sprachpaars abbilden sollte und somit den pragmatischen Hintergrund für die gestellten Aufgaben sichern (vgl. dazu Kußmaul 2006: 358f). Dennoch arbeiten die polnischen Bildungsinstitutionen in der Anfangsphase des Übersetzungskurses oft an Texten, die in der Übersetzertätigkeit nur ausnahmsweise zum Objekt eines Auftrags werden. Gemeint sind u.a. unterschiedliche Pressetexte1, die semesterlang in den Seminaren dem Übersetzernachwuchs als Ausgangstexte dargeboten werden (vgl. dazu Sieradzka 2010: 461f). In der Bildung der Germanisten liegt in der ersten Phase des Übersetzungskurses das primäre Ziel in der Progression sprachlicher Kompetenzen, jedoch tragen die Seminarleiter auch die Verantwortung dafür, welche Techniken und Prinzipien des Übersetzens (und Dolmetschens) den Studenten eingeprägt werden. In diesem Zusammenhang überlegen wir folgende Fragen: Sollten Texte, die außerhalb der Berufspraxis stehen im Bildungsprozess überhaupt eingesetzt werden? Welche Vorteile kann didaktische Arbeit an solchen Texten bringen? Welche Kompetenzen entwickelt und welches Wissen vermittelt die Übung an solchen Texten überhaupt? 1. Der Text in der Übersetzungswissenschaft Die Sprachwissenschaft hat sich bisher nicht geeinigt, welche Textdefinition für die breite Spanne der textlinguistischen Forschung hin zur Diskurslinguistik grundlegend ist. Die neuesten Untersuchungen weisen weiterhin auf die Textauffassung von Dressler/de Beaugrande (1981), auf das Texttualitätskonzept hin, und sie könnte als passender Grundstein für die translatorische Forschung betrachtet werden (vgl. dazu Fix, 2006: 18ff.; Dressler/de Beaugrande, 1981). Diese Auffassung versteht den Text als eine kommunikative Okkurrenz und die Textualitätskriterien2 konzentrieren sich auf den Austausch sprachlich codierter Inhalte zwischen dem Sender (Emittenten) und dem Rezipienten. Das kommunikative Übersetzungsmodel sieht ebenfalls das mündliche und schriftliche Übersetzen als „interlingualen Kommunikationsvorgang, wobei jeweils zwei alternierende Endkodierungs- und Dekodierungsvorgänge erforderlich sind“ (Kade, 1968: 48). Das Übersetzen besteht demnach aus drei Phasen: Der Kommunikation zwischen dem Autor (Sender) und dem Übersetzer, der zum vorläufigen Empfänger des Ausgangstextes wird; der Umkodierung (diese Phase vollzieht sich im Übersetzer) und der kommunikativen Okkurrenz zwischen dem Übersetzer (S') und dem Rezipienten (E') (vgl. dazu Koller, 1997: 106). Bei der schriftlichen Übersetzung besteht die Umkodierung aus drei Phasen: Analyse, Transfer und Synthese. Die drei Schritte erfordern vom Übersetzer 1 Texte der Presseagenturen werden alle möglichen Weltsprachen übersetzt, sobald darin etwas vom internationalen Belang berichtet wird (vgl. dazu Gawlas, 2006: 236f). Auf dem polnischem Pressemarkt sind zudem Zeitschriften zu finden, die hauptsächlich übersetzte Texte aus der fremdsprachigen Presse beinhalten. Es handelt sich dabei um Vertreter der Motorisierungspresse (z.B. „Auto Świat“, „Auto Motor i Sport“ mit den deutschen Vorlagen „Auto Bild“ und „Auto Motor und Sport“) sowie die Zeitschrift „Forum“ mit übersetzten populärwissenschaftlichen aus der fremdsprachigen Presse. Darüber hinaus vermittelt die polnische Fachpresse übersetze Texte, die an Vertreter der jeweiligen Berufsgruppen gerichtet sind (z.B. Texte der Kosmetikhersteller an Frisöre oder der Bauchemiehersteller an die Ausführungsfirmen auf dem polnischen Baumarkt). 2 Der Text ist „eine kommunikative Okkurrenz“ (engl ‚occurrence’), die sieben Kriterien der Textualität erfüllt. Wenn irgendeines dieser Kriterien als nicht erfüllt betrachtet wird, so gilt der Text nicht als kommunikativ. Daher werden nicht kommunikative Texte als Nicht-Texte behandelt“ (Dressler/de Beaugrande, 1981: 3). Die Textualitätskriterien sind die Kohärenz, Kohäsion, Intentionalität, Akzeptabilität, Situationalität, Informativität und Intertextualität. Der Übersetzer muss die Textualitätskriterien in allen drei Phasen des Übersetzungsprozesses berücksichtigen, und somit dem Skopos des Translats nachzugehen. 232 MARCIN MICHOŃ textuelle Kompetenzen, daher sind textlinguistische Probleme in der Übersetzungs- und Dolmetscherdidaktik zu berücksichtigen (vgl. dazu Nida/Taber, 1964: 68). 2. Intertextuelles Wissen als Kompetenz des Übersetzers Die Translationswissenschaft profitiert von den Erkenntnissen der Textlinguistik, vor allem hinsichtlich der normativen Erfassung vieler schriftlicher oder mündlicher Kommunikationsvorgänge als Textsorten. Diese Richtlichtlinien erleichtern sowohl ihre Produktion als auch Rezeption. In der Gesellschaft funktionieren diese Muster als Werkzeug zwecks Erfüllung bestimmter Absichten. Dressler/de Beaugrande (1981) fassen die Zugehörigkeit eines Textes zu einer Textsorte innerhalb des textuellen Kriteriums der Intertextualität auf. Die Textsorte ist dabei als Gruppe von Texten zu verstehen, die das jeweilige Textmuster abbilden. Die Tatsache, dass das Textsortenwissen dem Übersetzer bei der Rezeption der kodierten Botschaft hilft, und dass diese Kompetenz ihm ermöglicht, die gegebenen Eigenschaften des Ausgangstextes (AT) bei der Produktion des Zielsprachentextes (ZT) zu berücksichtigen, berechtigt die Frage: In welchem Grade ist die Textkompetenz für den Übersetzer notwendig? Kann diese Kompetenz im Rahmen eines Übersetzungskurses vermittelt und geübt werden oder muss sie als kognitive Erfahrung aus den wissenschaftlichen und alltäglichen Gebieten entwickelt werden? Die Berücksichtigung der Textualitätskriterien im Hinblick auf die Textanalyse und Textreproduktion ist ein wesentlicher Bestandteil des Übersetzungsprozesses. Natürlich sollten sich die Übersetzer das kognitive Wissen über Texte als Teil des breitverstandenen Weltwissens aneignen. Die meisten Studenten zeigen sich im Übersetzungskurs in vieler Hinsicht als unerfahren auch in Bezug auf die Textrezeption und -produktion. Daher wird im Rahmen des Übersetzungskurses oft ein textlinguistisches Seminar mit dem Fokus auf die Textgruppen und Textsorten angeboten, die für die Berufspraxis des jeweiligen Sprachpaars typisch sind. Da manche Fachtexte, z. B. Urkunden, Datenblättern, Gebrauchsanleitungen, Packungsbeilagen usw. aufgrund ihrer Struktur viele Voraussetzungen der Übersetzerwerkstatt nicht explizieren lassen, können als Ergänzung Texte dienen, die seltener in der beruflichen Praxis zum Objekt eines Übersetzungsauftrags werden. Gemeint sind z.B. Pressetexte. 3. Anwendung der Pressetexte im Übersetzungskurs 3.1 Zielgruppe Um feststellen zu können, welche Prinzipien der Übersetzung an Pressetexten erläutert werden können, wurde eine Fallstudie durchgeführt, bei der die Probandengruppe polnische Studenten bilden, deren Sprachkenntnisse eine Spanne vom B1 bis über zum C2 Niveau erstrecken. Es handelt sich dabei um Studenten, die erst ein BA-Studium absolvieren, und solche, die den Magistertitel und das Lehramt Deutsch bereits erworben haben und sich an einem Postgraduiertenstudium für Dolmetscher und Übersetzer weiterbilden. 3.2 Ziel der Studie Die Arbeit am gewählten Pressetext ist als Einstieg in den Übersetzungsworkshop gedacht, denn an dem ZT soll auf 1) das Wesen der Übersetzung, 2) Probleme des interkulturellen Transfers, 3) die Methoden der Übersetzung, 4) die Typen der Äquivalenz, 5) die möglichen Wege zu Problemlösungen, 6) die Fehlerklassen bzw. -typen und die Möglichkeiten ihrer Behebung, 7) die Werkstatt des Übersetzers u. a. hingewiesen werden. Als eine wichtige und durchaus positive Nebenwirkung ist hier natürlich die sprachliche Weiterbildung zu betrachten. Ziel der Studie liegt in der Feststellung, ob sich Texte, die nur selten in der Berufspraxis Anwendung finden, als Übungsstoff für die Übersetzung dienen können. Die Studie knüpft an die von Ramge (1993) unter Germanistikstudenten der Justus-Liebig-Universität Gießen durchgeführte Studie an, die beweisen konnte, dass durch die Lektüre von Zeitungen kognitives Textsortenwissen erworben wird. Bei dem Test handelte es sich um die Erforschung der TRANSLATION AND MEANING, PART 10, 2013, 231 - 238 233 textsortenspezifischen Merkmale des Kommentars i. e. S. Der hier gewählte Text ist ein Kommentar i. w. S., nämlich eine ironische Meinungsäußerung der Textsorte Glosse. Auf dieser Grundlage ist anzunehmen, dass die Übersetzer (als Zeitungsleser) ein bestimmtes Pensum an Textsortenwissen besitzen und die Formulierungsmuster sowohl in der Rezeption des AT als auch der Produktion des ZT einsetzen können. Die gestellte Aufgabe bestimmt den Skopos des Translats in Bezug auf die besonderen Eigenschaften des Textrezipienten. Hervorzuheben sind vor allem diejenigen Züge, die auf das Dekodieren des Inhalts beeinflussen könnten. Wir wollen hier die Arbeit an einem Pressetext mit der zugehörigen Aufgabenstellung besprechen und die Zieltexte evaluieren. Diese Übung ist als Einstieg in den Übersetzungsworkshop gedacht, bei dem die Studenten durch die Analyse der Problemstellen und der begangenen Fehler, die Prinzipien des interlingualen und interkulturellen Transfers entdecken sollen. Dazu ist seitens der Kursteilnehmer natürlich das Verständnis der gewählten Fehlertypologie von Bedeutung. Dies ist auch notwendig zur Erläuterung theoretischer Voraussetzungen des Übersetzens. 3.3 Zum Text Beim gewählten Beispieltext handelt es sich um eine Glosse aus Die Welt am Sonntag vom 20. Mai 2007. Bei der charakteristischen, rhetorisch ausgeprägten Struktur, ironischen Färbung, Meinungsbetonung der Textsorte ist es interessant zu beobachten, ob es den Übersetzern gelingt, die textsortenspezifischen Merkmale (Lexik, Syntagmen, Syntax, Idiomatik, die stilistischen Züge u.a.) im Zieltext widerzugeben. Zusätzlich ist die Bezeichnung „Glosse“ den meisten polnischen Deutschsprechenden nicht geläufig und bei der polnischen Äquivalenz „felietion“ besteht auch die Gefahr, dass man zur Opfer der sog. falschen Freunde wird, und dass falscher Textsortenwissen aktiviert wird. Viele der angehenden Übersetzer führen trotzt der Erfahrung eines Zeitungslesers die Synthese des ZT ausschließlich nach den wahrgenommenen Merkmalen des AT durch. Der hier gewählte Text äußert sich kritisch und ironisch über eine kontroverse Erfindung aus der Motorwelt. Die Kolumne „Mielke murrt“ verspottet die Idee eines Wagens, der sich ohne Insassen orientieren könnte. Der Ausgangstext: [1]3 Mielke murrt (Die Welt am Sonntag Nr. 20 vom Sonntag, den 20. Mai 2007: 1) [2] Im Straßenverkehr stört der Mensch nur [3] Berliner Informatiker haben ein Auto vorgestellt, das sich dank Computern und Sensoren ohne Insassen bewegen und orientieren kann. [4] Eine erfreuliche Nachricht für Familienväter mit Großraumlimousinen: [5] Bisher mussten sie ihre Angehörigen sonntags zu sinnfreien Ausfahrten drängen, „um den teuren Wagen mal zu bewegen". [6] Dann kutschte die Mischpoke ein paar Stunden in ein Ostseebad, flanierte die doofe Seebrücke einmal hoch und runter, und kutschte dann wieder zurück. [7] Dieses Elend ist vorbei. [8] Die Menschen können endlich zu Hause vor dem Fernseher statt im Auto trostlos herumsitzen. [9] Weil der Wagen leer unterwegs ist, spart man auch noch Treibstoff, insbesondere bergan. [10] Wer mit seinem neuen Auto angeben will, klebt ein Foto von sich auf die Windschutzscheibe. [11] Den Forschern ist wichtig, dass der Straßenverkehr trotz all der Technik nicht 3 Die Markierung der Sätze mit Nummern soll die Orientierung bei der Evaluierung im Punkt 3.4. erleichtern. 234 MARCIN MICHOŃ kalt und seelenlos erscheint. [12] Es soll feminine Modelle geben, die nicht einparken können oder von Berlin nach Paris über Moskau fahren. [13] Andere Varianten könnten über ein Surround-Lautsprechersystem verfügen, das vor Kreuzungen aus Richtung des Beifahrersitzes „Grüner wird's nicht!" absondert, während es aus dem Fond ununterbrochen quäkt: [14] „Sind wir schon da?" [15] Dass der Prototyp die Vorfahrtsregeln noch nicht beherrscht, ist kein Problem. [16] Erstens geht es sehr vielen Fahrern ebenso. [17] Zweitens verfügt der Wagen über sogenannte Scheibenwischer, die den Scheibenwischergruß imitieren können. [18] Für den Fall eines Unfalls gibt es den integrierten Injuriengenerator. [19] Damit können sich Kraftfahrzeuge stundenlang gegenseitig beschimpfen, während ihre Halter einen Likör trinken. [20] Ihre spektakuläre Präsentation hat den Entwicklern bereits einen weiteren Auftrag eingebracht: [21] Die Vorstände von SPD und CDU haben die „Konzeptstudie einer modernen Volkspartei" bestellt, die weitestgehend ohne Mitglieder und Wähler unterwegs sein kann. André Mielke Die Aufgabenstellung bei solchen Texten kann durch verschiedene didaktische Zwecke geprägt sein. In diesem Fall ist es wichtig hervorzuheben, was die intentionelle Ebene des Textes beinhaltet, und wie sie auf die Polnisch sprechenden Adressaten zu adaptieren wäre. Das bedeutet, dass die Übersetzer um den interkulturellen Transfer bemüht sein sollen, und die für den deutschsprachigen Kulturkreis spezifischen Elemente der Wirklichkeit durch polnische oder universelle Äquivalente ersetzten, so dass sich die Rezipienten des ZT durch den Kommentar betroffen fühlen. Das Ziel liegt dabei darin, sich dem behandelten Sachverhalt gegenüber ebenfalls im ZT kritisch zu äußern. Wenn der interkulturelle Transfer, diese Adaptation im Zielsprachentext in Bezug auf die Aspekte polnischer Wirklichkeit nicht stattfindet, wird möglicherweise die ironische Haltung gegenüber den Empfängern des AT erzeugt. An dem Text kann dem Übersetzernachwuchs erläutert werden, dass die Translation nicht in der wortwirklichen Umformung des AT in die ZS besteht, sondern es handelt sich beim Übersetzen darum „in der Empfängersprache das beste natürliche (sic) Gegenstück zur Ausgangsbotschaft zu schaffen, erstens was den Sinn und zweitens was den Stil anbelangt“ (Stolze, 2001: 94 nach Nida/Taber, 1969: 11). Diese Definition müsste um die Wiedergabe der Funktion des Ausgangstextes im Zielsprachentext erweitert werden, da die Intention des Senders ein wichtiger Faktor für den Skopos ist (vgl. Nord, 2002: 29ff). Der Übersetzungskurs ist der richtige Augenblick um aus Fehlern zu lernen. Eine richtig durchgeführte Evaluation der Fehler kann in Verbindung mit zutreffenden therapeutischen Maßnahmen, in vieler Hinsicht Erfolge bringen. 3.4 Zur Evaluation der Texte Nord (2006: 386) spricht sich zunächst für die Hervorhebung der wirklichen Übersetzungsfehler aus und meint dabei Verstöße gegen den Übersetzungsauftrag. Es wurde dabei ein Korrekturmuster vorbereitet, an dem die Typen der Fehler sichtbar werden, damit sich der bewertete Text für weitere therapeutische Arbeit eignet. Sie unterscheidet: • • • pragmatische Fehler (P) entstehen aus der unzureichenden Berücksichtigung der pragmatischen Ebene des Zieltextes; kulturelle Fehler (K) bei der ungenügenden Anpassung des Textes an Normen- und Konventionssysteme; formale Fehler (F) als Resultat mangelnder Beachtung formaler Vorgaben. TRANSLATION AND MEANING, PART 10, 2013, 231 - 238 235 Diese Fehler können mit einer zusätzlichen Bemerkung versehen werden, die sie einer der vielen Fehlerkategorien bzw. Fehlersorten innerhalb der genannten Gruppen zuordnet (z.B. P/Adr/Vorwissen, K/TS/Textaufbau4). In den Texten, die als Resultat der Übung entstanden sind, findet man Beispiele aller Typen der o.g. Fehler. Sie sind als Resultat der Nichtbeachtung wichtiger Prinzipien der Übersetzung oder der Aufgabe zu verstehen, und können exemplarisch für therapeutische Zwecke in der Übersetzungs- und ferner Dolmetscherdidaktik verwendet werden. Die pragmatischen Fehler in den ZT hängen in vielen Fällen mit kulturellen Fehlern zusammen und resultieren aus der Nichtbeachtung der Aufgabenstellung, die im didaktischen Prozess den Auftrag ersetzt. Einige Beispiele, die nun angeführt werden, dienen als Beweis für die These, dass das Kriterium der Praxisnähe nicht den Einsatz von Pressetexten ausschließt, zumal die Übung an solchen Texten mehrere wichtige Aspekte des Übersetzens explizieren lässt. In den korrigierten und evaluierten ZT sind unter anderem Verstöße 1) 2) 3) 4) 5) gegen die Textsortenspezifik, gegen den Still, gegen das Gebot des interkulturellen Transfers, gegen die Äquivalenz und Interferenzfehler, sowie Misserfolge auf Grund sprachlicher Inkompetenz und einige andere zu beobachten. Im Ausgangstext werden durch Fettdruck die Stellen hervorgehoben, die für die Probanden auf Grund der gestellten Aufgabe oder aus sprachlichen Gründen zu Problemstellen wurden. Zu 1) Die Glosse ist ein ironischer Kommentar, der in erster Linie auf Meinungsausdruck und –beeinflussung abgesehen ist und dessen Struktur nach dem erörternden Schema aufgebaut ist. Die Glosse bedient sich der stilistischen Färbung mit Ironie um das jeweilige Thema kritisch zu schildern und zur Einnahme einer bestimmten Haltung zu bewegen. Den meisten Studenten ist es gelungen, die textsortenspezifische Oberflächenstruktur des AT im ZT zu rekonstruieren. Der Teufel steckt jedoch bekannter Weise im Detail. Die Tiefenstruktur des Textes wurde nicht von allen Probanden gründlich genug analysiert, wodurch der Sinn einzelner Textelemente, allem voran idiomatischer Syntagmen nicht richtig oder nicht mit gleicher Qualität rekonstruiert wurde. Zu 2) Mit den textsortenspezifischen Merkmalen hängt in diesem Fall die stilistische Färbung zusammen. Die Ironie im AT ist außer der sog. Thematisierung in allen Elementen des Textschemas, d. h. in der Argumentierung und im Fazit zu erkennen, wobei sie oft idiomatisch impliziert wird. Mit diesem Textelement hatten die meisten polnischen Studenten Probleme. Nicht alle ironischen Implikaturen konnten von den Probanden auf Grund mangelnder sprachlicher und kultureller Kompetenz wahrgenommen werden. Und wenn es doch der Fall war, dann tauchten in der Rekonstruktionsphase Problemstellen auf, weil es nicht leicht fällt, die ironisch gefärbten idiomatischen Syntagmen des AT mit dem gleichen Inhalt in der ZS so zu formulieren, dass die Konvention der Glosse beibehalten bleibt. Es ist für die unerfahrenen Übersetzer sehr nützlich, dass sie bei der Evaluierung der ZT-e Lösungsvorschläge miteinander konfrontieren können und somit Wege der möglichen Überwindung von Problemstellen im Text kennen lernen. Die Evaluierung der Texte im Plenum ähnelt in dieser Hinsicht den Protokollen des lauten Denkens und ist gleichermaßen fruchtbar, auch wenn dabei andere Ziele erreicht werden (vgl. dazu Kußmaul/Hönig, 2006). 4 P/Adr/Vorwissen – zu hohe Forderung an das Vorwissen des Adressaten; K/TS/Textsorte – Abweichung von der Konvention der jeweiligen Textsorte. 236 MARCIN MICHOŃ Zu 3) Damit die Rezipienten den Sachverhalt des ZT nicht aus fremder Perspektive betrachten, müssten die einzelnen Syntagmen und Lexeme mit Denotaten, die auf den deutschsprachigen Kulturkreis hinweisen, in die polnischen oder universellen Äquivalente transformiert werden. Die Adaptation betrifft auch die geographischen Bezüge im Text. So könnte z.B. im Syntagma „von Berlin nach Paris über Moskau“ [Satz Nr. im Text 12] die deutsche Hauptstadt durch die polnische ersetzt werden. Auf die Idee sind die meisten Probanden nicht gekommen. Ähnlich ist es im Falle der Syntagmen: „Grüner wird’s nicht!“ [S. 13], „Scheibenwischergruß“ [S. 17], „Likör trinken“ [S. 19]. Alle drei Wendungen haben adäquate Äquivalente im Polnischen, die ähnliche spezifische Verhaltensweise in analogen Situationen bezeichnen. Die meisten Übersetzenden haben sich jedoch auch hier für die wortwirkliche Übersetzung entschieden. Im Falle der bildhaften Bezeichnung „Scheibenwischergruß“ besteht die Schwierigkeit der Übersetzung zusätzlich darin, dass im ersten Satz des Gefüges auf den Scheibenwischer als Teil eines Autos hingewiesen wird [S. 17]. Die ZT beinhalten ähnliche Fehler in Bezug auf die Aufgabenstellung, wenn es sich um die systempolitischen Unterschiede zwischen Deutschland und Polen handelt. In den Zieltexten vieler Probanden werden die Namen deutscher Parteien „SPD und CDU“ [S. 21] beibehalten, was wieder beim Rezipienten des ZT die Perspektive des Beobachters und nicht des Betroffenen hervorruft. Zu 4) Hinsichtlich des Transfers politischer Spezifik beider Länder taucht in vielen Texten ein interessanter Interferenzfehler aus der ZS auf: Den Begriff „Volkspartei“ [S. 21] haben die meisten Studenten als „partia ludowa“ übersetzt, was die Bedeutung des deutschen Lexems nicht wiedergibt und im Wiederspruch zu den zwei größten Parteien in Deutschland steht, die mit diesem Begriff bezeichnet werden. Ein interessantes Beispiel der Unzulänglichkeit in Bezug auf die Adäquatheit ist in der Thematisierung vieler ZT zu finden. Das Bedeutungsfeld des deutschen Lexems „Insassen“ [S. 3] ist breiter als beim polnischen Wort „pasażerowie“, dass in diesem Fall den Fahrer des Wagens ausschließt. So ist die Aussage: „auto … może poruszać się bez pasażerów” nicht äquivalent zur Erfassung der Intentionalität in dieser Mikrostruktur des Ausgangstextes. Zu 5) Dadurch dass es unter den Probanden viele Studenten gab, die sprachlich für die Übersetzerarbeit noch nicht reif sind, können in den ZT Aussagen gefunden werden, die von mangelnder Kompetenz zur Analyse des AT zeugen. So haben beinahe alle Studenten das ironische Wortspiel „Für den Fall eines Unfalls“ [S. 18] (zu übersetzten z.B. als „Na wypadek wypadku“) in den ZT nicht adäquat erfasst. Falsch übersetzt werden unter anderem die Syntagmen „unterwegs sein“ [S. 21], „Injuriengenerator“ [S. 18] oder „flanierte … hoch und runter“ [S. 6]. Dass die Studenten in vielen Fällen nicht auf die polnische Äquivalenz des Lexems „Mischpoke“ [S. 6] gekommen sind, zeugt so wie die obigen Beispiele von fremdsprachlicher Inkompetenz und ungeübter Handhabung zugänglicher Quellen. 4. Schlussbemerkungen: Die Evaluation der ZT in der dargestellten Fallstudie beweist: 1) dass die Übersetzung von Pressetexten dank der Vielfalt gemeinsprachlicher Ausdruckmittel die sprachlichen Kompetenzen fördert; 2) dass an Pressetexten viele Übersetzungsprinzipien erläutert werden können, wie etwa das Wesen der Äquivalenz oder die möglichen Wege der Lösung von Problemfällen, z.B. wie und mit Hilfe welcher Mittel die Übersetzer die informative und intentionale Ebene des Textes erschließen lassen; 3) dass sich durch die drei Übersetzungsmethoden Analyse, Transfer und Synthese viele Fehlgriffe bei der Translation vermeiden lassen; 4) dass sich viele Pressetexte für Adaptationsübungen sehr gut eignen und somit für literarische TRANSLATION AND MEANING, PART 10, 2013, 231 - 238 237 Übersetzerarbeit vorbereiten; 5) dass Text- und Textsortenwissen sowohl bei der Rezeption des AT als auch bei der Produktion der ZT in einem bestimmten Maße für die Übersetzer unbedingt sind. Es können anhand dieses einfachen Beispiels noch andere Argumente für den Einsatz von Pressetexten in der Ausbildung von Übersetzern anführen werden. Die hier exemplarisch dargestellten Resultate einer Fallstudie lassen für die Übersetzung von Pressetexten im Übersetzungskurs plädieren. Bibliographie Bednarczyk, A (2009), Podstawy Teorii Przekładu – Skrypt. IR UŁ, Łódź. Best, J./Kalina, S. (2002), Übersetzten und Dolmetschen. Eine Orientierungshilfe. A. Franke Verlag, Tübingen-Basel. Brinker, K. (2000a), „Textfunktionale Analyse“. In: Brinker, Klaus [et al.] (Hg.): Text- und Gesprächslinguistik: ein internationales Handbuch zeitgenössischer Forschung. Handbücher zur Sprach- und Kommunikationswissenschaft Band 16. De Gruyter, Berlin, New York, S. 175-186. Brinker, K. (2000b), “Textstrukturanalyse”. In: Brinker, Klaus [et al.] (Hg.): Text- und Gesprächslinguistik: ein internationales Handbuch zeitgenössischer Forschung. Handbücher zur Sprach- und Kommunikationswissenschaft Band 16. De Gruyter, Berlin, New York, S. 164-175. Brinker, K. (2005), Linguistische Textanalyse. Eine Einführung in Grundbegriffe und Methoden. ErichSchmidt-Verlag Berlin. De Beaugrande, R./Dressler, W. (1981), Einführung in die Textlinguistik. Niemeyer, Tübingen. Didaoui, M (2006), “Qualitätslektorat”. In: Mary Snell-Hornby (et al.): Handbuch Translation. Stauffenburg Verlag Brigitte Narr, Tübingen, S. 381-383. Fix, U. (2008), „Text und Textlinguistik“. In: Janich, Nina (Hg.): Textlinguistik. 15 Einführungen. Günter Narr, Tübingen, S. 15-34. Gawlas, C. (2006), “Texte von Presseagenturen”. In: Mary Snell-Hornby (et al): Handbuch Translation. Stauffenburg Verlag Brigitte Narr, Tübingen, S. 236-237. Kade, O. (1968), „Kommunikationswissenschaftliche Probleme der Translation“. In: Albrecht Neubert (Hg.): Grundfragen der Übersetzungswissenschaft. Leipzig. Kielar, Barbara Z. (2003): Zarys translatoryki. Katedra Języków Specjalistycznych UW. Warszawa. Koller, W. (2001), Einführung in die Übersetzungswissenschaft. Quelle & Meyer Verlag, Wiebelsheim. Kußmaul, P. (2006), „Textauswahlkriterien“. In: Mary Snell-Hornby (et al.): Handbuch Translation. Stauffenburg Verlag Brigitte Narr, Tübingen, S. 358-359. Kußmaul, P. (2007), Verstehen und Übersetzen. Ein Lehr und Arbeitsbuch. Narr Francke Attempto Verlag. Tübingen. Kußmaul, P./Hönig, H. (2006), „Einblicke in mentale Prozesse beim Übersetzen“. In: Mary Snell-Hornby (et al.): Handbuch Translation. Stauffenburg Verlag Brigitte Narr, Tübingen, S. 170-178. Majkiewicz, A. (2008), Intertekstualność - implikacje dla teorii przekładu. Wydawnictwo Naukowe PWN. Nida, E./Taber, C. (1969), Theorie und Praxis des Übersetzens, unter besonderer Berücksichtigung der Bibelübersetzung. Weltbund der Bibelgesellschaften. Nord, C. (2002), Fertigkeit Übersetzten. Ein Selbstlernkurs zum Übersetzenlernen und Übersetzenlehren. Editorial Club Universario, Alicante. Nord, C. (2006), “Transparenz der Korrektur”. In: Mary Snell-Hornby (et al.): Handbuch Translation. Stauffenburg Verlag Brigitte Narr, Tübingen, S. 384-387. Nord, C. (2007), Textanalyse und Übersetzen. Theoretische Grundlagen, Methode und didaktische Anwendung einer übersetzungsrelevanten Textanalyse. Julius Gross Verlag, Tübingen. Nowag, W./Schalkowski, E. (1998), Kommentar und Glosse. UVK Medien. Konstanz. Pieńkos, J. (2003), Podstawy przekładoznawstwa – od teorii do praktyki. Zakamycze. Ramge, H. (1993), „Wie simuliere ich einen Zeitungskommentar? Vorläufiges über die Einheiten und Elemente des Textschemas”. In: Kwartalnik Neofilologiczny 40, Warszawa, S. 217-229. Sieradzka, M. (2010), „Gutes Deutsch – schlechtes Polnisch. Übersetzung von Pressetexten als Versuchsgelände für die Prüfung und Schulung der fremdsprachlichen Kompetenz in der universitären Germanistenausbildung?” In: Anna Małgorzewicz (Hg.): Translation: Theorie – Praxis – Didaktik. Studia Translatorica. Atut, Neisse Verlag, Dresden, Wrocław, S. 461-471. Snell-Hornby, M. (2006) (et al.): Handbuch Translation. Stauffenburg Verlag Brigitte Narr, Tübingen. 238 MARCIN MICHOŃ Stolze, R. (1999), Die Fachübersetzung. Eine Einführung. Günter Narr Verlag, Tübingen. Stolze, R. (2001), Übersetzungstheorien. Eine Einführung. Günter Narr Verlag, Tübingen. Quellen: Die Welt am Sonntag Nr. 20 vom Sonntag, den 20. Mai 2007. TRANSLATION AND MEANING, PART 10, 2013, 239 - 243 239 EINIGE ARGUMENTE FÜR DIE VERWENDUNG DER REGENBOGENPRESSE ALS EINSTIEG INS ÜBERSETZUNGSTRAINING Mariusz Milczarek Akademie für Human- und Wirtschaftswissenschaften, Łódź, Poland Abstract: Übersetzen ist ein Prozess, bei dem neben linguistischen Aspekten auch psycho- und soziolinguistische Faktoren zu berücksichtigen sind. Die Rolle des Übersetzers ist daher mehr als nur die Übertragung eines Inhalts aus der Ausgangssprache in die Zielsprache. Der Ausbildungsprozess der Übersetzer ist daher sehr komplex. Im Zuge der Lehrveranstaltungen sollten bei Lernern Sprach-, Sach- sowie Übersetzungskompetenz gleichmäßig aktiviert werden. Dabei werden zuerst theoretische Ansätze vermittelt, die u.a. Aspekte der Kommunikation und die Hauptprobleme der Übersetzungswissenschaft betreffen, z.B. das Problem der Äquivalenz. Aber erst regelmäßiges Üben, nach Schwierigkeitsgrad abgestuft, kann im Endeffekt didaktische Erfolge bringen. Der Fokus des folgenden Artikels liegt auf den ersten Schritten des Übersetzungstrainings. Der Autor meint, dass vor dem richtigen Training eine praxisbezogene Einführung angebracht wäre, ohne die Lerner gleich mit komplizierten Übersetzungsproblemen zu konfrontieren. Dafür eignet sich die Übersetzung von Texten aus der Regenbogenpresse. Die Texte, die eher trivialer Unterhaltung als informativen Zwecken dienen (daher die oft gebrauchte, abwertende Bezeichnung ‚Boulevardpresse‘), können in dieser Funktion von großem Wert sein. Die Übersetzer-Lerner werden mit den Hauptproblemen des Fachs vertraut gemacht, ohne dass sie ein zu großer Schwierigkeitsgrad und die Kompliziertheit des Ausgangstextes überfordern. Ein positiver Nebeneffekt ist der psychologische Aspekt: Kleine Erfolge, d.h. die gelungene Übersetzung, werden Anfänger ermuntern, sich mit komplizierteren Texten auseinanderzusetzen. Im Artikel wird deshalb anhand von einigen Beispielen ausführlicher auf diese These eingegangen werden. Dem komplexen Prozess der Übertragung eines Inhalts aus der Ausgangssprache in die Zielsprache liegen viele – auch außersprachliche – Faktoren zu Grunde, die der Übersetzer zu berücksichtigen hat. An dieser Stelle möchte ich z.B. Łabno-Falęcka (1995, S. 11) zitieren, die darauf hinweist, dass „eine Translation nicht nur ein sprachlicher, sondern immer auch ein ‚kultureller Transfer‘ ist.“ Vom Übersetzer wird primär erwartet, dass er - so Grucza (1970, S. 42) - „wenigstens bestimmte Teile zweier Sprachen beherrscht“. Eine perfekte Kenntnis einer Sprache wäre ein Idealfall, aber die kommt laut Karl Dedecius (1986, S. 142) eigentlich nie vor: „Es gibt niemanden, der seine Muttersprache vollkommen kennt – mit all ihren phonetischen, lexikalischen, syntaktischen, idiomatischen und anderen Mehrsprachigkeiten – um so weniger gibt es jemanden, der sich anmaßen könnte, in dieser Vollkommenheit zwei oder noch mehr Sprachen synchron zu beherrschen. Gerade aber dies ist es, was man naiverweise stillschweigend bei Übersetzern voraussetzt.“ Die sprachliche Kompetenz allein ist noch keine ausreichende Prämisse für die gelungene Übersetzung. Darüber hinaus muss der Übersetzer auch über Sachkompetenz verfügen, die für die richtige Interpretation des Originaltextes unentbehrlich ist. Last but not least ist die Übersetzungskompetenz (das sog. Fingerspitzengefühl) zu nennen, die von Weigt (2002, S. 396 ) mit Recht als „Krönung aller Kompetenzen“ bezeichnet wird. Es ist daher wichtig, dass in der Übersetzerausbildung Sprach-, Sach- sowie Übersetzungskompetenz gleichmäßig bei Lernern aktiviert werden sollen. Es sollte im Unterricht auch betont werden, dass das Übersetzen eigentlich nie abgeschlossen ist, auch wenn das Translat bereits fertig ist, sondern als eine Approximation angesehen werden sollte- dazu Reiß (1991, S. 113) „Es gibt nicht ‘die‘ optimale Translation. Es gibt nur das Streben nach Optimierung unter den jeweils gegebener aktuellen Bedingungen.“ 240 MARIUSZ MILCZAREK Ebenso wie der Prozess der Translation, ist auch die Ausbildung der Übersetzer kompliziert.1 In den Vorlesungen werden den Lernern zuerst theoretische Ansätze, das notwendige Know-how beigebracht, indem ein Überblick über Geschichte, Übersetzungsmodelle und Hauptprobleme der Disziplin verschafft wird. Diese Etappe in der Ausbildung der Übersetzer halte ich für unerlässlich, obwohl es auch viele Gegenstimmen gibt.2 Aber erst die Praxis, ein durchdachtes, nach Schwierigkeitsgrad abgestuftes Training kann die Fähigkeiten der Lerner vervollkommnen. In meinem Referat möchte ich mich auf die ersten Schritte des Übersetzungstrainings konzentrieren. Ich vertrete die Ansicht, dass vor dem richtigen Training eine praxisbezogene Einführung angebracht wäre, ohne die Lerner gleich beim Einstieg mit komplizierten Übersetzungsproblemen zu konfrontieren. Es werden Argumente herangezogen, warum die Übersetzung der Texte aus der Regenbogenpresse keinesfalls als Unterforderung der Lerner angesehen werden sollten, sondern als eine Etappe, die die weitere Übersetzungspraxis positiv beeinflussen kann. An dieser Stelle ist nachdrücklich hervorzuheben, dass es dabei nicht das Ziel ist, diesen Typ der Zeitungen in irgendeiner Form aufzuwerten und dass die Verwendung der Regenbogenpresse im Übersetzungstraining nur in der Einstiegsphase vorgeschlagen ist. Einleitend möchte ich in aller Kürze die Merkmale dieser Textsorte3 nennen, die meistens v.a. mit auffälligem Layout und griffigen Schlagzeilen assoziiert wird, wie z.B. „Die DDR hat mir die Mutter geraubt.“ (Auf einen Blick, Nr.5/2009) „Aufatmen! Bleibt ihre süße Tochter von der schlimmen Erbkrankheit verschont?“ (Freizeit Aktuell, Nr. 1/2009) „Baby-Glück: Was ihr Auftritt im grünen Kleid verrät.“ (Woche heute, Nr. 51/2008) Jegliche Pressetexten sind an breites Publikum gerichtet und sollen berücksichtigen, dass sich ihre Rezipienten der alltäglichen Rede bedienen, in der phraseologische Wendungen einen festen Bestandteil bilden.4 Phraseologische Wendungen, die u.a. zur Plastizität der Sprache beitragen, erhöhen in der Regenbogenpresse den Leseanreiz. Sehr oft werden sie deshalb in den Schlagzeilen verwendet, z.B.: „Endlich Arbeit – aber das Amt lässt mich im Stich“ „Auf einen Blick“ Nr.49, S.13 jmdn. im Stich lassen – ‚jmdn. [in einer Notlage] allein lassen, jmdm. nicht helfen.‘ (D 11, S.692). „Trotz aller Liebe sind wir wie Hund und Katze“ „Auf einen Blick“ Nr.23, S.20 leben wie Hund und Katze – ‚sich ständig streiten‘ (D 11, S. 440). „Diese Frau bläst Politikern den Marsch“ „Bild-Online“, Zugriff am 15.09.2010 1 Vgl. Grucza Franciszek, 1970, S. 32. Vgl. dazu Reiß Katharina, 1976, S. 127. 3 Stella Neumann weist darauf hin, dass für empirische Studien der Begriff ‘Textsorte’ bevorzugt wird, dagegen In den theoretischen Arbeiten ‘Texttyp’. Vgl. Neumann Stella, 2003, S. 11. Auch Adamzik Kirsten (2004, S. 102) wendet sich dem Problem der Begriffsbestimmung bei der Klassifikation der Texte. Dabei sei der Terminus ‚Textsorte‘ für die nicht – literarische Texte vorbehalten und im deutschsprachigen Raum bevorzugt, dagegen ‚Texttyp‘ als ein Internationalismus, der die internationale Verständigung erleichtern soll. 4 Vgl. Lehr-Spławiński, 1987, S. 388: „[…] język dziennikarski powinien się w wyższym stopniu niż inne działy języka piśmiennego wiązać z żywą mową szerokich kół społeczeństwa, z którymi prasa codzienna powstaje w jak najściślejszej łączności.” 2 TRANSLATION AND MEANING, PART 10, 2013, 239 - 243 241 jmdm. den Marsch blasen – ‚jmdn. ausschimpfen, tadeln‘. (D 11, S. 477). Hinsichtlich ihrer Funktion dienen Zeitungen und Zeitschriften, die als Regenbogenpresse bezeichnet werden, eher trivialer Unterhaltung als informativen Zwecken5, weswegen sie oft unter der abwertenden Bezeichnung ‚Boulevardpresse‘ bekannt sind. Das Wesen dieser Textsorte charakterisiert treffend u.a. Adamzik (2004, S. 131), indem sie schreibt: „Bekanntlich wird z.B. in der sog. Boulevardpresse besonders Gewicht auf personenbezogene (HumanInterest-)Angaben gelegt, die die seriösen Blätter, Nachrichtenagenturen (und wahrscheinlich auch Leute, die das Thema ableiten bzw. eine Zusammenfassung erstellen wollen) als vom Hauptthema abführende Nebensächlichkeiten definieren.“6 Im Nachstehenden möchte vorschlagen, diese Texte aus einer anderen Perspektive zu beurteilen. Ich werde Argumente nennen, die nach meiner Ansicht den Einsatz der Regenbogenpresse im Übersetzungstraining motivieren. 5 − Ich möchte mit der Übersetzungsmethode anfagen. In seiner Klassifikation der Übersetzungstypen unterscheidet Karl Dedecius (1986) zwischen Übersetzung, Übertragung und Nachdichtung. Im Fall der journalistischen Texten wird der erstgenannte Übersetzungstyp vorgeschlagen, bei dem „das Original genau und ohne ‚künstlerisches Obligo‘ nach dem Kriterium der Genauigkeit wiedergegeben wird.“ (Dedecius, 1986). Für die Anfänger ist es also zweierlei von Vorteil, wenn sie an den Texten aus der Regenbogenpresse arbeiten, weil bei ihrer Übersetzung eine künstlerische Tätigkeit nicht nötig ist. − Das nächste Argument ist die Vielfalt an Themen der Artikel, wodurch die Sprachkompetenz der Lerner erweitert werden kann. Die Texte handeln von Ereignissen aus den unterschiedlichen Bereichen wie Sport, Wirtschaft, Gesundheit, Politik, Partnerschaft, Unterhaltung, Reisen, Livestyle, Ratgeber, usw. − Der nächste Vorteil ist die unkomplizierte Sprache der Texte, die bei der Interpretation eher keine Schwierigkeiten bereiten soll. Die Autoren der Artikel bevorzugen die Alltagsrede und greifen nach stilistischen Mitteln, die Expressivität steigern sollen, z. B. dem Gebrach von Phraseologismen7. − In der Regenbogenpresse sind oft Bezeichnungen in Gebrauch, die für die Inländer üblich - den Ausländern dagegen unbekannt sind, auch wenn ihre Bedeutung verständlich ist. Diese haben häufig einen okkasionellen Charakter oder beziehen sich auf Ereignisse aus dem Alltagsleben in dem jeweiligen Land. Was ist z. B. unter PKW-Maut zu verstehen? Wer wird als „Mövenpick-Minister“ bezeichnet und warum? Wer ist „Bundescremer“ oder „Jogi“ ? In diesem Sinne kann die Regenbogenpresse einem auch dazu verhelfen, die Sachkompetenz der Lerner zu erweitern. − Obwohl diese Texte aus der Sicht erfahrener Übersetzer unkompliziert sind, werden die Lerner bereits in dieser Textsorte mit den Hauptproblemen des Fachs vertraut gemacht, ohne dass sie ein zu großer Schwierigkeitsgrad und die Kompliziertheit des Ausgangstextes überfordern. Bei der Übersetzung von Texten der Regenbogenpresse kann u.a. die Äquivalenz, besonders wegen der Anglizismen und Phraseologismen, zum Problemfall werden. Vgl. Duden Deutsches Universal Wörterbuch, 1996, S. 1230. Die Autorin ergänzt aber weiter, dass es gegenwärtig schon üblich sei, dass die unterhaltende Funktion der Berichterstattung mit dem informativen Zweck einhergeht. (Adamzik Kirsten, 2004, S. 115). 7 Vgl. z.B. Fleischer Wolfgang 1982, S. 202 6 242 MARIUSZ MILCZAREK Phraseologismen: (z.B. Das zugrundeliegende Bild des Phraseologismus kann irreführend sein- besonders, wenn der Kontext auf die wörtliche Bedeutung einer der Komponenten im Kontext aktualisiert wird oder einfach unbekannt ist.) „Auf einen Blick“ Nr. 49/2008 „Werden wir Verbraucher gnadenlos über den Tisch gezogen?“ (S. 5) – ‚jmdn. übervorteilen‘ (D 11, S. 724) „Die Verbraucher haben jedenfalls die Nase voll“ (S. 5) – ‚jmds., einer Sache überdrüssig sein‘ (D 11, S.507) „Auf einen Blick“ Nr.49/2009 „Wer mittels einer Agentur Amor auf die Sprünge helfen will, sollte vorher das Institut prüfen […]“ (S.5) – ‚jmdm. weiterhelfen‘ (D 11, S. 677) „So gehen Sie auf Nummer sicher“ (S. 5) – ‚sich in jeder Hinsicht absichern‘ (D 11, S. 521) „Immer öfter geraten wir uns in die Haare“ (Seite 23) – ‚in Streit geraten‘ (D 11, S. 285) Anglizismen: (Das Problem in diesem Fall liegt darin, dass sie in der Alltagsrede häufig Anwendung finden, was im Polnischen nicht der Fall ist.) „Da halfen auch etliche Schreiben und all die teuren Handy-Telefonate mit der Hotline nichts.“ („Auf einen Blick“ Nr. 36/2008, S. 5) „Mit einem Schock für die Crew vom „Großstadtrevier“ beginnt der Auftakt von insgesamt 16 Wiederholungen der ARD-Kultserie“ („Auf einen Blick“ Nr. 32/2008, S. 73) „Stets top gestylt – nicht nur auf dem Rasen ist Joachim Löw ein erfolgreicher Mann.“ („Freizeit aktuell“ Nr. 1/ 2009, S. 2 ) „Marc und Sarah sind zwar nicht mehr ,in love‘, aber für die Kinder Summer und Tyler kommt dieses Jahr der Weihnachtsmann noch einmal im (noch) gemeinsamen Haus.“ („Freizeit aktuell“ Nr. 1/ 2009, S. 2 ) − Im Laufe der Einstiegsübungen mit den Texten aus der Regenbogenpresse wird mit einfachen Beispielen die Eigenart der Tiefenstrukturen der Fremdsprache erfasst und auf die Interferenz aufmerksam gemacht. Bei der Übersetzung aus dem Deutschen ins Polnische kommt oft das Problem der Syntax zum Tragen. Die Studenten erliegen nicht selten der Versuchung, z.B. die „typisch deutschen“ Schachtelsätze genau ins Polnische zu übertragen. Das gleiche gilt auch für die Passivkonstruktionen. Geübt auf diese Hürden zu achten, können Lerner allmählich mit weiteren Problemen der Übersetzung konfrontiert werden, indem schwierigere Texte eingesetzt werden, z.B. Fachtexte oder Literatur. Dann muss der Übersetzer, mit Dedecius (1986) gesprochen, mehr „intellektuelle Eigenleistung“ erbringen. Man könnte in der nächsten Phase des Trainings auch Artikel, die ein bestimmtes Thema behandeln mit demselben Text aus Nachrichtenmagazin, z.B. „Der Spiegel“ vergleichen, indem man z.B. der Frage nachgeht, was zum Thema gesagt wird, welche Aspekte des Ereignisses im Vordergrund ste- TRANSLATION AND MEANING, PART 10, 2013, 239 - 243 243 hen etc.8 Solch eine Übung kann das Problem der Funktionalstile und Übersetzungstypen sehr gut veranschaulichen. (Nach der Übersetzung kann man die Bewertung des Translats hinsichtlich seiner intra- und intertextuellen Kohärenz9 durchführen und daraus Schlussfolgerungen für die weitere Arbeit ziehen.) Abschließend möchte ich noch einmal auf meine These zurückkommen, dass durch den Einsatz dieser Textsorte in mannigfacher Weise die Studierenden profitieren können. Die Lerner werden sich bewusst, dass Übersetzung als Prozess aus mehreren Etappen besteht und durch mehrere Faktoren determiniert wird, wie z.B. die kommunikative Funktion des Textes, die Intention des Autors des Ausgangstextes etc. Hinzu kommt der psychologische Aspekt – der kleine Erfolg wird das Selbstbewusstsein der Lerner stärken und sie zu weiteren Erfolgen im Übersetzen ermuntern. Literaturverzeichnis Adamzik, K (2004), „Textlinguistik. Eine einführende Darstellung.“ Tübingen. Bausch, K./Weller, F. (Hrsg.) (1981), Übersetzen und Fremdsprachenunterricht. Frankfurt am Main. Filipowicz-Rudek, M./ Konieczna-Twardzikowa, J. (Hrsg.) (2002), Między oryginałem a przekładem VII. Radość tłumaczenia. Przekład jako wzbogacanie kultury rodzimej. Kraków. Fleischer, W. (1982), Phraseologie der deutschen Gegenwartssprache. Leipzig. Fleischer, W./ Michel, G./ Starke, G. (1993), Stilistik der deutschen Gegenwartssprache. Frankfurt am Main. Grucza, F. (1970), „Fremdsprachenunterricht und Übersetzung“ in: Bausch Karl-Richard/Weller Franz-Rudolf (Hrsg.) (1981), Übersetzen und Fremdsprachenunterricht. Frankfurt am Main. Lehr-Spławiński, T. (1987), Język polski. Pochodzenie, powstanie, rozwój. Warszawa. Łabno-Falęcka, E. (1995), Phraseologie und Übersetzen: eine Untersuchung der Übersetzbarkeit kreativinnovativ gebrauchter wiederholter Rede anhand von Beispielen aus der polnischen und deutschen Gegenwartsliteratur. Frankfurt am Main. Neumann, S. (2003), Textsorten und Übersetzen. Eine Korpusanalyse englischer und deutscher Reiseführer. Peter Lang GmbH Europäischer Verlag der Wissenschaften. Frankfurt am Main. Reiß, K./ Vermeer, H. (1991), Grundlegung einer allgemeinen Translationstheorie. Tübingen. Reiß, K. (1976), „Didaktik des Übersetzens: Integration der Sprachwissenschaft in den Übersetzungsunterricht“ in: Bausch Karl-Richard/Weller Franz-Rudolf (Hrsg.) (1981): Übersetzen und Fremdsprachenunterricht. Frankfurt am Main. Snell- Hornby, M. (Hrsg.) (1986), Übersetzungswissenschaft- Eine Neuorientierung: zur Integrierung von Theorie und Praxis“. Tübingen. Weigt, Z. (2002), „Ausgewählte Probleme bei der Interpretation und Übersetzung deutsche Und polnischer Fachtexte”. In: Lewandowska-Tomaszczyk Barbara/ Thelen Marcel: (2002) Translation and Meaning. Part 6.“ Maastricht. 8 9 Vgl. Dazu Adamzik,Kirsten, 2004, S. 131. Vgl. dazu Reiß Katharina, 1991, S. 114. 244 TRANSLATION AND MEANING, PART 10, 2013, 245 - 251 245 TEACHING, TRAINING, EDUCATING? TERMINOLOGICAL AMBIGUITY IN TRANSLATOR EDUCATION THEORY Paulina Pietrzak University of Łódź, Poland Abstract: The concepts of translators, translation and its teaching recur in the discourse of the field but the terminology for these concepts used in translator education theory is by no means unified. Two questions among many arise, namely who translators are and where they come from. In this study, in regard of the process of teaching translation, a translator will be broadly regarded as the end product of translation education. This rather one-dimensional approach has been adopted primarily to emphasise the duration and complexity of the practice of translation competence development. The translator, regarded as the product of translation training, should, ideally, be a full-fledged professional, aware of the purpose of the work commissioned, well-versed in translation techniques and computer software available, proficient and efficient in both source and target language, working in accordance with the code of ethical conduct with a view to serving the function allotted by the society. However, on attempting any definition of either translators or the path of their creation and the people and techniques serving to create them, it appears vital to consolidate the terms used to describe these notions; only when that corpus of theoretical devices has been fully delineated, analysed and disambiguated, does it become feasible to direct the critical focus onto translators and their education. Yet, the terminology used in translator education theory to date remains to a significant extent chaotic. Some translation scholars opt for the term translation teacher, some of them favour the term translation trainer but a great number of theoreticians tend to use them interchangeably. The usage of these terms and the framework for this usage will therefore come under scrutiny in order to investigate whether they are fully interchangeable or maybe the contexts in which they are used differ. The following analysis will determine whether referring to one term only can suffice and, if not, will advocate the usage of more than one term but in a systematized way. It is commonly accepted to describe the person responsible for translator education using the term translation teacher. Some scholars, like Adab (1996, 2000), Dodds (1999), Schäffner (2000) combine the two terms into an agglutinated teacher/trainer; some of them, however, decide to use only one of the terms or propose to resort to some other terms. In the present analysis the attention will be devoted to the term teacher, used by such scholars as Wills (1982), Newmark (1991, 1993), Kussmaul (1995), Kiraly (1995), Hatim (2001), Munday (2001), Beeby (2004), González Davies (2005), Hatim and Mason (2007). The term is normally defined as “a person who teaches in a school” while the verb to teach means to: • • • • impart knowledge to or instruct (someone) in how to do something, especially in a school or as part of a recognized programme give instruction in (a subject or skill) cause to learn by example or experience advocate as a practice or principle (COD 1999: 1469) So teachers are the people who impart, instruct, give, cause and advocate; therefore the impression which is conveyed here is that they are executive, decisive and very influential. Newmark explains that: “Often the translation teacher practises the systematic routines of a craft teacher. He does not lecture; he teaches in a classroom. The difference is (here he resembles the literary critic) that every text presents a new task that challenges his resourcefulness. He is teaching, not a new subject (translation was the staple, first of classics, then of so-called ‘modern’ language teaching), but an old one that has been transformed in the last twenty years, when professional translation courses may be said to have started. (The first, appropriately for its international significance, was in Geneva.) The subject that used to be 246 PAULINA PIETRZAK ‘academic’ in a grammar school sense is now basically a craft subject in a technical college sense, but stretches to areas of linguistics, language philosophy, language criticism and cultural studies” (Newmark, 1991: 137). He therefore implies the practical aspect of the job. It might be called a craft subject but it is worth repeating after González Davies that: “Most other subject categories taught as undergraduate degrees have been learnt at school with textbooks and appropriate activities. This is not the case with translation. It is not a school subject nor is it directly related to one, such as Medical Science and Biology for future doctors or biologists. Its closest relative could be Language Learning, both native and foreign, so it seems logical to draw from the teaching approaches and ideas from this field and then select, integrate and adapt the special characteristics of translation studies” (González Davies, 2004: 11). Consequently, when compared to language learning or – from a pedagogical perspective – language teaching, the person in charge of such a course should be called a teacher. However, as González Davies goes on to explicate, “the approaches cannot be identical because translation competence goes beyond language learning and has different objectives, but bridges can be built” (2004: 11). It is far more complex because, inter alia, simple communication, which is the main aim of teaching languages, here, in translation process, is constrained and guided to such an extent that teaching becomes far too insufficient a term to denote it. Trying to find the proper term suitable for referring to the person in charge of creating transition professionals, the aims of translation education should be taken into consideration to recognize the nature of the process of translation teaching. In other words, what has to and what can be taught during this process? We all have at least some partial translational competence, if not innate then simply gained in the process of learning a language. The fact is that each individual has diverse skills and thus also different mental structures of understanding. Bearing this universal truth in mind, Presas declares that the development of translation competence consists basically of three kinds of processes: (1) the acquisition of previously non-existent competences; (2) the restructuring of already existing competences in order to facilitate transfer competence; (3) the acquisition of strategic competence. (2000: 29) In this way, she emphasizes the need to approach each translation trainee individually and take his or her needs into consideration in the process of arranging and conducting the translation course they attend. Robinson comments pertinently that “we all learn in different ways, and institutional learning should therefore be as flexible and as complex and rich as possible, so as to activate the channels through which each student learns best” (2003: 49). Therefore, when translation educators, facing the multifarious challenge of translation competence development, additionally take a more personal approach, the organization of translation class seems rather arduous. Bearing in mind that each translation task requires and employs a different set of skills, a combination of the universal competences crucial for novice translators must be decided on. The delineation of the universal competences is an attempt to answer the nagging question posed by Kiraly who wonders “what skills and knowledge are common to all translators, regardless of context” (1995: 14). The skills required as the basis for a successful translation are bound to a large variety of factors. According to Presas, the overall factors that influence the choice of skills to be taught in a translation course are directionality, modality, specific language combination, and such text-oriented factors as text type, subject matter or language competence (2000: 21). Not only linguistic factors matter here, but also a wide range of interdisciplinary ones like extensive memory, code-switching, ability to control interference, self-confidence in performance, communication skills, persistence in problem sol- TRANSLATION AND MEANING, PART 10, 2013, 245 - 251 247 ving and so forth. All the factors combined affect the choice of the skills to be taught or developed in translation training. Nonetheless, the question presents itself whether the necessary abilities are at all feasible to be taught to a translation trainee. This uncertainty is what undermines the validity of the innumerable inventories of the necessary skills to be acquired. The skills are not typical or inherent in all users of two languages, so can it be assumed that the activation of certain resources may be learnable? Be it so, the only legitimate assumption would be to attempt to sensitize students and provide them with guidance rather than patterns of behaviour to follow. Practice is therefore the very process whereby students establish their own individual ways of dealing with translational problems; instead of listing the skills to be developed, it is therefore more reasonable to focus on particular components of translation competence which may be necessary for performing a planned translational task. While performing certain tasks involving the activation of the existent but latent resources, the intended components of translation competence are developed. Gile observes that some components of translation competence that are considered “prerequisites for admission into translation schools, but do not form an objective or component of training, although training should improve the subjects’ capacity to use them more fully” (1995: 5). Among the most vital components of translation competence, there are unquestionably not only translation and interpreting skills. As Piotrowska argues, traditional roles of a translator or a philologist have been changing, making way for graduates who act as bilingual text editors, multimedia designers, IT specialists, cultural mediators, localizers, terminologists and others (2007: 103). Hence, to satisfy the demand for specialists of this kind, there are also computer- and media-related techniques in which translators should be well-versed to rise to the challenge of, for instance, voice-over, subtitling, interviewing, interpreting performed on the Internet or telephone and so forth. Additionally, if the list of all the skills required for a successful completion of a translation is to be enriched, it should also include such abilities as proofreading, revising, editing and, last but definitely not least, self-assessment. It all serves one purpose- attaining the goal of a job well done. Needless to say, apart from the structure of a translated text, its content is what predetermines translation competence as well. Not only does translation competence require the ability to form a text with all the necessary devices used in the process of translation but also a great deal of knowledge in the field the text is written in. Bearing in mind all the skills needed to translate, a comprehensively structured list should be provided at this point. However, according to Kearns, “to attempt to systematize all of these variables in the style that is sometimes adopted by industrial human resource development models is impractical” but, instead, “what is required is an holistic approach to curriculum renewal which does not presume needs at the outset, but which identifies them as contextually (‘situationally’) dependent on manifold variables” (2008: 209). Rather than attempting to formulate a holistically binding list of the necessary skills, competences, subcomptences and the like, it is perhaps most plausible to face the fact that translation education is a solution-seeking process whereby the contingently arising problems are dealt with in an equally contingent manner; what is of paramount importance is to realize that no strategy, theory or, to be sure, knowledge can ever be assumed to be unquestionable and incontrovertible. Getting back to the terminological disambiguation undertaken at the beginning of the present study, a closer look must be cast on the term teaching. If translation education is a process of perpetual search for better ways of facilitating the students’ knowledge and ability, the term teaching seems to be far too insufficient as it is associated with not much activity on the part of students. Kiraly is one of the scholars who opt for the analysis not of the transmissionist teaching process but rather the learning process. He observes that “the teacher assumes a variety of roles to allow learner participation in a 248 PAULINA PIETRZAK variety of communicative situations” (1995: 33). According to Kiraly’s social constructivist approach, “instead of filling them with knowledge, teachers should serve as guides, consultants and assistants who can help set the stage for learning events” (2000: 18). Having assumed a variety of roles, the teacher is no longer just a teacher. Translation, as a complex process to be familiarized, requires practical experience which makes students educated enough to perform it. This student-oriented approach seems to be reflected in the term trainer or in the process of training since it involves this practical aspect of being educated, being trained. The term trainer is, in general lexicon, defined as “a person who trains people or animals” and the verb to train means to “teach (a person or animal) a particular skill or type of behaviour through regular practice and instruction” or “be taught in such a way” (1999: 1520). In this definition, trainers, on the one hand, seem to be rather instructional, supervisory and apparently a bit automatic but, on the other hand, they need to allow trainees to act and thus gain experience; they must instruct and then withdraw and be there, in the background, just to provide feedback and more instructions. Certainly, this practice is more student-oriented than teaching and thus used by Dodds (1999), Bowker (2003), Pym (2003), Kiraly (2005), Tennet (2005), Kelly (2005) but none of them sticks only to this term. The question should be posed whether there is any difference between the two terms, more particularly as the terms tend to be used interchangeably. Goad states that “one of the primary differences [between the terms] is that training takes an adult approach to learning theory, recognizing the differences between the traditional way children learn and how adults learn” (2010: 58). Stereotypically, this would suggest the application of a more serious approach, strict discipline and mandatory consequences but what is suggested here by Goad may be analysed in a few categories he mentions: •Knowledge. This is largely discovered in training rather than passed on, as in teaching. Rather than have the teacher tell learners what the “truth” is, the trainer facilitates learners’ discovery of the truth. •Improvement. Teaching emphasizes technical advancement […] Training emphasises improvement through developing the people. •Orientation. The teaching approach is oriented toward the teacher; the training approach is oriented toward the learner. •Behaviour. Teaching is concerned with measurable behaviour. So is training, but it also cares about attitudes. Positive reinforcement is a boon to trainers. •Objectives. Teaching focuses on precise behavioural objectives, emphasising the acquisition of information. Training is also concerned with precision but adds an emphasis in interpersonal skills competencies (learning how to learn). •Content. Teaching relates content to technical skills. Training adds personal skills, such as decision making and critical thinking. •Methodology. Teaching is subject oriented, perhaps emphasising the principles toward the end of the list of sixteen learning principles, and employs the lecture approach. Training, more personalised and geared to the situation, stresses more involvement on the part of the learner (2010: 58, 59). Teachers are therefore perceived as more orderly and finite in terms of the job performed, while neither performing translation nor teaching it can be treated as a fixed, learnable set of rules that translators apply. Indeed, when the aforementioned variety of roles that the teacher or trainer assumes is taken into consideration, neither of these terms appears to suffice. Dodds stipulates that a good teacher of translation: TRANSLATION AND MEANING, PART 10, 2013, 245 - 251 249 “is or should be the young, inexperienced trainee’s first friend and guide, even though the trainer/trainee relationship does not always seem to be one of friendship. But it is within this relationship that the inexperienced learn the basics of their profession, they learn safety first and only then start thinking about possibly facing the hazard of changing the words and why and when and how to do so. It should not be forgotten either that professional translators have a job to do, they have to be taught how to do it well, quickly and efficiently with minimum risk for maximum output” (Dodds, 1999: 64). This “job to do” pertains to the “particular skill” which is taught in the process of training, at least according to the dictionary definition already mentioned. Hence, there is an amount of teaching first, and later when translation trainees become more aware, there needs to be an amount of training. Yet are these two terms, teaching and/or training enough to describe such a complex process encompassing both of them? Training, as mentioned above, is also included in the process, so it should also be mentioned in a general term for this processes. Thus probably, both teaching and training included, the discipline should be referred to as translation teaching and training. Certainly, the whole process of designing and conducting the course in translation should not be limited to these two separate terms, either teaching or training, as it involves both of these activities. The process is a complex progression from teaching, which is more theoretical and teacher-oriented, to training, given that it is based not on examples but on students’ own experience, which constitutes the more practical aspect of the translation competence development on a course in translation. There is a high correlation between the term chosen to discuss a person teaching translation together with the process itself and the definition of what translation is. Should it be decided that translation is a skill, the teacher should rather be called a guide or a trainer; if translation is regarded as an art, then the better term to refer to is master; if it is just an activity it may be an instructor or facilitator, and when translation is considered to be a science, the translation teacher is a teacher or a lecturer. These terms render translation a lifeless concept while, as a subject being taught, it should be considered from the perspective of students or apprentices or trainees or disciples and so forth. Thus, translation is no longer a skill but a skill to be taught, not an art but an art to meet and engage with, not a science but a new scientific challenge. From this angle, translation must be treated as a challenge for those who are taught or trained since it is translation with all its tricks that is to be grasped by them. In the light of such a wide range of terms to use in order to delineate the common path, a more general term must be introduced; a term referring to both aspects, describing both the process of teaching and the agent of this process. Having scrutinized these terms, I would venture an opinion that the whole process could be subsumed under one term, namely translation education which encompasses both teaching and training. There are critics of the opinion that the term educator is too general; for instance, Sin-Wai clarifies that the term translation educator is a “collective term for anyone involved in translation education, including curriculum planners, textbook compilers, and translation teachers” (2004: 249). Education, however, is best described as a broad process, focused on an individual’s development with a general background provided, which is advocated by Chau (2001), Kiraly (1995, 2003, 2005), Sofer (2006). Training may be more practical, specific and short-term but the ability to translate definitely cannot be called just a skill to acquire quickly during an intensive, goal-oriented course. It takes time and requires more theoretical and thorough approach, additionally combined with all the sound aspects of training. The terms teacher and trainer are so widely used by translation scholars that it is hardly feasible to eschew these two terms or enforce and accept only the term educator, most telling but also unappreciated. The most reasonable solution would be to use these two terms as well but to apply them only in specific, justifiable contexts as they apparently differ depending on the perspective. The suggested delineation may therefore be presented as follows: 250 PAULINA PIETRZAK The distinction between teaching and training depending on the particular context of translation education discussed process-oriented approach of teaching translation • organizing a course • identifying market demands • planning the aims of the course • choice of texts to translate • designing translation activities • adopting a particular approach to correction of translations • implementation of activities and assessment planned product-oriented approach of training translators • recognizing translation trainees’ needs • identifying skills necessary to develop • developing students’ translation competence • practicing translation activities with students • assessing students’ translations • assessing translation trainees’ progress • identifying progress in students’ translation quality. To recapitulate, taking the aim of educating, conducting a course and focusing on its goal and outcomes, a perfectly valid idea would be to refer to teaching translation; whereas, when using the more personal aspect of being educated and discussing students’ personal needs or outcomes, then training translators may be used as an equally pertinent term. References Adab, B.J. (1996), Annotated Texts for Translation. Clevedon: Multilingual Matters. Beeby, A. (2004), “Language Learning for Translators: Designing a Syllabus”, in Malmkjær, K. Translation in Undergraduate Degree Programmes. Amsterdam/Philadelphia: Benjamins, pp. 39-65. Bowker, L. (2003), “Corpus-based Applications for Translator Training: Exploring the Possibilities”, in Granger, S., Lerot, J. and Petch-Tyson, S. (eds.) Corpus-based Approaches to Contrastive Linguistics and Translation Studies. Amsterdam: Rodopi, pp. 169-184. Chau, S.S.C. (2001), “Translating for the ‘New Age’: Theoretical, Professional and Pedagogic Implications”, in Sin-Wai, Ch. (ed.) Translation in Hong Kong: Past, Present and Future. Hong Kong: Chinese University of Hong Kong, pp. 219-226. COD, The Concise Oxford Dictionary (1999). New York: Oxford University Press. Dodds, J.M. (1999), “Friends, False Friends and Foes or Back to Basics in L1 to L2 Translation”, in Anderman, G., Rogers, M. (eds.) Word, Text, Translation: Liber Amicorum for Peter Newmark. Clevedon: Multilingual Matters, pp. 56-65. Gile, D. (1995), Basic concepts and models for interpreter and translator training. Amsterdam/ Philadelphia: Benjamins. Goad, T.W. (2010), The First-Time Trainer. New York: American Management Association. González Davies, M. (2004), Multiple Voices in Translation Classroom. Amsterdam/Philadelphia: John Benjamins. González Davies, M. (2005), “Minding the Process, Improving the Product: Alternatives to Traditional Translator Training”, in Tennet, M. (ed.) Training for the New Millennium: pedagogies for translation and interpreting. Amsterdam/Philadelphia: Benjamins, pp. 67-83. Hatim, B. (2001), Teaching and Researching Translation. Harlow: Pearson Education. Hatim, B., and Mason, I. (2007), Discourse and the Translator. London and New York: Longman. Kearns, J. (2008), “The Academic and the Vocational in Translator Education”, in Kearns, J. (ed.) Translator and Interpreter Training. London: Continuum International Publishing Group. Kelly, D. (2005), A Handbook for Translator Trainers: A Guide to Reflective Practice. Manchester: St. Jerome Publishing. TRANSLATION AND MEANING, PART 10, 2013, 245 - 251 251 Kiraly, D. (1995), Pathways to Translation: Pedagogy and Process. Kent: Kent State University Press. Kiraly, D. (2000), A Social Constructivist Approach to Translator Education. Manchester: St. Jerome Publishing. Kiraly, D. (2003), “From Instruction to Collaborative Construction: A Passing Fad or the Promise of a Paradigm Shift in Translator Education?”, in Baer, B.J., Koby, G.S. (eds.) Beyond Ivory Tower: Rethinking Translation Pedagogy. Amsterdam/Philadelphia: Benjamins, pp. 3-32. Kiraly, D. (2005), “Project-Based Learning: A Case for Situated Translation”, in Clas, A. (ed.) Meta: Translators’ Journal vol. 50, no. 4, Dec, pp. 1098-1111. Kussmaul, P. (1995), Training the Translator. Amsterdam/Philadelphia: Benjamins. Munday, J. (2001), Introducing Translation Studies: Theories and Applications. London and New York: Routledge. Newmark, P. (1991), About Translation. Clevedon: Multilingual Matters. Newmark, P. (1993), Paragraphs on Translation. Clevedon: Multilingual Matters. Piotrowska, M. (2007), Proces Decyzyjny Tłumacza; Podstawy Metodologii Nauczania Przekładu Pisemnego. Kraków: Wydawnictwo Naukowe akademii Pedagogicznej. Presas, M. (2000), “Bilingual competence and Translation Competence”, in Schäffner, Ch., and Adab, B. (eds.) Developing Translation Competence. Amsterdam/Philadelphia: Benjamins, pp. 19-31. Pym, A. (2003), “Redefining Translation Competence in an Electronic Age: In Defence of a Minimalist Approach”, in Clas, A. (ed.) Meta: Translators’ Journal vol. 48, no. 4, Dec, pp. 481-497. Robinson, D. (2003), Becoming a Translator: An Introduction to the Theory and Practice of Translation. Oxon: Routledge. Schäffner, Ch. (2000), “Running Before Walking? Designing a Translation Programme at Undergraduate Level”, in Schäffner, and Ch., Adab, B. (eds.) Developing Translation Competence. Amsterdam/ Philadelphia: Benjamins, pp. 143-156. Sin-Wai, Ch. (2004), A Dictionary of Translation Technology. Hong Kong: Chinese University Press. Sofer, M. (2006), The Translator’s Handbook. Maryland: Schreiber Publishing Rockville. Tennet, M. (2005), Training for the New Millennium: pedagogies for translation and interpreting, Amsterdam/Philadelphia: Benjamins. Wills, W. (1982), The Science of Translation: Problems and Methods, Tübingen: Narr. 252 TRANSLATION AND MEANING, PART 10, 2013, 253 - 262 253 TRANSLATIONAL COMPETENCE AND THE NATIVE SPEAKER Pius ten Hacken Swansea University, United Kingdom Abstract: In the translation industry throughout Europe, the assignment of work to translators is dominated or at least heavily influenced by what I will call the Native Speaker Principle (NSP) in (1). (1) Native Speaker Principle (NSP) A translator must be a native speaker of the target language. In this paper, I will take as an example the language pair Polish-English. According to (1), only native speakers of English should be given work in this language pair with English as the target language. I will argue that (1) is a fairly arbitrary restriction that is not supported by theory. In section 1, I will review the reasons why (1) is widely accepted. In section 2, I will analyse the notion of native speaker. In doing so, I will use some notions from the general framework of generative grammar. However, only minimal assumptions will be made and they will be stated explicitly. In section 3, I will turn to translational competence and relate it to the model of language use described in section 2. Finally, section 4 summarizes the conclusions. 1. Motivation for the Native Speaker Principle In Europe, languages are often connected to nations. Since the 19th century, nationalism has in many cases centred around a single language. This is certainly the case for Germany and Italy, where the political unification was fed by a sense of unity based on the feeling of having a shared national language. It also applies to the states emerging in the Balkan region, as Mackridge (2009) documents for Greece. A more recent example is the the emergence of Serbian and Croatian from what until the breakup of Yugoslavia used to be Serbo-Croatian, cf. Greenberg (2005). Even if multilingualism is widespread throughout the world, the link between language and nation made monolingualism the standard. Further languages can be acquired as a ‘foreign language’ at most. Let us now consider the implications for translation. Translators need of course to have a good command of both the source language (SL) and the target language (TL). The standard view that people are monolingual implies that translators have learned (at least) one of the SL and TL as a foreign language. In our example language pair, the choice is then whether we take a native speaker of Polish or of English to translate texts from Polish into English. A native speaker is taken to have a better command of the language, in particular where it comes to intuitions about the fine-grained aspects of meaning. The question is then whether the additional level of knowledge that a native speaker has is more important in decoding the source text or in encoding the target text. The adoption of the NSP in (1) points to a general choice for the latter. This tradition is much older than the formulation of the Skopos Theory by Reiß & Vermeer (1984), but their emphasis on the target text in the evaluation of translation seems to provide further support for the NSP. Anecdotal counterexamples to the NSP often take the form of cases where people become less fluent in their native language after living in another speech community for a long time. A Polish woman marrying a British man and moving to Britain will at some point be better at translating into English than into Polish. The most common reaction among translation agencies to this scenario is to make the NSP stricter, as in (2). (2) Native Speaker Principle (modified version) A translator must be a native speaker of the target language, living in an area where this language is generally spoken. 254 PIUS TEN HACKEN There are also variants of (2) where the translator is allowed to be away from the country of the TL for a certain period of time, e.g. no more than five years. Adopting a stricter version of the NSP is generally seen as a quality measure. If we broaden our view and consider bi- and trilingual people, however, we find a first piece of empirical counterevidence. A bilingual person is someone who has grown up with two languages, for instance a British-born daughter of Polish migrants. The initial motivation for the NSP predicts that a bilingual speaker of English and Polish will be a better translator than a monolingual speaker of English who learned Polish as a foreign language. The bilingual speaker will have all the advantages of the monolingual speaker in formulating the English target text, and in addition have a better access to the various layers of meaning of the Polish source text. Translator trainers generally agree, however, that bilingual speakers are often remarkably poor translators. Even if not supported by systematic experiments and quantified evaluations of translations, the robustness of this observation is interesting. The motivation for (1) or (2) leads us to expect the contrary. Therefore an explanation is asked for. 2. The concept of native speaker In this section, I will analyse what it means to be a native speaker of English. There are three aspects to this concept. First, native points to language acquisition in childhood. This is addressed in section 2.1. Then, speaker indicates language use. In section 2.2, I will set up a basic model of language use, relating it to the various types of underlying factors. Finally, English is an example of what I will call here a named language. This concept is discussed in section 2.3. The conclusion in section 2.4 summarizes how native speakerhood can be defined. 2.1 Language acquisition The expression native speaker is misleading in the sense that a newly-born baby is not a speaker. The process of language acquisition takes several years. In these years, two things happen to the child. First, it grows. This involves physical and cognitive development. Secondly, the child gets linguistic input from the environment. This determines which language the child will acquire. A Polish child adopted by monolingual English-speaking parents in Britain will grow up speaking English, not Polish. There is a strong case for the existence of a language faculty. Jackendoff (1993) formulates it very accessibly and convincingly. However, for our purposes it is not necessary to accept that there is a function-specific, genetically determined language faculty. The same argument could be made with the assumption that the human properties necessary for language acquisition do not constitute a language faculty, but are distributed over various components of the human mind. What is essential is that the Polish child adopted by English-speaking parents in Britain will become a speaker of English, whereas their cat will not. There is a large amount of empirical evidence for this. Whereas the idea of native speaker is not literally based on birth, it is still linked to age. If the child grows up in Poland with Polish-speaking parents and comes to Britain as a student, they will not become a native speaker of English. Children have a way of learning language that is different from how adults go about this task. There are various factors involved in this difference. First, a two-year old will develop general cognitive abilities at the same time as a language. An adult will have these cognitive abilities in place. Second, an adult will already know a language when starting to learn a new one. Arguably, both these factors should make it easier for adults to learn a new language. That children are nevertheless more successful is often attributed to the so-called critical period. A human being has a certain period in which to learn a (first) language and if that period is not used, it will be irrecoverably lost. This happens, for instance, when hearing-impaired children grow up without getting any linguistic input, cf. Mayberry (1994). For a discussion of the critical period in neurosci- TRANSLATION AND MEANING, PART 10, 2013, 253 - 262 255 ence in general, cf. Hensch (2005). Ten Hacken (2007: 290-93, 307-09) gives a brief overview of the discussion in the context of first and second language acquisition. Without taking any definite position on issues such as the speed of the deterioration of the receptiveness for linguistic input, the homogeneity of this process among a population, and the precise influence on second language acquisition as opposed to first language acquisition, we can state a first tentative definition of native speaker in (3). (3) A native speaker of a language L is someone who completed the acquisition of L before the end of the critical period. The definition in (3) is only tentative and leaves a number of issues open. One of these issues is the precise status of L. For our purposes, we are interested in the case where L is English or Polish. English and Polish are, what I will call, named languages. Before addressing the question of the status of named languages, however, we need to clarify and analyse what happens when language is used. 2.2 Performance A fundamental distinction in the discussion of the nature of language is the one between competence and performance. This distinction was introduced by Chomsky (1965) in a famous passage quoted in (4). (4) We thus make a fundamental distinction between competence (the speakerhearer’s knowledge of his language) and performance (the actual use of language in concrete situations). [Chomsky (1965: 4)] In its original context, (4) was a part of an argument intended to show that corpus-based linguistics was not the only possible way of studying language synchronically. In particular, a corpus will contain errors and judgements based on competence are necessary to identify these errors. This is probably the reason why Chomsky & Halle (1968: 3), when explaining the difference between competence and performance, mainly highlight “memory restrictions, inattention, distraction” as factors that characterize performance. In the context we are dealing with here, performance is arguably an issue of central concern. In translation, the target text produced by the translator is an instance of TL performance. In language acquisition, the data used as input by the child are also performance data. In trying to explain performance, we only make very limited progress by invoking the negative factors mentioned by Chomsky & Halle (1968: 3). They only determine how the intended performance differs from the actual performance. It is therefore worth considering which factors determine the intended performance. Much of the criticism of Chomsky’s notion of competence is directed against the fact that grammatical competence is not the only type of knowledge necessary for an adequate performance. On this basis, Hymes (1971) proposed to replace the concept of grammatical competence by a new concept of communicative competence. Chomsky (1980) argued that, instead of a conflation of the various types of competence underlying performance into one type, we should rather aim to analyse the different types of knowledge involved. He proposed a concept of pragmatic competence (1980: 224), which interacts with grammatical competence to produce performance. Kasher (1991) elaborated this concept in more detail. The easiest way to see the nature of the interaction of the various factors in performance is to consider what they result in when added one by one. If grammatical competence alone determined performance, we would go around producing random grammatical sentences. This is the basis of Hymes’s (1971) observation. When we add pragmatic competence, we can account for the fact that what we 256 PIUS TEN HACKEN say or write is appropriate for the situation it is said in. However, if no other factors than grammatical competence and pragmatic competence were involved in determining performance, we would react like robots, whose reaction is determined entirely by the situational and linguistic input. In fact, another factor contributing to performance is free will, which explains that we can choose whether and what to say or write. Together, these three factors determine the intended performance. Constraints of the type mentioned by Chomsky & Halle (1968: 3) explain why the intended performance may differ from the actual performance. 2.3 Named languages In speaking about translation, we say that a text is translated from one language, e.g. Polish, to another, e.g. English. Polish and English are languages. The notions of Polish and English are not at all as straightforward as we might think. I will call them named languages here. In considering the nature of named languages, it is first of all obvious that English is different from competence. Whereas competence is individual to a particular speaker, English is the language of a speech community. Similarly, it is clear that English is not a performance. A performance is temporally and spatially limited in a way that English is not. These observations raise the question of what kind of entity English is. Uriagereka (1998: 27) states that “English doesn’t really exist.” This statement is provocative and plays on a precise, terminological sense of exist. Whereas competence and performance are natural, empirical entities, named languages are not. Competence is the natural result of language acquisition and performance is the natural result of the use of competence. Named languages can only come into existence as a result of metalinguistic reflection. Competence and performance can exist and function successfully without any notion of English or other named languages being operative. In this context, Mackridge’s (2009) description of the Greek language question makes interesting reading. The whole point of this question was to determine a version of the language that could serve as a national standard. As long as there is no such standard, it is not obvious how to produce one. Therefore, different contenders each proposed and defended their own approaches to using the existing language as spoken by (different categories of) Greek people and (different varieties of) Ancient Greek as a basis for a new standard. In order to determine what is English, the first step is to determine whose competence will be classified as English. Then their judgements can be used to find the rules of English. Vaugelas (1647) describes this method in very much these terms when he explains how he determined what was good French. Of course not only competence but also performance can be classified as (good) English or not. Classification and evaluation go together here. Given that named languages are not natural objects but derived from (grammatical) competence, they cannot be part of this competence. To the extent that they influence performance, they do so by the speaker’s knowledge of the standard. This knowledge is part of the pragmatic competence, along with conventions about, for instance, the appropriate topics for small-talk. 2.4 Conclusion The purpose of this section was to come up with a clear concept of native speaker. On the basis of the discussion of the critical period in language acquisition, we tentatively defined it as in (3). However, this definition depends on an entity corresponding to the named language L. Given the nature of named languages as it emerged from the discussion in section 2.3, (3) can no longer be maintained. Instead, we should rather define it as in (5). In order to avoid overly complex formulations, (5) defines native speaker of English rather than native speaker as such. TRANSLATION AND MEANING, PART 10, 2013, 253 - 262 (5) 257 X is a native speaker of English iff X’s competence is classified as close enough to English. A first expression to highlight in (5) is “close enough”. Named languages are not natural entities, so we should not expect anybody’s competence to be equal to English. This raises the issue of how to determine native speakerhood. There are two main criteria, performance and linguistic biography. Language testing centres all over the world have developed a wide range of performance-based tests to assess someone’s level of competence in a language. These tests normally take as their anchor point a native speaker standard. Linguistic biography as a criterion returns us to (3). Crudely speaking, if X was in an English-speaking environment during the critical period, X is a native speaker of English. This criterion is much easier to apply, but less reliable than performance-based criteria. First of all, it depends on the classification of an environment as English-speaking. In general, the quality of the input a child gets is judged rather crudely on the basis of geographical parameters. Secondly, the critical period is a notion that is hardly understood precisely enough to figure in a reliable procedure. Apart from being poorly understood, the critical period is also not directly involved in the acquisition of named languages. As we concluded in section 2.3, knowledge of a named language is part of the pragmatic competence. Whereas there is good evidence for a critical period in first language acquisition, this evidence only concerns grammatical competence. In the acquisition of pragmatic competence, learning (as opposed to unintentional, automatic acquisition) plays a much larger role. The critical period concerns in particular the natural acquisition, not the aspects of language that have to be learned more explicitly. This explains also that second language acquisition is possible after the critical period. Therefore, the conclusion must be that native speaker is not a natural concept but a label for a high level of (grammatical and pragmatic) competence. The sense of language acquisition that takes place naturally does not result in being a native speaker of a named language. Someone’s competence in English is the result of the interaction between their grammatical competence and their knowledge of the English standard.1 3. Translational competence The concept of translational competence is not often used in the context of linguistic theory. When we consider the types of competence involved in translation, translational competence includes those aspects of competence that are necessary for a good translation but cannot readily be attributed to the SL or to the TL. In order to approach the question of the nature of translational competence, it is worth taking as a basis Holz-Mänttäri’s (1984) model of translational action. Fig. 1 gives the six roles she distinguishes, using Munday’s (2001: 77) translations. 1 As Lüdi & Werlen (2005:7) state, in the census held in Switzerland every ten years, the concept of Muttersprache (‘native language’) was replaced by Hauptsprache (‘main language’) after 1980. The relevant census question in 2000 was “Welches ist die Sprache, in der Sie denken und die Sie am besten beherrschen?” (‘What is the language in which you think and which you have the best command of?’). This reflects a similar scepticism to the concept of native language. 258 PIUS TEN HACKEN Fig.(1): The model of translational action In Fig. 1, roles are represented as rounded rectangles and documents are rectangles. There are three types of arrows. The wide arrows, such as the one from ST producer to ST, represent text production. The pointed arrows, such as the one from ST to TT producer, represent text reception. The small arrow from initiator to commissioner represents communication of an unspecified nature. In Fig. 1 it is assumed that the translation brief is an actual document. Depending on the context of translation, the specifications of the translation job may also be less explicit. As an example, let us assume that the ST is a brochure in Polish about the touristic sites and accommodation in the województwo of Łódź to be translated into English. The ST producer is the author of the Polish original, the TT producer the Polish-English translator. The initiator is, for instance, the publicity officer of the województwo and the commissioner the project manager in the translation agency that gets the job. Depending on the translation brief, the TT user may be, for instance, someone at the Polish stand of a holiday fair, whereas the TT receiver would be, in that case, a visitor of such a fair. In analysing translational competence, the best way to proceed is to compare the types of competence involved in producing the ST and the TT. In section 2.2, we concluded that the intended performance is the result of the interaction of three factors, linguistic competence, pragmatic competence, and free will. I will ignore here the difference between intended and actual performance. For published texts offered for translation and for translations themselves, the degree of monitoring and proofreading minimizes the distortion compared to casual spoken language. Of the three factors involved in performance, pragmatic competence is most in need of elaboration. In our example of the tourism brochure, the ST is written in Polish, presumably for Polish tourists and by a Polish author. The author’s pragmatic competence has the function of taking into account what this readership expects, what background knowledge it has, etc. As argued in section 2.3, also knowledge of the standard language, i.e. the normative rules of Polish grammar and vocabulary use, is part of the pragmatic competence. The TT is written in English by a translator. The target readership is determined by the translation brief. It may be British visitors of holiday fairs, but could well be wider than this, including Americans or Dutch and Scandinavians. The intended target readership will determine the variety of English and the type of language use that is acceptable. Clearly, the choice between British or American English spelling and vocabulary is a matter of knowledge that should be part of the translator’s pragmatic competence. It may be useful to avoid words that are restricted to one variety if both readerships should be addressed. If Dutch and Scandinavian readerships are also targeted, ex- TRANSLATION AND MEANING, PART 10, 2013, 253 - 262 259 pressions that are hard to understand for these groups should be avoided. This may even involve knowledge whether certain idioms in these languages use the same images as their English counterparts. However, all of this knowledge is ideally also part of the pragmatic competence of an English text writer who writes original texts for such purposes. Unsurprisingly, the main difference between writing the ST and writing the TT is that the writer of the latter has the ST as input. Whereas the sections in the translation brief specifying the intended target readership may have a counterpart in the specification of the purpose and readership of the ST, the Polish ST provides the translator with a degree of detailed input specification that is not comparable to that available to the author of the ST. This is the reason why translation is considered a fundamentally different activity to authoring a text. The extra information available to the translator comes in the form of a text (the ST), rather than in the form of explicit instructions. Therefore, compared to writing the Polish ST, writing the English TT involves two extra types of skill. The first is the skill to recognize the ST author’s intention by interpreting the text in relation to linguistic and cultural background assumptions. In a sense, the translator has to reverse-engineer the influence of the ST author’s pragmatic competence on the ST. Questions that arise include how much of the ST is determined by the linguistic and cultural rule system and how much constitutes the author’s choice. If the ST author had different options for expressing something, the translator can infer aspects of the author’s intention by analysing the possible reasons leading to this choice. To some extent this process is an inherent part of any type of communication, as highlighted in Sperber & Wilson’s (1995) Relevance Theory, but in the case of translation it tends to assume a special significance. The second additional type of skill in writing the TT is resolving conflicts between competing constraint systems. An example of the type of conflict that arises in our example is the use of articles and demonstratives. In Polish, there are no articles. In cases where English requires a particular article, there is no real problem. In many contexts, however, English offers the choice between a definite and an indefinite article or the option of leaving out the article entirely. Whereas in Polish the contrast can be resolved by underspecification, English forces the translator to choose one of the three options. A similar problem arises in the translation of Polish demonstratives. Because Polish does not have articles, demonstratives are used more frequently than in English. This is not a matter of the grammatical system, but of the preferred use. Therefore, in the translation of Polish demonstratives the translator will have to choose an article or demonstrative in English depending on the interpretation of the author’s intention in writing the Polish ST. Another example of conflict resolution in the scenario of tourist brochures is the use of first person plural pronouns and verb forms. In the Polish ST, we may find expressions such as “our country” referring to Poland, using inclusive our. In writing the TT, “our country” can be reinterpreted as involving exclusive our, but the use of inclusive and exclusive first person plural pronouns is subject to different constraints. Using (exclusive) our where the ST used a first person plural interpreted implicitly as inclusive our changes the tone of the TT, which may or may not be intended or desirable. An issue that is particularly complex for tourist brochures is the amount of background knowledge that can be assumed. For a British readership, the geographic location of Łódź with respect to Warsaw, its approximate size, its industrial history, etc. will not be common knowledge, whereas for a Polish readership they are. The translator will have to decide how to deal with the conflicts this triggers between the constraints of translating the ST and of creating a meaningful TT for the intended readership. A German translation, or an English translation used in Germany, should probably mention the German name Lotsch, if only because it occurs in a well-known German song, but there is no point in doing that for a British (or indeed for a Polish) readership. 260 PIUS TEN HACKEN What these examples illustrate is that translational competence involves two aspects, formulated in (6). (6) a. b. Recognizing which information should be interpreted as constraints on the TT to be written. Strategies for resolving conflicts between constraints of different sources. The information referred to in (6a) includes the interpretation of the ST in relation to the SL rules, conventions of language use, and cultural expectations, the knowledge of the TL rules, conventions, and culture, and the knowledge of the target readership’s background knowledge. It is interesting to note how Reiß & Vermeer’s (1984) Skopos Theory approaches the conflict resolution in (6b). In view of the model in Fig. 1, Skopos Theory highlights the needs of the initiator, commissioner, TT user, and TT receiver. In this way, the TT is taken primarily in its role as an original. For an original tourist brochure written in English, the author should also consider these four roles. Considerations of faithfulness to the ST, which make translation different from authoring, are placed at a lower level of priority in Skopos Theory. They are only used to select expressions within the space determined by the needs and expectations of the initiator, commissioner, TT use, and TT receiver. If we consider the TT as the performance produced by a translator, translational competence as characterized in (6) is a specialization of pragmatic competence. The translator uses the SL competence to understand the ST, the TL competence to produce a grammatical TT, and the translational competence to balance the different constraints arising from all sources of relevant knowledge. Arguably, by reducing the role of the ST, Skopos Theory makes translational competence less different from regular pragmatic competence. 4. Conclusion Our starting point was the question to what extent the Native Speaker Principle in (1) is supported by insights from linguistic theory. Sections 2 and 3 raised problems for the NSP from two different angles. In section 2, we saw that native speaker is a rather poorly defined concept. There are two possible interpretations correlating with different interpretations of the concept of language. In the first, everyone is a native speaker of the language corresponding to their linguistic competence. Linguistic competence is a part of the speaker’s mind, hence personal. In the context of translation, the translator’s linguistic competence is of course of eminent importance, but because of its individual nature it cannot serve as a standard for determining whether the translator is a native speaker of the TL. The TL is no person’s competence. In order to apply the NSP, we need to invoke the second interpretation of the concept of language, where language is interpreted as a named language, e.g. Polish or English. Named languages are not the product of a natural process. In order to set up a named language, the linguistic competence of individual speakers is classified into groups with a certain degree of similarity. Decisions about the criteria for similarity are necessarily subjective. The decision who has the authority to apply these criteria therefore becomes crucial. Given the artificial nature of named languages, the knowledge of their ‘proper use’ is not of the same type as naturally arising competence in language acquisition by children. Therefore, using native in this context is misleading. In section 3, we saw that translational competence can be seen as a special component of pragmatic competence used in writing a TT. It consists of information collection and conflict resolution skills that require special training. Therefore, there is a rather long chain of reasoning hidden in the formulation of the NSP in (1). TRANSLATION AND MEANING, PART 10, 2013, 253 - 262 • • • 261 There is a qualitative difference between naturally arising competence and knowledge of named languages. Explicit rules play a much more important role in the latter than in the former. Knowledge of these explicit rules is part of pragmatic competence. This knowledge is to a much greater extent the result of explicit learning rather than natural acquisition. The concept of native speaker relies on a rather simplistic view of the critical period. There is no absolute, uniform critical period for language acquisition, after which no new language can be learned. Moreover, the critical period concerns grammatical competence. Pragmatic competence, as well as lexical competence, are much less sensitive to the critical period. Translation quality depends not only on SL and TL competence, but also on translational competence, involving skills of inference and conflict resolution. There is no obvious relation between native speakers and this kind of skills. Translational competence is a specialized set of skills that has to be learned separately from linguistic competence. In view of these observations, there is no linguistic motivation for the NSP as formulated in (1). Whereas the variant in (2) addresses some obvious counterexamples to (1), it does not address the main problems for the NSP. These problems stem from the incongruity between the concepts of native speaker and of named language and from the lack of attention for the nature of translational competence. In this context, the observation that bilinguals are not necessarily good translators can be explained straightforwardly. Bilinguals may be native speakers of two languages, but this does not give them privileged knowledge of the relation of these languages to the respective norms and the need to acquire translational competence is the same for them as for other translators. The NSP assumes that someone’s linguistic biography is a reliable guide for predicting the quality of their translations. Even in a neatly organized world in which every country has its own national language and nobody ever migrates to another country, the NSP has only a very limited predictive force, because it neglects translational competence. Moreover, the idealizations made in the NSP are not theoretically justified. People with special linguistic biographies are overrepresented in the translation profession. With current trends of increasing mobility and international contacts, the number of cases of discrepancy between the expectations based on the NSP and the actual performance of a translator will only increase. In the light of these observations, it is interesting to note that the European Union translation services have adopted a system entirely based on performance. According to the frequently asked questions page at http://europa.eu/epso/discover/faq (28-12-2010), the “main language” or “language 1” is “most likely to be your mother tongue” and “[c]andidates found by the selection board to have an inadequate knowledge of “language 1” will be excluded.” The selection board is not interested in the linguistic biography, but bases its judgement entirely on the performance in the concours. This means that the biggest translation service in the world does not use the NSP. In sum, the NSP is not a guarantee for translation quality. The widespread use of the NSP excludes large numbers of (potential) translators with less straightforward linguistic biographies. Therefore, the continued use of the NSP in the translation industry is harmful both to individual translators and to the industry as a whole. References Chomsky, N. (1965), Aspects of the Theory of Syntax. Cambridge (Mass.): MIT Press. Chomsky, N. & Halle, M. (1968), The Sound Pattern of English. New York: Harper & Row. Chomsky, N. (1980), Rules and Representations. New York: Columbia University Press. Greenberg, M. (2005), ‘Serbo-Croatian and South Slavic Languages’, in Strazny, Philipp (ed.), Encyclopedia of Linguistics (2 vol.), New York: Fitzroy Dearborn, pp. 956-958. ten Hacken, P. (2007), Chomskyan Linguistics and its Competitors. London: Equinox. 262 PIUS TEN HACKEN Hensch, T.K. (2005), ‘Critical Period Plasticity in Local Cortical Circuits’, Nature Reviews Neuroscience 6:877-888. Holz-Mänttäri, J. (1984), Translatorisches Handeln: Theorie und Methode. Helsinki: Suomalainen Tiedeakademika. Hymes, D. (1971), ‘Competence and Performance in Linguistic Theory’, in Huxley, Renira & Ingram, Elisabeth (eds.), Language Acquisition: Models and Methods. London: Academic Press, pp. 3-24. Jackendoff, R. (1993), Patterns in the Mind: Language and Human Nature. New York: Harvester/ Wheatsheaf. Kasher, A. (1991), ‘Pragmatics and Chomsky’s Research Program’, in Kasher, Asa (ed.), The Chomskyan Turn. Oxford: Blackwell, pp. 122-149. Lüdi, G. & Werlen, I. (2005), Eidgenössische Volkszählung 2000: Sprachenlandschaft in der Schweiz. Neuchâtel: Bundesamt für Statistik. Mackridge, P. (2009), Language and National Identity in Greece. 1766-1976, Oxford: Oxford University Press. Mayberry, R. (1994), ‘The Importance of Childhood to Language Acquisition: Evidence from American Sign Language’, in Goodman, Judith C. & Nusbaum, Howard C. (eds.), The Development of Speech Perception: The Transition from Speech Sounds to Spoken Words. Cambridge MA: MIT Press, pp. 57-90. Munday, J. (2001), Introducing Translation Studies: Theories and Applications. London: Routledge. Reiß, K. & Vermeer, H. (1984), Grundlegung einer allgemeinen Translationstheorie. Tübingen: Niemeyer. Sperber, D. & Wilson, D. (1995), Relevance: Communication and Cognition, second edition. Oxford: Blackwell (orig. 1986). Uriagereka, J. (1998), Rhyme and Reason: An Introduction to Minimalist Syntax. Cambridge (Mass.): MIT Press. de Vaugelas, Claude Favre (1647), Remarques sur la langue françoise. ed. Jeanne Streicher, Paris: Droz, 1934. 263 Section IV: Terminology 264 TRANSLATION AND MEANING, PART 10, 2013, 265 - 273 265 THE USE OF GERMAN TERMINOLOGY AS A EUPHEMIZING STRATEGY IN HEBREW 1930-19701 Nitsa Ben-Ari The Porter School of Cultural Studies, Tel Aviv University, Israel Abstract: In this paper, I will deal with the use of German terminology in Hebrew as a euphemizing strategy. I will first describe this phenomenon in non-fiction para-medical texts such as translated sexual guides or encyclopedias. I will then dwell on the effect the use of German euphemisms had on literature and the literary repertoire, since, for lack of other options, modern Hebrew literature found itself using German-origin lexemes or derivatives thereof in fiction as well. The greatest Israeli writers are still struggling to free erotic descriptions of this old “medical” hue. 1. Introduction One of the most common ways of avoiding direct use of sexual/erotic language is euphemism, that is cleansing lexemes until they become “presentable”, their unambiguous denotative meaning substituted with one that is less “offensive”. It is the use of a mild, neutral, evasive, or vague term in place of one considered taboo, offensive, blunt, or unpleasant. Euphemisms may be regarded as a safer way of saying questionable things, but it may well be regarded as a form of censorship, mostly self-censorship. There is one aspect of euphemism in translation, however, that I find particularly intriguing, which is used to keep the discussion of the “questionable” topics strictly in the hands of those allowed or appointed to do so by society/establishment/ authority, while keeping “outsiders” away. British tradition employed Latin as a form of keeping the discussion of sexual matters in the “right” hands. In the case of Hebrew literature in pre-state and even post-state Israel, translations often used German as a way of avoiding direct use of sexual lexemes, thus euphemizing, and mystifying texts about sex, at the same time giving them a medical character. This was done in accordance with the puritan tendency of presenting an idealized model of the Sabra, the New Hebrew, as a pure creature, whose erotic drives should be “sublimated” to conquering the land, sacrificing his personal desires for the common good. The conquest of the land [adama, feminine gender in Hebrew] was eroticized in the new Zionist ethos. Euphemizing sexual lexemes was also done in accordance with a long Hebrew tradition of euphemization in the Talmud, where all aspects of sexual life were discussed in detail, yet in metaphorical language. It was done in sharp contrast, though, with ideals of the new sexual revolution (inherited from the Bolshevik revolution), which Zionism was supposed to entail. And last but not least, it was done in a period where Hebrew was being revived by linguists, lexicographers, men of letters, who refrained from dealing with the erotic lexicon. 2. Some basic terms for this research The behavior of people in culture does not rely on a set of well-organized, systematic rules, but on models, or ready-made options that dictate their behavior, according to the area they function in and to their position. My research is based on the functional dynamic approach developed by Even-Zohar that sees repertoire as a cultural model, as an imagined system of rules that produce behavior or products that people who belong, or wish to belong, to a certain habitus, will obey, not necessarily knowingly. This system can also produce distinctions concerning repertoires, provide legitimization for repertoires, means to explain them, classify and evaluate them. (Even-Zohar, 1990: 39-43; 2071 Based on a paper presented in the 5th International Maastricht-Lodz Duo Colloquium on “Translation and Meaning” Lodz, 16-19 September 2010. 266 NITSA BEN-ARI 218; also Bourdieu, 1984, 97-256 about “taste” as a cultural product). I have analyzed the phenomenon of building an erotic-free repertoire in my book Suppression of the Erotic in Modern Hebrew Literature (Ben-Ari, 2006), and in this essay I would like to describe euphemism as a particular form of censorship that screens “unwanted” readers out while, seemingly, addressing a select public. The etymology of euphemism is, as we know, the Greek euphemos (auspicious), deriving from eu (good) + pheme (speaking). How does the use of German terminology fall into the category of saying things in a more auspicious way? The questioned should be phrased differently: when is there a guarantee that using foreign terminology would enhance a text? Obviously, when two conditions are fulfilled: the foreign language is difficult/rare/ancient; b) the foreign language is prestigious or considered so by the particular sector that masters it. In the case of Latin used as euphemism all preconditions apply. In the case of German – the term 'antique' is irrelevant, to be replaced by some other epithet, such as “learned”, or accessible to a select few only. These statements will be questioned and analyzed below. 3. The non-translation strategy In a paper presented in Leuven 20092, Carol O'Sullivan spoke about “The Decent Obscurity of a Learned Language: Multilingualism as an expurgatory strategy in Victorian Britain.” She described how the British publisher Henry Bohn established the “Classical Library” in 1848 and made translation accessible to a mass readership at the then low price of three shillings and sixpence or five shillings. While doing so, he was careful to avoid clashes with the puritanical censor, which meant expurgating the texts (O'Sullivan, 2009: 107). In fact, his colleague, publisher Henry Vizetelly was prosecuted in 1880 for publishing (expurgated) translations of Zola, which led to his ruin. The most usual form of expurgation in Bohn's Libraries was leaving whole passages in languages other than English. Sections from The Golden Ass, recounting the narrator's dalliance with Milo and Pamphile's maid Fotis, and, after his transformation, with the wealthy Roman matron, were printed in the original Latin, either in the body of the text or in the footnotes (Works of Apuleius 1853: 35-36, 209-210). Passages in Ovid and Petronius were similarly treated. In several cases the translation was supplied in footnotes. Sometimes more than one foreign language was used. In Kelly's new translation of Boccaccio's Decameron, two problematic stories were partially told in the original Italian, accompanied by a French translation. A number of Epigrams of Martial were left in Latin and accompanied by Graglia's 18th century Italian translation, “in those instances where an English translation given faithfully would not be tolerable” (Bohn, 1860: 4, quoted in O'Sullivan: 114). His regular translator Walter K. Kelly remarked that a literal translation of texts such as Catullus 33 and six other poems would be “insufferable in English” (Poems of Catullus 1854:31, O'Sullivan: 115). These instances clearly come under the heading of self-censorship, and stem from the fact that publisher/ translator have internalized the norms pertinent to the culture, and [use] them as a constant monitoring device. This translation – or rather non-translation – technique had two types of target-readers in mind: the masses who were excluded, that is, could not read foreign languages, and were therefore not entitled to reading the racy passages, and the select few who were included, those who could read “learned” foreign languages. The results of this procedure were manifold. One of them was inciting the lay- 2 21 May 2009, “Translation in Multilingual Cultures”, Catholic University of Leuven. See “Translation within the Margin: the “Libraries” of Henry Bohn, Carol O'Sullivan 2009 in: Agents of Translation, John Milton and Paul Bandia (eds.), Amsterdam and Philadelphia: John Benjamins 107-130. Also available on the web. TRANSLATION AND MEANING, PART 10, 2013, 265 - 273 267 men, such as curious students, for instance, to learn Latin.3 Another more far-reaching result was the medicalization of sex, since Latin had traditionally become the doctor's jargon. One must bear in mind, though, that full unexpurgated translations of Boccaccio, Martial or Catullus did exist in Victorian England; they were expensive and printed for private circulation only, as in the case of John Payne's Decameron from 1886, or the Index Expurgatorius of Martial from 1868 (O'Sullivan: 115), but they did exist. And so did a whole repertoire of erotic language of all registers. Such was not the case in the Hebrew culture revived in the 19th and 20th century. 4. German as euphemization Euphemizing/censoring via a different language occurred in Hebrew translations of the early 20th century, with three main differences: (1) the relevant texts were non-fiction – sex manuals or encyclopedias. (2) the euphemizing/expurgating language was German, not Latin. (3) a total non-translation strategy was not an option, for Hebrew and German are not compatible, not even in fonts. Some historical background is necessary to understand how, in a period where the erotic was culturally “undesirable”, a whole set of ideological shaping of “the right taste” was construed in Hebrew culture, a distinction between “refined” and “cheap” taste in literature, side by side with a system of values that dictated a very clear set of “purity” norms. It is the key to understanding how these models were “translated” or how they were applied in texts to be produced in the center or in the periphery of that culture. A ban, legislative but mostly normative, pushed erotica to the periphery (which is more or less a normative procedure in many cultures), and the only “legitimate” niche to discuss it (or rather dissect, analyze, psychoanalyze it, much in the Foucauldian sense of the word) was in pseudo-scientific texts. In these texts, most of them translated in the early pre-state period, the euphemizing language was German, not only when the original was a German texts. The fact that we are dealing with pseudo-medical texts seems to indicate that this is where the analogy with the English/Latin case described by O'Sullivan stops. True, but only partially so, for two reasons: first, sex manuals of the period often used literary language and/or quotations from world literature. Second, when Hebrew fiction eventually yielded to the need of the erotic repertoire, it had to fall back on the sex manuals for a “modern” inventory, where it found German or Germanized lexemes and expressions. It could not fall back on the erotic repertoire developing in colloquial Hebrew. In fact, side by side with the “clinical” “scientific” inventory developing somewhat artificially in the sex guides, a “filthy” vernacular consisting mainly of loans from foreign languages or adaptations thereof was developing – independently – in spoken Hebrew; but it was unimaginable, in norms of the period, to use it in literary texts. The development of erotic repertoires has a much larger role to play in culture than is immediately obvious, for it reflects the external world’s rhetoric, activities, and didactics, with many ideas and beliefs springing from occurrences and attitudes in general culture. It is influenced by but also shapes contemporary notions about socio-sexual relations between the sexes, depicting men's and women’s hierarchical roles, existing, imagined or recommended. The regulation of sexual expression may be an integral, even expressive part of society; in fact, in puritanical societies, it may become the carrier of social meaning (Ingebresten, 2001: 23). The culture that developed in pre-state Israel did not admit to being puritanical. Quite on the contrary, it carried banners of sexual freedom and gender equality. One of its main priorities was the revi- 3 Harish Tribedi then noted that in a translation from Sanskrit, where creation/fertilization was described – though in very noble terms – the translator translated it into Latin! The visibility of it almost demanded attention. 268 NITSA BEN-ARI val and development of the Hebrew language4, a task zealously taken up by major culture shapers, men of letters, with or without formal linguistic education. Yet, the pre-state Language Committee and later the Academy for the Hebrew Language hardly contributed to the development of the erotic repertoire, due to a strong puritanical aversion to material concerning sex. The “medical” lexicon was generally confided to physicians like Dr. Mazia or Dr. Tchernichovsky5, and they refrained from dealing with vocabulary pertaining to sex. Dictionaries of the period lacked the most basic terms, which meant that translators were mostly left to figure out ways of dealing with lexical problems on their own. The lack of sexual or erotic terminology is apparent in the German-Hebrew dictionaries of the first decades of the 20th century. Neither Shulbaum’s dictionary from 1904 nor Laser-Torczyner’s from 1927 have any entries for 'penis' or 'vulva' (though, strangely enough, the popular term dagdegan for clitoris does appear there). Baruch Krupnik’s 1939 Hebrew Dictionary does not include the words pin [penis], pot [vulva] or nartik [vagina]. Neither does it include the term dagdegan. It was up to translators and innovators from the 1930s and on to find suitable solutions, since this was the time when demand for pseudo-scientific books began to increase. This ascetic attitude did not change much with the years; the Academy’s contribution to the lexical repertoire of Gynecology, for instance, was published in the late 1990s only, not necessarily the period in which the terms had been coined. The Committee for Gynecological Terminology had been active since 1989 (more than two decades ago) and the terms were coined and confirmed 27 years ago. Furthermore, etymological research on new, renewed and invented lexemes, particularly lively and flourishing in Israel in other fields, has been quite rare when sex terminology was concerned. For instance, a daring essay on the source of the colloquial zayin (cock, dick), written by two serious academic researchers – Assaf and Bartal – was rejected time and again by the puritanical linguistic journal Leshonenu La’am [literally: Our Language for the People], to be finally published, not surprisingly, in the carnival Purim issue of 1993, against the better judgement of the chief editor, who extracted his name from this issue. Shunning the sex repertoire had far-reaching consequences for, while languages/cultures with a tradition of colloquial or written repertoire had something to fall back on when puritanical periods have passed,6 not so Hebrew culture, which had some mild poetic Biblical erotica and a vast tradition of Talmudic euphemisms developed in the 3rd and 4th centuries AD.7 Moreover, with time, the few Biblical and Talmudic terms have become associated with sin. Women’s sexual parts, probably out of fear of incest, were the greatest taboo. The old term for female genital erva has accumulated negative connotations in early Hebrew literature. The prohibition to listen to women singing coined the expression “Kol be-isha erva” – a woman’s voice is obscenity, creating the analogy erva=obscene. Canonic terms remained general and literally obscure: 4 The revival officially started in pre-state Israel in 1890, with the establishment by Eliezer Ben-Yehuda and friends of the “Hebrew Language Committee”. Ben-Yehuda's life's work was his “complete Dictionary of Ancient and Modern Hebrew, published by Langenscheidt in 18 volumes between 1910 and 1959. By 1918, Hebrew had been established, along with English and Arabic, as one of the official languages in Palestine. 5 Dr. Aharon Mazia (1858-1930) was a renowned physician. Dr. Shaul Tchernichovsky (1875-1943) was a physician and an acclaimed poet. 6 Consider, for the sake of comparison, the findings in a language whose development was “normal”, or rather “normally” subdued by State and Church. The French language has a lexicon for erotica, Pierre Guiraud’s 1993 Dictionnaire érotique. It was not compiled in 1993; this updated version is based on the 1978 version, in itself based on 19th century editions. This dictionary boasts of 7000 terms, out of which more than 1300 synonyms for intercourse (coït), more than 600 terms for the female sexual organ and about 600 for the male organ. Since the terms are inter-connected, the dictionary claims to include 3000 terms only for the very narrow area of sex in various strata. It provides, apart from their etymology, important examples and citations. These testify to the richness of the French language from the Middle-Ages and until today, but also to the need to circumvent medical/technical terminology and develop a diversity of terms of all kinds: lewd, filthy, expurgating, poetic, allusive, funny, ambiguous, popular, elitist etc. 7 For instance, in order to prove adultery, so the Talmud scholars decreed, it was necessary to catch the culprit with “brush in tube” (an expression which exists in Arabic as well). TRANSLATION AND MEANING, PART 10, 2013, 265 - 273 269 the penis was referred to as the male organ (ever ha’zahrut), and terms as ezor ha’boshet, ezor ha’starim [shameful parts, hidden parts] implied shame and secrecy. Thus, the only niche in which sexual terminology could be updated and dealt with in a more or less legitimate way – the sexual manuals or encyclopedias, was not entrusted either to professionals or to eminent men of letters, but was somewhat reluctantly (to judge by the many prefaces and apologetic translator notes) taken up by mediocre unprofessional translators, who may have been “professional” in sexual matters but not in lexicography. They were “sexologists”, physicians, psychologists, pedagogues, who had immigrated to pre-state Israel from central Europe, and/or had gone to Vienna or Berlin for their studies. In fact, central-European pre-Freudian and Freudian schools predominated the cultural scene in Israel until well into the 1970s, when American texts began to penetrate it. Generally speaking it was the “social” approach to sex which dominated the local scene. The sex guides acquired a marked ideological hue in the 1930s and 1940s, with books written by and for pedagogues of the socialist youth movements. Zionist ideology, considering itself close to the Austrian Marxist line, adopted the combination Marx+ Freud, favored by Herbert Marcuse (see his Eros and Civilization from 1955). The ideology of the youth movements was ambivalent, on the one hand brandishing banners of liberty, freedom and equality of women, and on the other calling for “purity” and self-discipline. Unwritten prohibitions demanded that youth preparing for Aliyah (immigration) should abstain from sexual relations until they were allowed to immigrate. In 1.6.1929 the Warsaw Ha’shomer Ha’tzair paper featured a series of essays by the pedagogue A. Ravitzki called “On the Sexual Question of the Adolescent”. The Shomer “purity” is presented nearly as a precondition for participating in a pioneer movement (see also Almog, 1997: 342-343, translation mine, emphases mine): “Till this age the question of sexual abstention must be presented with all the severity with which we consider smoking or drinking alcohol. It must be clear that although Ha’shomer Ha’tzair does not wish to become a movement for ascetic monks, yet it has been proved beyond doubt that temporary erotic asceticism, or rather postponement of sexual life until the possible limit, is for us not just desirable, educationally, but necessary. This stems not only from the wish to fight immature relations, but it is a formidable means to forging a man’s will. A strong will, not afraid of suffering and always ready for privation and suppression of needs, this is after all the educational goal of a pioneer movement”. Ravitzky could have found the recommendation in August Forel's admonition against masturbation. The famous social sexologist Forel inserts a eulogy to the “hero” who can be “his own master” and reach the age of twenty-five without sex (Forel, 1930: 448; m.t., e. m.): “We offer the middle way, the golden way: the youth who has enough power to brake his sexual urge, or the one whose sexual urge is not as strong and allows him to live ascetically until the age of twentyfive, without falling to the prostitution offense, without using substitute sexual relations and masturbation – this youth is sure to have the upper hand in life”. What were then the prototypes of the popular sex guides? They were translations of (a) mostly books well accepted, if not canonized, in Europe. (b) they were, consequently, not controversial. (c) they did not preach the “joys of sex” (such as Havelock Ellis's 7 volume Psychology of Sex, published between 1897 and 1910). (d) they were preferably books that verbalized the clinical, sociological and pathological aspects of sex. (e) they appeared in anonymous publishing houses bearing pseudo-scientific names like the Hadassa Health House, Jerusalem. Such was in fact August Forel's Die Sexuelle Frage, published in 1905 and translated into Hebrew in the early 1930s as Ha'sheela Ha'minit (literal translation). Such was Dr. Fritz Kahn (or Cohen) guidebook translated in 1962 as Ha’higiena Ha’minit shel Ha’naar [The Sexual Hygiene of the Boy] and its sequence Ha’higiena Ha’minit shel Ha’naara [The Sexual Hygiene of the Girl] translated in 270 NITSA BEN-ARI 1935. Such was Van de Velde's book The Ideal Marriage from 1926, translated around the same time (no date) as Misterei ha'nisuin [The Mysteries of Marriage]. With time, more modern solutions were found, so that Costler & Willy's book from 1934 The Encyclopedia of Sexual Knowledge,8 translated in 1946 as Encyclopedia le madaey ha'min [Encyclopedia for the Sexual Sciences (!)], sounds much less German, and Fritz Kahn's second book Unser Geschlechtsleben from 1937, translated in 1962 as Chayenu Ha'miniyim [Our Sexual Life] relies even less on German lexemes. Let me bring a few examples to this Germanizing tendency from the earlier translations. In Henrik van de Velde's Mysteries of Marriage9 the solutions for sexual terminology are as follows: 1. Hebrew neologisms with no explanation (most of them not adopted eventually): beitziot for testicles 2. Hebrew with Hebrew explanation: pota – inner sexual part of the female 3. Hebrew calques with foreign synonyms: hizdakrut ha'zera be'term zman Praecox) (Ejaculation 4. Foreign words left in the original, with or without foreign translation: Gentlemen prefer blonds, Blondine wird bevorzugt vollschlank, 5. Foreign words in transliteration with Hebrew explanation: libido, ereqtzia, orgasma, ejaqulatzia, impotentzia, defloratzia 6. Transliteration without Hebrew explanation: sexuologia, deqolteh 7. Foreign words in transliteration explained with foreign terms 8. Hebrew phrases in literal translation from German: teva (die Nature) in feminine gender, although it is masculine in Hebrew. The result is often a compilation of transliterated words which is hardly legible: infantilism ke'inormaliyut constitutzyonalit August Forel's book from the turn of century is an excellent example for the kind of social/clinical approach to sex, diseases and perversions, as well as an example for the Germanic solutions. I first chose four pages (45-50) in the translation, describing the human body, to analyze the solutions thematically. 1. Thirteen out of twenty terms are old/new/renewed Hebrew terms: ama, ashachim, pota, orla, erva, veset 2. Five terms are accompanied by foreign terms using hyphens or brackets: dagdegan-clitoris, etzem ha'cheq-os pubis, eviona-orgasmus venericus 3. Two are localized foreign terms: ereqtzia, individuomim When you move to the part where pathology, perversion, and venereal diseases are discussed, transliteration becomes dominant: pederastia, erotomania, nimfomaia, steriasis, anestesia, hipestesia. What confuses/mystifies most in the book, apart from its German character, is the inconsistent use of terminology. 8 The author was in fact the famous writer Arthur Koestler, but – not surprisingly – he chose to write this sexual manual under the pen-name Dr. A Costler. 9 The translation is so far removed from the original, that it is hardly a translation, rather an adaptation to what the translator M. Ben-Yoseph judged as the local needs. It concentrates on the “problems” of adjustment, transliterated as “problema”: the Biological Aspect of the Problema, the Psycholgical Aspects of the Problema and the Social Aspects of the Problema are the three first chapters of the Hebrew translation (unlike anything in the original). TRANSLATION AND MEANING, PART 10, 2013, 265 - 273 271 To sum up the techniques or strategies used by the translators we can state the following: a) The most common solution was transliteration: pederosis, qondom, perservativ, urninge, dioninge, homosexualismus. b) Localization was often necessary, done via the Russian model. Nouns ending with the suffix -on received the feminine suffix -ia: ilusia, masturbatzia, prostitutzia, avtoerotiqa, psichiqa. c) Calques, literal translations, were common, mostly for phrases: protzes modifiqatzioni, monthly tzyclus. They were sometimes put in inverted commas, with the original term in brackets: “ha'mistaklim” (voyuers). d) non-systematic mixture of all solutions. Since translators were left to find solutions for themselves, the end-result was usually not systematic. A Biblical term would be used together with a German/Russified term, followed by an asterisk or brackets providing a neologism or an explanation, etc. The message the books were to transmit to youngsters was as foreboding as the language, usually in the nature of an implied warning. For the boy, puberty was described as a dangerous stage to be overcome with the help of self-discipline. A young girl’s life was dominated by the [hygienic] problem of menstruation and by the fear of unwanted pregnancy. Sex was described in sterile, functional terms. G. Cabaillon, who researched the development of sexual education in pre-state Israel claims that the sterile approach was part of the patronizing attitude of knowledgeable adults towards unruly children [of all ages, I would add, for newcomers from “primitive” cultures were treated as children] who had to be educated (Cabaillon, 2000, 531-548). 5. The function of sex manuals Seemingly, the manuals were mostly meant for the use of “professionals”, in that they were pseudoscientific, pseudo-medical. As such, there was no incentive for developing Hebrew alternatives or insisting on using or developing Biblical/Talmudic terminology; the said “professionals” were acquainted with the foreign, mostly German, terms. Whether or not this was indeed the legitimization is not clear, however. Publishers would not confine themselves to such a restricted target-public, well knowing that for lack of legitimate erotic reading material, youngsters (and not so young) read and used sexual manuals as erotica. Erica Jong describes the same phenomenon in Puritan North America of her youth in the 1950s (quoted in St. Jorre, 1994, 125): “It was impossible to obtain a copy of John Cleland’s Memoirs of a Woman of Pleasure outside the rare-book room of a college library or a private erotica dealer. (I tried.) Henry Miller’s Tropics and D.H. Lawrence’s Lady Chatterley’s Lover could not be purchased at your local bookstore. The raciest sex manual available to the panting adolescent was Love Without Fear by Eustace Chesser, MD”. The fact that sex manuals functioned as erotic literature is not surprising in itself. Since the mid 19th century there was a blurring of boundaries between pornographic and pseudo-scientific literature, first and foremost because the two genres were printed in the same underground presses, distributed by the same agents and sold together under the table. They were often written by people who preached free love, divorce, women rights, abortion and contraception, and were therefore considered subversive (Lefkowitz Horowitz, 2003: 272). Translators and writers were aware of this “other” reading public, a fact that transpires from their introductory notes. Such is the case of Dr. A.B. Matmon's book from 1969 Chayei ha'min shel ha'adam [Human Sex Life], this time originally written in Hebrew. In his introduction Dr. Matmon publishes the following warning (translation mine): […] Many textbooks had been translated by non- 272 NITSA BEN-ARI professional people, for “commercial reasons”. This book, however, “is entirely scientific and was written on the basis of clean scientific principles, and those who are looking for titillation would kindly not look in it for piquant material to inflame their sickly urges.” How did readers cope with the mystifying euphemized texts? As best they could, though unlike the British students mentioned by O'Sullivan they did not rush to study German. The German expressions taken in were received with discomfort, if not parodied, first because the subject was taboo, then because Germanized lexemes could not possibly be taken seriously by a generation that grew up on a puristic, chauvinistic approach to Hebrew. Much more far-reaching was the result for the Hebrew language. For years, up to the present, Hebrew has had either a pseudo-medical terminology for sex, or a slang repertoire of “dirty” terms, mostly borrowed from foreign languages: Russian, Polish, Moroccan, Arabic. Hebrew writers who eventually turned from the commune to the individual found they did not possess an updated erotic repertoire. In Reuben Kritz's novel Sartan Ba'laila [Crab at Night] from 1966 (p. 206), a long list of sex guides is mentioned as an inhibiting factor in the sex life of an inexperienced young couple (translation mine): Well, the first obstacle must have come from “professional literature”: you of course read “Man and Woman” and “Human Sex Life”, and probably leafed through the “Encyclopedia for Sexual Sciences” once or twice, although the name was strange and a little repulsive, you also must have browsed through all kinds of booklets and pamphlets that you saw at your friends’ occasionally. Intercourse, so they wrote somewhere (and you loathed the name but could not offer another, for “love-making” sounded just as repulsive), comprises of three parts […] In their joint novel Tandem from 1973, Yonat and Alexander Sened complain of the German sounding technical jargon that accompanies erotic moments. When they look for a direct term for the “pleasure of life”, as they call it, the text almost automatically drifts into German (p. 8, translation mine): […] To find a soft word that would slip into us as the word Liebeslied, or an exact term for the pleasure spasms, and not to have the explosive sounds of Der sogennanten Orgasmus mix in it. How come the words popped up in German? Only in the 1970's, mainly with David Reuben's Everything You Always Wanted to Know about Sex, translated by one of the notorious translators of subversive erotica, Eliezer Carmi, Americanization filtered in. Together with new notions of “happy sex”, the vocabulary underwent two main changes: it became more updated, and English terminology took over where German used to predominate. Hebrew erotica remained lacking and lame, and it was not till the late 1980's that young women writers took up the challenge and began to deal with the lacuna. 6. Conclusion There are numerous kinds of euphemistic strategies, all stemming from the normative tendency of the translator/editor to avoid using the “blatant”, “dirty” or “shameful” word. In this particular case euphemism was achieved through the use of professional scientific sounding German. The use of German gave the general impression of a scientific endeavor, and, what's more, precluded “undesirable” readers. However, it is problematic to discern where self-censorship ends and the anomaly of the development of the Hebrew language begins. In other words whether it was mainly euphemistic tendencies that played a part in the transference of German vocabulary into Hebrew sex guides of the 1930s-1970s, and how far the lack of sex/erotic vocabulary in modern Hebrew may have played its part. The fact that from the 1970s and on Americanization took over where German used to dominate may serve as an indication that the reluctance on the part of innovators to touch problematic material still lingered on. As I have shown elsewhere (Ben-Ari 2006: 337-361) the results of prolonged de- TRANSLATION AND MEANING, PART 10, 2013, 265 - 273 273 ligitimization of dealing with an erotic repertoire over such a long period – and such a crucial period – in modern Hebrew culture had, and still has, long-lasting consequences on Hebrew literature and indirectly, on cultural identity. References Almog, O. (1997), Ha’tzabar: dyokan [The Sabra: A Profile]. Tel Aviv: Am Oved. Assaf, D. & Bartal, I. (1993), “Gilgulo shel zanav: me'chatzrot ha'chasidim el va'slang ha'yisraeli.” [The Metamorphosis of a Tail: from the Hassidic communities to Israeli slang]. Leshonenu La'am, 44, 73-79. Ben-Ari, N. (2006), Suppression of the Erotic in Modern Hebrew Literature. Ottawa: University of Ottawa Press. Ben-Ari, N. (2006a), “The Role and Responsibility of the Anonymous: the historic function of mass translations. TIS (Translation & Interpreting Studies), vol. 1, no 2, Fall, pp. 73-90. Bourdieu, P. (1980), Questions de sociologie. Paris: Minuit. Cabaillon, G. (2000), “Ha'yeled ke'havnaya tarbutit: ha'dugma shel ha'chinich ha'mini ve ha'hasbara ha'minit ba'yishuv ha'yehudi bi'tchilat ha'mea ha'esrim” [The Child as Cultural Product: the example of sexual education and sexual explanation in the Jewish Yishuv in the beginning of the 20th century]. Megamot 40, 3 458-531. Costler [Koestler], A. & Willie, A. (1946), Encyclopedia le'madaei ha'min [Encyclopedia for Sexual Sciences]. Translated from English by Yaacov Ben-Gera. Jerusalem: Shahar. Ellis, H. (1890), The New Spirit. London: Walter Scott Publishing. Even-Zohar, I. (1990), Polysystem Studies, Poetics Today, vol 11, no 1. Even-Zohar, I. (1997), “The Making of Culture Repertoire and the Role of Transfer.” Target, 9 (2): 373-381. Forel, A. (1931), Ha-sheela ha’minit [The Sexual Question]. Translated by Dr. A Bar Natan. 3 vols. Tel Aviv: Mitzpah. Foucault, M. (1976), Histoire de la sexualité, I. La Volonté de Savoir. Paris: Gallimard. Guiraud, P. (1993), Dictionnaire érotique. Paris: Payot et Rivages. Kahn, F. (1962), Chayenu ha’miniyim: madrich ve’yoetz le’chol ish ve’isha [Our Sex Life: A Guide and Counselor for Everyone]. Translated by Baruch Karu. Jerusalem: Achiasaf. Ingebresten, E. (2001), “Wigglesworth, Mather, Starr: Witch-Hunts and General Wickedness in Public. In Tracy Fessenden, Nicholas F. Radel and Magdalena J. Zaborowska, (eds.), The Puritan Origins of American Sex: Religion, Sexuality, and National Identity in American Literature, 21-40. New York, London: Routledge. Kritz, R. (1966), Sartan ba'laila [Crab at Night]. Tel Aviv: Pura. Laser, S.M., Torczyner, H. (1927), Deutsch-Hebraïsches Wörterbuch. Berlin und Wien: Bejamin Harz. Lefkowitz Horowitz, H. (2002), Rereading Sex: Battles over Sexual Knowledge and Suppression in NineteenCentury America. New York: Vintage Books. Marcuse, H. (1955), Eros and Civilization: A Philosophical Inquiry into Freud. New York: Vintage Books. Mazia, A. (1963), “Shemot machalot” [Names of Diseases]. Sefatenu, Vol. 2, 1-6. O'Sullivan, C. (2009), “Translation within the Margin: the 'Libraries' of Henry Bohn. In: Agents of Translation, John Milton and Paul Bandia (eds.), Amsterdam and Philadephia: John Benjamins 107-130. Reuben, D. (1969), Kol ma she'ratzita ladaat al ha'min [Everything You Always Wanted to Know about Sex]. Translated by Eliezer Carmi. Tel Aviv: Bustan. Sened, Y., and A. Sened. (1974), Tandem. Tel Aviv: Ha’kibbutz Ha’meuchad. De St Jorre, J. (1994), The Good Ship Venus: The Erotic Voyage of the Olympia Press. London: Hutchinson. Talmon, A.B. (1938), Chayey ha’min shel ha’adam [Human Sexual Life]. Tel Aviv: Institute for Hygiene and Sexual Sciences. Toury, G. (1995), Descriptive Translation Studies and Beyond. Amsterdam and Philadelphia: John Benjamins. Van de Velde, H. Misterei ha'nisuin [The Mysteries of Marriage. Originally The Ideal Marriage]. No date. Originally: 1926. Translated by M. Ben-Yosef. Jerusalem: Hotzaa Amamit. 274 TRANSLATION AND MEANING, PART 10, 2013, 275 - 279 275 DIE FALLBEARBEITUNG EINE MÖGLICHKEIT DES EINSTIEGS IN DIE DEUTSCHE RECHTSSPRACHE Magdalena Duś Wyższa Szkoła Lingwistyczna, Częstochowa, Poland Abstract: In the center of attention of contemporary studies devoted to specialist languages are specialist texts with their characteristic textual and linguistic features. Specialist texts constitute a specific realization of specialist languages, at the same time being, due to their widespread availability, the crown of specialist communication. Although they have to conform to all the same requirements as non-specialist texts, their form is affected by professional activity of experts in a given field, as well as the specialist subjects they discuss and the extralinguistic knowledge they refer to. The aim of the present article is to characterize German legal language as exemplified by case solutions. Due to the fact that specialist texts are not only mere collections of specialist expressions, they should also be analyzed with the view of problems encountered with regard to their translation. In der heutigen germanistischen Ausbildung kommt dem Fachtext ein hoher Stellenwert zu, insbesondere in Bezug auf die translatorische Tätigkeit. Die zunehmende Bedeutung der Fachsprachen verursacht durch die ständige Entwicklung der internationalen Kooperationsbeziehungen setzt die Beschäftigung mit den Fachtexten bei der übersetzerischen Arbeit der Studenten voraus. Die meisten Schwierigkeiten, nach Weigt1, bereiten den Studierenden die Rechtstexte, deren Verwendung im Unterricht spezielle fachkommunikative Übersetzungsfähigkeiten verlangt. In diesem Beitrag wird die Aufmerksamkeit auf eine andere Art der Arbeit mit und an dem Gesetzestext gelenkt, die bei der Fallbearbeitung zustande kommt. Viele Berufe bearbeiten „Fälle“. Dies macht einen wesentlichen Teil ihrer Arbeit aus. Die Fallbearbeitung steht auch im Zentrum der juristischen Arbeit, sowohl in der Ausbildung als auch in der Praxis. Weder der Rechtsanwalt noch der Richter können sich bei ihrer rechtlichen Analyse mit allgmeinen Ausführungen zu bestimmten rechtlichen Aspekten begnügen, sondern werden beauftragt, konkrete Fälle zu lösen, indem sie diese klären. Die Falllösung, deren Endeffekt ein Gutachten bildet, besteht in erster Linie in der Anwendung von Rechtssätzen auf vorgegebene Sachverhalte, für die eine Lösung zu finden ist. Die Rechtssätze werden also nicht isoliert präsentiert, sondern sie befinden sich „in Aktion“ und werden in ihrer Vernetzung angewendet. Dadurch wird bewusst gemacht, dass die einzelnen Rechtssätze oft noch nicht zur Beurteilung eines Falles ausreichen. Es bleibt dann zu prüfen, ob ein bestimmter Sachverhalt den Tatbestandsmerkmalen einer gesetzlichen Norm entspricht. So kann die Struktur des Rechtssatzes in Text- und Satzstrukturen überführt werden. Dabei dürfte die Antwort auf die Frage wesentlich sein, worin die Vernetzung der Rechtssätze, die zur Klärung beitragen können, bestehen kann. Aus der einschlägigen Literatur geht hervor, dass der Rechtssatz „einen angenommenen Fall, sofern er sich in Zukunft ereignen sollte, in einer Weise, die er vorherbestimmt“, entscheidet2. Der Rechtssatz ordnet dem generell umschriebenen Sachverhalt, dem „Tatbestand“, eine ebenso generell umschriebene „Rechtsfolge“ zu. Der Sinn dieser Zuordnung 1 Weigt, Z. (2010): Fachtext in studentischer Übersetzung. In: M. Duś/G. Zenderowska/Korpus (Hrsg.) Fachsprachenpropädeutik im Germanistikstudium, Częstochowa: Wydawnictwo Wyższej Szkoły Lingwistycznej, S. 54. 2 Rehbinder, M. (1991): Einführung in die Rechtswissenschaft: Grundfragen, Grundlagen und Grundgedanken des Rechts. 7. Auflage. Berlin, New York: de Gruyter, S. 14. 276 ist, dass immer dann, wenn der im Tatbestand bezeichnete Sachverhalt vorliegt, die Rechtsfolge eintritt, d.h. im konkreten Fall gilt. Darauf bezogen weist der Rechtssatz eine bestimmte logische Struktur auf, nämlich: Wenn T dann R, das bedeutet nach Larenz3: Wenn der Tatbestand T in einem konkreten Sachverhalt S verwirklicht ist, gilt für S die Rechtsfolge R. Dieser Rechtssatz wird auf einen konkreten (geschehenen) Sachverhalt angewandt, so dass man – ganz allgemein und stark vereinfacht – sagen kann: Wenn T dann R, und wenn Sachverhalt dem T entspricht, dann tritt R ein. Die Arbeit an Rechtssätzen aus den Gesetzbüchern unterscheidet sich wesentlich von der in der Fallbearbeitung. Um deutlich zu machen, dass man zur Fallklärung die Rechtssätze in ihrer Vernetzung ermitteln muss, wird im Folgenden näher darauf eingegangen. Wesentlich dürfte noch die Erkenntnis sein, dass es außer den soeben erwähnten „vollständigen“ Rechtssätzen auch noch „unvollständige“ Rechtssätze gibt, die eine Hilfsfunktion erfüllen und als „Teile von Geltungsanordnungen“ aufzufassen sind, in dem Sinne, dass man nach ihnen noch keinen Rechtsfall klären kann. Hierbei werden erläuternde, einschränkende und verweisende Rechtssätze4 unterschieden. Manche dieser unvollständigen Rechtssätze „dienen nur dazu, den Tatbestand, ein Tatbestandselement oder die Rechtsfolge eines vollständigen Rechtssatzes näher zu bestimmen; manche schränken einen weit gefassten Rechtssatz ein, indem sie eine bestimmte Fallgruppe von seiner Anwendung ausnehmen; wieder andere verweisen hinsichtlich eines Tatbestandselements oder der Rechtsfolge auf einen anderen Rechtssatz“5. Sie sind sprachlich vollständig, aber als Rechtssätze unvollständig. Ihre konstitutive, Rechtsfolgen begründende Kraft erhalten sie nur in Verbindung mit anderen Rechtssätzen. Die Überführung der logischen Struktur des Rechtssatzes in Text und Satzstrukturen erfolgt zum ersten durch Ausnahme (Exzeption). Die Ausnahme von der Regel kann in einem selbständigen Satz zum Ausdruck kommen, z.B. im § 833 oder § 834 BGB, in dem es sich um die Haftung des Tierhalters im ersten Falle und des Tieraufsehers im zweiten handelt: Die Ersatzpflicht tritt nicht ein, wenn … § 833 Haftung des Tierhalters. ¹Wird durch ein Tier ein Mensch getötet oder der Körper oder die Gesundheit eines Menschen verletzt oder eine Sache beschädigt, so ist derjenige, welcher das Tier hält, verpflichtet, dem Verletzten den daraus entstehenden Schaden zu ersetzen. ²Die Ersatzpflicht tritt nicht ein, wenn der Schaden durch ein Haustier verursacht wird, das dem Berufe, der Erwerbstätigkeit oder dem Unterhalte des Tierhalters zu dienen bestimmt ist, und entweder der Tierhalter bei der Beaufsichtigung des Tieres 3 Larenz, K. (1992): Methodenlehre der Rechtswissenschaft. 2. Auflage. Berlin, Heidelberg: Springer Verlag, S. 159. 4 Ebenda, S. 145-149. 5 Ebenda, S. 145. TRANSLATION AND MEANING, PART 10, 2013, 275 - 279 277 die im Verkehr erforderliche Sorgfalt beobachtet oder der Schaden auch bei Anwendung dieser Sorgfalt entstanden sein würde. § 834 Haftung des Tieraufsehers. ¹Wer für denjenigen, welcher ein Tier hält, die Führung der Aufsicht über das Tier durch Vertrag übernimmt, ist für den Schaden verantwortlich, den das Tier einem Dritten in der im § 833 bezeichneten Weise zufügt. ²Die Verantwortlichkeit tritt nicht ein, wenn er bei der Führung der Aufsicht die im Verkehr erforderliche Sorgfalt beobachtet oder wenn der Schaden auch bei Anwendung dieser Sorgfalt entstanden sein würde. In vielen Fällen wird jedoch die Angabe der Ausnahme in den Rechtssatz aufgenommen, und zwar durch einen Gliedsatz, der durch es sei denn, dass … oder durch es sei denn, es ist … eingeleitet wird. Die konjunktionswertige Formel es sei denn, (dass) wird in der Bedeutung von wenn nicht, ausgenommen verwendet; in den Texten wird sie auch durch wenn nicht, sofern nicht, soweit nicht ersetzt. § 145 Bindung an den Antrag. Wer einem anderen die Schließung eines Vertrags anträgt, ist an den Antrag gebunden, es sei denn, dass er die Gebundenheit ausgeschlossen hat. Die Vernetzung der Rechtssätze äußert sich auch in ihrer Einschränkung (Restriktion). Die den Gliedsatz einleitenden Konjunktionen wie (nur) wenn, soweit, insoweit als (weiterhin: falls, sofern) geben ein Signal, dass das im Hauptsatz Ausgedrückte, die Geltungsanforderung G, durch das im Gliedsatz Formulierte (=q) eingeschränkt wird. Die Restriktion kommt im Falle von § 398 BGB zustande: „Eine Forderung kann von dem Gläubiger durch Vertrag mit einem anderen auf diesen übertragen werden (Abtretung)“. Dieser Satz wird eingeschränkt durch die §§ 399 und 400, nach denen Forderungen unter gewissen Umständen nicht abgetreten werden können. So heißt es im § 399 BGB: „Eine Forderung kann nicht abgetreten werden, wenn die Leistung an einen anderen als den ursprünglichen Gläubiger nicht ohne Veränderung ihres Inhalts erfolgen kann, oder wenn die Abtretung durch Vereinbarung mit dem Schuldner ausgeschlossen ist“. Die weiteren Beispiele für die Restriktion finden wir z.B. in § 581 Wesen des Pachtvertrags. ¹Durch den Pachtvertrag wird der Verpächter verpflichtet, dem Pächter den Gebrauch des verpachteten Gegenstandes und den Genuss der Früchte, soweit sich nach den Regeln einer ordnungsmäßigen Wirtschaft als Ertrag anzusehen sind, während der Pachtzeit zu gewähren. § 109 HGB Gesellschaftsvertrag. Das Rechtsverhältnis der Gesellschafter untereinander richtet sich zunächst nach dem Gesellschaftsvertrag; die Vorschriften der §§ 110 bis 122 finden nur insoweit Anwendung, als nicht durch den Gesellschaftsvertrag ein anderes bestimmt ist. Dass der eingeführten verbindlichen Rechtssprache das Primat zukommt, geht auch daraus hervor, dass sich die Autoren bei der sprachlichen Gestaltung der Fallbearbeitung, also der Erörterung, der in der Rechtssprache üblichen Fachausdrücke bedienen. Sie erbringen damit auch den Nachweis, dass sie sich in der Rechtssprache auskennen und sich der Spezifik der Bedeutung des jeweiligen Fachausdruckes bewusst sind. Für die Arbeit an den Fachtexten ist es wichtig, das betreffende sprachliche Zeichen sowohl unter dem Aspekt des gemeinsprachlichen Gebrauchs als auch unter dem Aspekt des 278 fachsprachlichen, in unserem Falle rechtssprachlichen, Gebrauchs zu betrachten und dessen Spezifik für die Rechtssprache – als Benennung des juristischen Begriffs – bewusst zu machen. Die angehenden Übersetzer sollen die Erkenntnis gewinnen, dass die ihnen bekannten Bedeutungsvarianten eines Lexems nicht voreilig auf die Rechtssprache übertragbar sind. Das Problem der Arbeit mit und an Fachausdrücken hat Baden6 treffend formuliert: „Sehr bald wird der Leser eines Gesetzes … über die juristischen Termini und andere Fachausdrücke stolpern. Einige darunter, die ich vielleicht als ‚offene’ Fachausdrücke bezeichnen würde – wie etwa ‘Hypothek’, ‘Verzug’, ‘Verfügung‘ u.a. – sind leicht als solche zu erkennen. Andere dagegen, wie ‘Urkunde’, ‘Gewalt’, ‚Miete‘ und ‚Leihe‘, ‚Interesse‘, müssen im Gegensatz hierzu gleichsam als ‚verdeckte’ Fachausdrücke erscheinen. Das hier benutzte Wort ist jeweils auch in der Umgangssprache bekannt, wird dort aber in engerem oder weiterem, teilweise aber auch in gänzlich anderem Sinne (‘Interesse’) benutzt.“ Als Beispiel für das Obige kann das Lexem Auftrag angeführt werden: Auftrag nach DUW bedeutet so viel wie: ‚Weisung, eine zur Erledigung übertragene Aufgabe’; ‚Bestellung von Waren oder Dienstleistungen‘ während in der Rechtssprache (gemäß § 662 BGB) in der Bedeutung von ‚schuldrechtlichem Vertrag zwischen Auftraggeber und Beauftragtem, durch den sich der Beauftragte verpflichtet, ein ihm vom Auftraggeber übertragenes Geschäft für diesen unentgeltlich zu besorgen’7 Anwendung findet. § 662 Vertragstypische Pflichten beim Auftrag. Durch die Annahme eines Auftrags verpflichtet sich der Beauftragte, ein ihm von dem Auftraggeber übertragenes Geschäft für diesen unentgeltlich zu besorgen. Für weitere Gegenüberstellung eignet sich der Begriff Leihe. Laut Duden wird Leihe bestimmt als ‚(gegen das Versprechen der Rückgabe) vorübergehend aus seinem Besitz zur Verfügung stellen; ausleihen’. Im Fachlexikon Recht finden wir eine andere Bedeutungsvariante, die sich nach § 598 BGB richtet: Leihe (Leihvertrag): Vertrag, durch den sich der Verleiher einer (beweglichen oder unbeweglichen) Sache verpflichtet, dem Entleiher den Gebrauch der Sache unentgeltlich zu überlassen.8 § 598 Vertragstypische Pflichten bei der Leihe. Durch den Leihvertrag wird der Verleiher einer Sache verpflichtet, dem Entleiher den Gebrauch der Sache unentgeltlich zu gestatten. Für die Zwecke dieses Referates ist es nicht entscheidend, möglichst viele Erscheinungen, die bei der Fallbearbeitung auftauchen, zu behandeln. Vielmehr wichtig ist es, diese Erscheinungen bewusst zu machen und die Lerner/Übersetzer für die angesprochene Problematik zu sensibilisieren. Für unser 6 Baden, E. (1975): Zur Sprachlichkeit der Gesetze. In: Rödig, J.; Baden, E.; Kindermann, H.: Vorstudien zu einer Theorie der Gesetzgebung. Bonn: Gesellschaft für Mathematik und Datenverarbeitung (Berichte der Gesellschaft für Mathematik und Datenverarbeitung), S. 99-106, Zitat - Abschnitt 3.2.2.2. 7 Fachlexikon Recht (2005): 2. Auflage. Hrsg. von den Verlagen Alpmann & Schmidt Juristische Lehrgänge und F. A. Brockhaus, S. 143. 8 Fachlexikon Recht (2005): 2. Auflage. Hrsg. von den Verlagen Alpmann & Schmidt Juristische Lehrgänge und F. A. Brockhaus, S. 863. TRANSLATION AND MEANING, PART 10, 2013, 275 - 279 279 Anliegen war interessant zu untersuchen, worin der Vorzug der Fallbearbeitung in Bezug auf die Vermittlung/Anwendung der Rechtssprache besteht. Es kann ohnehin festgestellt werden, dass die Fallbearbeitung einen umfangreichen Stoff für die Arbeit mit einem Fachtext und einen Einstieg in die deutsche Rechtssprache bietet. Der Umgang mit den rechtlichen Fällen gibt den Studierenden einen Einblick in das Wesen des deutschen Rechts, ermöglicht ein besseres Verstehen der deutschen Rechtssätze und kann durch die Arbeit an einem Fachausdruck eine große Hilfe in ihrer translatorischen Tätigkeit leisten. Literaturverzeichnis Baden, E. (1975), „Zur Sprachlichkeit der Gesetze“. In: Rödig, J.; Baden, E.; Kindermann, H.: Vorstudien zu einer Theorie der Gesetzgebung. Bonn: Gesellschaft für Mathematik und Datenverarbeitung (Berichte der Gesellschaft für Mathematik und Datenverarbeitung), S. 99-106. Duden Deutsches Universalwörterbuch. (2003) 5. Auflage. München; Leipzig; Wien; Zürich: Dudenverlag Fachlexikon Recht (2005), 2. Auflage. Hrsg. von den Verlagen Alpmann & Schmidt Juristische Lehrgänge und F. A. Brockhaus. Larenz, K. (1992), Methodenlehre der Rechtswissenschaft. 2. Auflage. Berlin, Heidelberg: Springer Verlag Rehbinder, M. (1991), Einführung in die Rechtswissenschaft: Grundfragen, Grundlagen und Grundgedanken des Rechts. 7. Auflage. Berlin, New York: de Gruyter. Weber, S. (Hrsg.) (1997), Recht & Sprache. Reader für die fach- und berufsorientierte Ausbildung/Wieterbildung auf dem Gebiet Deutsch als Fremdsprache. Typoskript. Chemnitz. Weber, S. (Hrsg.) (2001), Der deutschsprachige Rechtssatz: inhaltliche und sprachliche Merkmale. Typoskript, Chemnitz. Weigt, Z. (2010), „Fachtext in studentischer Übersetzung“. In: M. Duś/G. Zenderowska/Korpus (Hrsg.) Fachsprachenpropädeutik im Germanistikstudium, Częstochowa: Wydawnictwo Wyższej Szkoły Lingwistycznej, S. 49-56. 280 TRANSLATION AND MEANING, PART 10, 2013, 281 - 290 281 LE LEXIQUE DE SPECIALITE DANS L’APPRENTISSAGE ET DANS LA TRADUCTION Jacek Florczak University of Łódź, Poland Abstract: The aim of this paper is to analyze different types of semantic relations occurring between Polish and French specialized terminology. It is assumed that appropriate knowledge of specialized vocabulary of a foreign language involves not only comprehensive knowledge of the given field, but also linguistic knowledge necessary to notice differences in semantic valuations of such vocabulary, both from the perspective of one language, and the languages undergoing comparison. With this assumption in mind, the most significant differences between concepts, conventions and profiles of selected specialized French and Polish items will be discussed and analyzed, and an appropriate model for such an analysis will be presented for the (future) translators of specialized languages. Nous analyserons différents types de relations sémantiques entre les mots appartenant aux lexiques de spécialité des langues française et polonaise. Nous partons du principe que pour aboutir à une connaissance appropriée d’un lexique de spécialité en langue étrangère, il faut posséder non seulement un savoir approfondi sur la problématique, dans laquelle le lexique en question s’inscrit, mais aussi un savoir-faire linguistique, i.e. un savoir sur la langue permettant de voir les différentes valeurs des mots de spécialité aussi bien au sein d’une seule langue qu’au niveau interlinguistique. 1. Approche linguistique – concepts principaux Les différences entre le lexique de spécialité et le lexique général se définissent souvent suivant les critères sémantiques (univocité, précision, concision) et fonctionnels (p.ex. au niveau des valeurs pragmatiques). Avec la plupart des chercheurs (recensés dans Kacprzak, 2000: 16), nous admettons cependant que les différences entre la compétence lexicale en langue générale et la compétence « terminologique » en langue de spécialité sont d’ordre graduel plutôt que catégorique. Autrement dit, la compétence ne se situe pas sur un pôle ou un autre, mais sur un continuum gradué entre deux pôles. Du point de vue linguistique les mêmes outils d’analyse sémantique peuvent donc être appliqués aux lexiques général et de spécialité, et les différences éventuelles dans les modèles de leur acquisition portent sur l’accentuation des difficultés choisies qui paraissent plus ou moins fréquentes dans un lexique par rapport à l’autre. De là, nous allons donc proposer une technique d’analyse linguistique des différences les plus frappantes entre les concepts, les conventions et les mises en profil des mots appartenant aux lexiques de spécialité choisis dans la langue française et polonaise. Le modèle de cette analyse sera ensuite conseillé aux (futurs) traducteurs de langues de spécialité en tant qu’outil d’analyse sensibilisant à différents problèmes et difficultés auxquels peut être confronté tout traducteur interprète de langue de spécialité. 2. Convention et conventionnalisation Le premier concept employé porte sur la différence entre ce que l’on PEUT dire et ce qui SE dit. Pour prouver que nos élèves ne comprennent pas toujours la différence entre forme conventionnelle et non conventionnalisée, voire usage occasionnel (hapax legomena), il suffit de citer l’argument utilisé par un étudiant en vue de défendre « l’appropriation » de la forme qu’il a employée : Cela SE dit. Je l’ai entendu dire UNE FOIS dans le métro. Les problèmes relatifs à l’assimilation de la convention des mots étrangers peuvent avoir deux origines distinctes. La première est relative aux interférences des degrés de conventionnalisation des formes de langue maternelle, qui prouve par ailleurs que les apprenants ne sentent pas les degrés de convention(nalisation) de leurs équivalents étrangers. La deuxième dépasse le cadre lexical et concerne le rôle qu’on attribue à l’enseignement de la grammaire « à l’état pur », et dont l’appropria- 282 JACEK FLORCZAK tion est censée donner à l’élève un outil permettant de transmettre pratiquement chaque pensée visée à l’aide d’un nombre de règles restreint. On sait pourtant que du nombre indéfini des structures grammaticalement générables, la langue en privilégie certaines aux dépens des autres. Dans ce cas donc, même si l’élève est à même de créer correctement les structures qu’il veut, il reste de savoir si cellesci SE disent au sein de la communauté linguistique donnée. Nous proposons une double approche du concept de conventionnalisation. La première est de nature quantitative. Nous y admettons que le degré de conventionnalisation d’une forme est proportionnel à la fréquence de son emploi. La deuxième est de nature qualitative. Nous y admettons que le degré de conventionnalisation dépend de l’interprétant de son emploi, p.ex. du sociolecte. Il est en fait fréquent qu’une forme considérée comme conventionnalisée dans un sociolecte ne le soit plus en dehors de lui. La question de savoir si telle ou telle forme SE dit doit donc être complétée par la question : OÙ et COMMENT elle s’emploie. 3. Concept et conceptualisation Sachant déjà ce qui SE dit, nous pouvons passer à l’analyse de ce que cela VEUT dire. L’analyse des concepts se fait dans deux perspectives. Dans la perspective interlinguistique, nous analysons les ressemblances et différences entre les expressions comparées (p.ex. uczciwość vs honnêteté). Dans la perspective intralinguistique, nous analysons les sens des mots au sein de la seule langue étrangère (p.ex. honnêteté vs justesse). Différentes techniques linguistiques peuvent être proposées dans ce but. Leur description dépasserait cependant le cadre de ce travail. Nous nous bornerons donc à constater que leur objectif principal est de faire apprendre comment faire des analyses sémantiques pertinentes et/ou comment vérifier la pertinence des résultats obtenus. L’accès cognitif aux conceptualisations de langue étrangère se fait à l’aide des concepts propres aux formes d’expressions maternelles. Bien que la querelle des universaux sémantiques n’ait jamais cessé, tout modèle comparatif implique l’existence d’un ensemble d’unités conceptuelles communes aux langues comparées, avec la possibilité de combinatoires différentes dans les structures considérées comme équivalentes. Les rapports entre les concepts appartenant aux mêmes champs lexicaux peuvent être illustrés à l’aide des diagrammes (Trier 1934, dans Lyons 1980) rendant compte des ressemblances et différences entre les fragmentations d’une même substance du contenu par les langues différentes. a b c 1 vs a a b b c c 2 3 a vs b c a a vs b b c Dans le premier cas (1), il n’y a aucune différence entre les fragmentations du même champ lexical : les deux systèmes sont isomorphes (pol. milimetr, centymetr, metr ↔ fr. millimètre, centimètre, mètre). Dans le deuxième cas (2), nous avons le même nombre de lexèmes recouvrant les champs lexicaux équivalents, mais leurs partages sémantiques ne sont pas isomorphes : la répartition des traits sémantiques n’est pas identique dans les lexèmes comparés (pol. wiedza, znajomość, umiejętność ↔ fr. savoir, connaissances, facultés). Dans le troisième cas (3), les mêmes champs lexicaux sont couverts par un nombre différent de lexèmes ; leurs structures internes sont donc forcément différentes (pol. tak vs nie, fr. oui vs non vs si). Enfin, les cas (2) et (3) peuvent se superposer. 4. Profil et mise en profil Le concept de profil et de mise en profil a reçu de nombreuses définitions. On l’illustre le plus souvent avec de simples images d’un verre à moitié plein et à moitié vide ou bien d’un triangle et d’un trilatère. La sémantique structurale se contente de constater que ces expressions, bien qu’elles ne soient pas synonymes, car elles actualisent d’autres traits sémantiques, elles sont équivalentes, car TRANSLATION AND MEANING, PART 10, 2013, 281 - 290 283 elles désignent le même référent. Dans l’approche cognitive, par contre, nous rendons compte de la possibilité d’une perception différente d’un même objet, tout en soulignant que deux perceptions linguistiques différentes ne créent plus les mêmes objets linguistiques (différents éléments de la même base ont été mis en profil). Ce n’est donc plus l’objet tel quel qui compte ; c’est plutôt la façon de le percevoir par le biais de la convention de la langue. Approche logique (désignative) Approche cognitive (de mise en profil) vs vs Le succès du concept de (mise en) profil est certainement dû à sa grande fonctionnalité et forte suggestivité. La pratique de son emploi fait constater qu’il facilite une appropriation intuitive de nombreuses relations entre les langues. C’est avec lui que l’on comprend facilement qu’elles peuvent attribuer des poids différents aux différents éléments des mêmes réalités (extra)linguistiques et que ce qui se trouve en premier plan dans une langue peut passer en arrière plan dans une autre. C’est aussi grâce à lui que nous pouvons attirer l’attention sur les divergences entre les mots isolés p.ex. pol. kwartał (1/4 de l’année) – fr. trimestre (3/12) ; entre les concaténations syntagmatiques plus ou moins courtes, p.ex. le Polonais miał włamanie ('a eu un cambriolage'), tandis que le Français s’est fait cambrioler ; ainsi qu’entre les constructions syntactiques au niveau des formes phrastiques complètes, p.ex. : pol. Nie wszystko złoto, co się świeci ('Pas tout x est y') vs fr. Tout ce qui brille n’est pas or ('Tout x n’est pas y'). 5. Interprétant La notion d’interprétant est due à Ch.-S. Peirce (1931-1958 [1978]) dont la réception n’a jamais été facile. Nous simplifions donc sa version originale et entendons par interprétant tout ce qui compte pour l’interprétation adéquate de la signification actualisée. L’interprétant sera par exemple le contexte d’emploi : une autre signification sera attribué à la tigresse d’un documentaire animalier {DA}, une autre dans le contexte des potins sur notre directrice {PD}. Les {DA} et {PD} sont ici les interprétants. La phrase Qui veut prendre la parole aura des lectures différentes suivant qu’elle est prononcée avec une prosodie interrogative ou menaçante. Dans ce cas, l’interprétant est l’intonation. L’interprétant peut être de nature intrinsèque, relativement à l’environnement lexical, et extrinsèque, relativement par exemple, aux conventions les plus diverses. Il peut être aussi une situation de communication, y compris exolingue. Les interprétants sont encore le récepteur et l’émetteur du message et les rôles qu’ils jouent en communication exolingue sont bien particuliers. 6. Approche cognitive – analyse CCP1 Les différentes langues peuvent procéder aux découpages CCP différentes dans les structures considérés comme équivalentes. La conceptualisation Cp1 d’une pensée 'P' conforme à la convention Cv1 de la langue L1 peut nécessiter, par exemple, une mise en profil Pr1 différente de celle Pr2 que nécessite la conceptualisation Cp2 conforme à la convention Cv2 de la langue L2. Les cas de figure sont nombreux et peuvent être, tous, représentés comme suit: 1 La ressemblance est purement fortuite entre les sigles CCP pour Charge Culturelle Partagée (Galisson) et pour Concepts, Conventions et Profils. 284 JACEK FLORCZAK L1 : 'P' Cv1 Cp1, Pr1 = ≈ ≠ L2 : 'P' Cv2 Cp2, Pr2 On voit bien que la combinatoire fort élevée des éléments susmentionnés (p.ex. Cv1=Cv2, Cp1=Cp2, Pr1≠Pr2 ; Cv1≠Cv2, Cp1=Cp2, Pr1=Pr2 ; Cv1≠Cv2, Cp1=Cp2, Pr1≠Pr2 ; ainsi de suite) fait exposer tout apprenant de L2 au grand nombres de ratés éventuels, et la pratique prouve qu’il n’y échappe pas. Les langues intermédiaires des étrangers abondent en profils « étranges », dans des phrases « étrangement » construites, et leur présence régulière est perçue par les usagers natifs comme une spécificité de l’interlangue d’une communauté étrangère donnée. Les ratés de ce type peuvent porter sur tous les plans linguistiques et concerner les valeurs (sémantique, pragmatique, etc.) et grandeurs (mot, syntagme, phrase, etc.) les plus diverses. Dans ce qui suit, nous nous limiterons au lexique. 7. En langue générale Il est tout à fait naturel que l’apprenant de LE perçoive les mots étrangers par le biais CCP de sa langue maternelle. Le verbe embrasser, par exemple, sera perçu avec un profil des bras, ce qui lui fera restreindre l’étendue de son emploi aux seules actions où les bras sont activés ('accolades'). Suite à une acquisition spontanée de la langue étrangère (cf. loi du moindre effort, motivation secondaire) pour les actions où les bras ne sont pas employés et où est mis en profil, par contre, le 'contact de la bouche avec qqn ou qqch', il tentera de choisir une dénomination différente ; par exemple baiser. C’est ainsi qu’on peut expliquer l’un des ratés classiques dans le français des Polonais, observés notamment dans la première étape d’apprentissage de la langue, et dont les effets sont particulièrement néfastes, car le verbe baiser a un effet pragmatique fort vulgaire. Bien évidemment, l’élève qui y sera sensibilisé évite d’employer ce mot. Ceci ne l’empêche pas cependant de continuer à chercher dans son interlangue une expression qui lui permette de mettre en profil le 'rapprochement des lèvres'. L’expression donner un baiser, qui n’est plus vulgaire et qui désigne, effectivement, le fait de 'poser sa bouche sur qqn ou qqch', peut lui venir en aide. C’est là que finit l’étape de ses recherches ; l’élève se construit un ensemble d’expressions qui lui permettent de soutenir le partage polonais de la réalité (extra)linguistique française. On sait bien, pourtant, que les structures CCP peuvent être fort différentes dans différentes langues : l’expression française donner un baiser est moins conventionnalisée que le verbe polonais (po)całować, parce que la forme la plus conventionnalisée est celle d’embrasser. Malgré la présence des 'bras' dans sa racine, par suite de sa grande fréquence d’emploi dans ce contexte, elle ne met plus en profil leur action de manière aussi forte que cela paraît à l’apprenant polonais. Le problème d’appropriation adéquate des structures CCP se laisse également expliquer par le biais de l’analyse ontogénétique du processus d’acquisition du lexique maternel effectuée du point de vue sémasiologique. La plupart de nos énoncés ont un caractère figuré. Or, les enfants perçoivent les premières occurrences des mots nouveaux de manière très concrète. Un enfant, qui entend parler pour la première fois d’un homme homme-grenouille, « verra » donc un 'homme grenouille'. Combien de fois on entend demander nos enfants : Pourquoi tu dis qu’il prend ses jambes à son cou, alors qu’il les pose par terre ? Ce n’est qu’avec le temps et à force d’entendre dire ces expressions dans des contextes particuliers qu’elles s’enracinent dans son esprit si bien qu’il finit par oublier leurs profils métaphoriques (catachrétiques) et commence à les interpréter conformément à la convention de sa langue. Dans l’approche structurale nous dirons qu’elles ont été lexicalisées, dans l’approche cognitive, que la langue les a conventionnalisées si bien qu’elles sont perçus comme « tout à fait normales ». L’homme-grenouille devient un simple 'plongeur', prendre ses jambes à son cou devient une façon plus expressive de dire 's’enfuir précipitamment'. Nous nous souvenons de ces profils quand les mots qui en sont porteurs sont employés dans les interprétants dans lesquels nous n’avons pas pris l’habitude de les utiliser ; ne serait-ce que dans une langue de spécialité. La notion polonaise de odchody utilisée en comptabilité ('dépenses') fera rire tous ceux qui ne sont pas comptables, compte tenu son profil particulier actualisé en langue générale ('excréments'). Mais vu la grande fréquence TRANSLATION AND MEANING, PART 10, 2013, 281 - 290 285 (et donc la normalité) de son emploi dans le langage des comptables, ceux-ci ne riront plus lorsqu’ils le verront apparaître, ayant en fait petit à petit oublié son profil actualisé dans le langage général. Les phénomènes susmentionnés ont des conséquences très importantes pour le processus d’acquisition du lexique de la langue étrangère. La première concerne la perspective onomasiologique ; il est en fait fréquent que nous cherchions à conceptualiser nos pensées de manière à mettre en profil ce que nous considérons comme étant conventionnalisé dans notre esprit, ou bien que nous évitions l’emploi des formes mettant en profil ce à quoi la convention de notre langue ne nous a pas habitués. Exemple de la première situation : un Polonais demande à son invité français : déshabillez-vous, en voulant transmettre ce qu’exprime l’expression polonaise proszę się rozebrać, qui ne fait pas forcément admettre que l’effet de cette action est de rester nu, ce que fait admettre l’expression française. Exemple de la deuxième situation : bien qu’on n’arrête pas d’expliquer aux étudiants polonais qu’en français ON dit embrasser, ils n’arrêtent pas d’employer donner un baiser.2 La deuxième conséquence porte sur la façon dont nous percevons les mots étrangers dans la perspective sémasiologique, en leur attribuant les profils qu’ils n’ont pas ou bien en n’interprétant pas leurs profils comme il faut.3 Les exemples de ratés relatifs à la seconde situation sont fournis par les blagues que nous ne comprenons pas. Le premier cas sera illustré par une situation authentique où un étudiant polonais pendant un cours en France a éclaté de rire, seul dans toute la classe, quand une étudiante, arrivée en retard, a donné pour excuse : J’ai oublié de me réveiller (en polonais on oublie qqch si et seulement si on en est conscient). Quelques exemples de listes d’équivalents permettent de prendre conscience des relations décrites cidessus. Pour les besoins de cet article la description des rapports CCP sera fort simplifiée et les unités comparées seront divisées en fonction de leur grandeur: Proverbes et dictons : Mieux vaut tard que jamais Cvfr = Cvpl, Cpfr = Cppl, Pfr = PPl Pierre qui roule n’amasse pas mousse Cpfr = Cppl, Pfr = PPl, Cvfr ≠ Cvpl Chien qui aboie ne mord pas Cpfr = Cppl, Cvfr = CvPl, Pfr ≠ Ppl Qui se ressemble, s’assemble Cvfr = Cvpl, Pfr ≈ PPl, Cpfr ≠ Cppl Lepiej późno niż wcale Kamień toczący się nie obrasta mchem Krowa, która ryczy mało mleka daje Kto z kim przestaje, takim się staje Expressions phraséologiques et idiomatiques : Le jeu en vaut la chandelle Gra warta świeczki Cvfr = Cvpl, Cpfr = Cppl, Pfr = PPl La mariée est trop belle Skarżyć się nie w porę Cpfr = Cppl, Pfr ≠ PPl, Cvfr ≠ Cvpl Faire d’une pierre deux coups Upiec dwie pieczenie na jednym ogniu Cpfr = Cppl, Cvfr = CvPl, Pfr ≠ Ppl Avoir du foin dans ses bottes Słoma z butów wyłazi Pfr ≈ PPl, Cvfr ≠ Cvpl, Cpfr ≠ Cppl 2 Et les exemples fourmillent sur tous les plans de la langue : lexical (saluer, pour 'witać', au lieu de dire bonjour), syntactique (pas toutes les femmes sont naïves, pour 'nie wszystkie kobiety są naiwne' au lieu de toutes les femmes ne sont pas naïves), pragmatique (Madame Alicja, pour 'Pani Alicjo', au lieu de Madame Nowak). 3 Nous réagissons aux mots étrangers tout comme nous réagissions aux mots de la langue maternelle au moment de leur première perception ; nous pouvons les trouver ridicules, mal appropriés, tout comme si leur profil avait eu quelque chose de particulier, alors qu’en langue étrangère il peut n’en être rien. 286 JACEK FLORCZAK Expressions simples ou composés : Grippe porcine Cvfr = Cvpl, Cpfr = Cppl, Pfr = PPl Narcotiques Cpfr = Cppl, Pfr = PPl, Cvfr ≠ Cvpl Souffre-douleur Cpfr = Cppl, Cvfr = CvPl, Pfr ≠ Ppl Essuie-mains Cvfr = Cvpl, Pfr ≈ PPl, Cpfr ≠ Cppl Świńska grypa Narkotyki Popychadło Ręcznik 8. En langue spécialisée Les langues de spécialité concernent les domaines les plus divers : depuis les sciences dites dures (physique, chimie) jusqu’aux sciences molles (sociologie, psychologie). Le lexique de spécialité est également présent dans la langue quotidienne ; tel est le cas des lexiques culinaire, sportif, pharmaceutique, etc. Chaque langue de spécialité concerne un domaine particulier, qui, lui, crée un interprétant dans lequel les structures CCP se forment en quelque sorte à nouveau. Nous y avons affaire soit aux mots nouveaux (néologismes), soit aux mots qui sont passés de la langue générale à la langue de spécialité (néosémantismes). Nous avons cité le cas de odchody ('dépenses') qui s’opposent aux przychody ('recettes') dans la comptabilité et qui dans la langue générale ne font penser qu’aux excréments. On ne sera donc pas étonné de voir la surprise de celui qui n’a pas pris le temps de s’habituer au langage des comptables. Il en va de même quand nous apprenons un lexique de spécialité d’une langue étrangère. Quelqu’un qui a pris l’habitude de parler de rachunek zysków i strat cherche une expression française avec les profils analogues et il lui faut du temps pour s’accoutumer au fait que l’expression compte de résultat ne mette en profil ni les 'bénéfices' (zyski) ni les 'pertes' (straty). 8.1 Concepts On admet que le lexique de spécialité diffère du lexique général par la précision de ses concepts et que les différences conceptuelles qui se manifestent tout particulièrement entre les lexiques généraux appartenant aux langues différentes n’existent pas dans les termes de spécialité, car les référents qu’ils désignent sont toujours les mêmes quelle que soit la communauté linguistique. Autant le premier constat décrit effectivement une particularité typique de toute langue spécialisée, autant le second fait prendre en compte les différences entre les domaines appartenant aux sciences « sociodépendantes » et « socio-indépendantes ». Les secondes (chimie, physique, biologie, etc.) décrivent une réalité partagée par les différentes communautés linguistiques. L’enseignement/apprentissage de leur terminologie se ramène à une description précise de ses concepts et aux choix des équivalents d’étendues identiques. Les seules différences concernent les conventions et les mises en profil (p.ex. włókna wolno- i szybkokurczliwe – fibres rouges et blanches plutôt que fibres contractiles lentes et rapides). Font partie des premières les sciences telles que droit, administration, comptabilité. Les divergences conceptuelles peuvent y être fréquentes. L’exemple classique nous est fourni par le droit français et polonais, dont l’incompatibilité a été bien visible lors des premières traductions de l’acquis communautaire vers le langage du droit polonais. La situation la plus embarrassante a lieu lorsqu’une langue possède ce que l’autre n’a pas. Ainsi, l’impôt polonais podatek akcyzowy dont l’équivalent français droit d’accise n’est connu que par un groupe restreint de fiscalistes communautaires. Un autre problème concerne les différents principes de fonctionnement des choses en apparence identiques. Les sociétés spółki z o.o. polonaises ont une responsabilité limitée au seul capital social. Les SARL françaises fonctionnent dans ce cadre un peu différemment. Il y a encore le problème des lacunes dans un lexique par rapport à l’autre. Un terme polonais konkurencyjny trouve deux équivalents français : concurrentiel et compétitif, où le premier réfère à la présence même d’une concurrence sur le marché, tandis que le second évoque l’idée de compétition entre sociétés concurrentielles. Mais l’acquisition la plus compliquée semble être celle où les mots équivalents forment des réseaux de concepts qui ne se recouvrent qu’en partie. Seule une TRANSLATION AND MEANING, PART 10, 2013, 281 - 290 287 lecture très attentive des définitions des termes polonais : zbrodnia, przestępstwo, wykroczenie, występek et français : crime, délit, infraction, contravention permettra de comprendre les écarts sémantiques plus ou moins significatifs entre les uns et les autres. 8.2 Conventions Dans les dictionnaires techniques bilingues un seul mot de L1 est accompagné de quelques équivalents de L2 sélectionnés en fonction des domaines (i.e. interprétants) choisis. Par exemple obrót a pour équivalents : chiffre d’affaires dans le domaine de la comptabilité et échange dans le domaine d’échanges commerciaux. Parfois, un seul mot trouve quelques équivalents dans le même domaine, mais certains sont plus, d’autres moins conventionnalisés. Par exemple dans le milieu des statisticiens on parle de odchylenie standardowe – écart type, quoiqu’on puisse également employer le mot odległość – distance. Dans le milieu des diététiciens on parle de błonnik, fibres ou féculents, quoiqu’on puisse employer le mot cellulose. Différents mots peuvent être aussi employés en fonction de leur interprétant. Un mot français presbytie trouve deux équivalents polonais : prezbicja et starcza ślepota, le premier étant employé dans le milieu des ophtalmologistes, le second, lorsqu’ils s’adressent à leurs patients. Cela ne veut pas dire que, pour starcza ślepota, il faut créer un terme français nouveau cécité sénile, car la presbytie française n’a pas la convention scientifique qu’a la prezbicja polonaise. Cette remarque peut être généralisée, car les formes françaises d’origine latine ont un degré de conventionnalisation moins élevé dans l’interprétant de la langue générale que ce n’est le cas dans les langues slaves. La personne qui se sert du lexique de spécialité étranger est exposée au risque d’emploi d’expressions peu connues dans le milieu donné. Les dictionnaires bilingues ne l’aident que dans un cadre limité, car ils ne définissent pas les degrés de conventionnalisation. Un néophyte ignorant la convention du lexique étranger essaie, cependant, de rendre compte de toutes les valeurs qu’il considère comme étant importantes. Il pourra donc employer une forme qui est si faiblement conventionnalisée qu’elle en devient incompréhensible. Tel est, par exemple, le cas de capital roulant employé pour désigner kapitał obrotowy, laquelle expression a déjà longtemps disparu du lexique de la comptabilité française, même si elle est toujours enregistrée par les dictionnaires bilingues. 8.3 Profils Les différences entre les (mises en) profils des termes de spécialité équivalents sont bien fréquentes.4 Par exemple, le vocabulaire de la cuisine française se sert des unités de décilitres, centilitres et grammes, tandis que le lexique polonais préfère les dénominations exprimées en litres, décagrammes où en ½, ¼ de verre. Vu l’incompatibilité entre ces deux conventions lexicales, le recours aux grammes et centilitres, dans les recettes de cuisine polonaise sera perçu comme une nécessité de mesurer exactement le produit en question (cf. mise en profil), sans mentionner le fait que nombre de cuisiniers polonais ne savent pas ce que représente 10 cl ou 10 dl. L’apprenant de lexique de spécialité se laisse fréquemment tromper par les profils conventionnalisés dans l’interprétant de sa langue maternelle. Un spécialiste du droit de sociétés polonais aura tendance à opposer les personnes physiques aux personnes juridiques, conformément à leurs dénominations 4 A titre d’exemple (d’après : Rutkowski, 2001) voici une simple analyse du signe internet « @ » qui est « perçu » différemment par différentes communautés linguistiques. Les plus neutres sont les dénominations anglaise : at (de : 5 @ $10 – 5 pièces à 10 dollars) et romanes : arobase (fr.), arroba (esp.) (mesure de 10,5kg12,5kg dénommée par @). Les plus intéressantes sont les expressions motivées par la forme. Par exemple : singe (pol. małpa, serbe majmun, slovène afna), queue de singe (afrikaans aapstert, holland. apestaartje, roum. coada de mamuta, allem. Klammeraffe), queue de chat (suéd. Kattsvans, finl. kissanhäntä), queue de cochon (allem. Schweinekringel, dan. grisehale), escargot (hebr. shablul, it.. chiocciola), petit chien (rus. sobačka), oreille d’éléphant (Suisse elefantöra), ver (hongr. kukac), caneton (grec. papaki). 288 JACEK FLORCZAK polonaises osobowość fizyczna vs prawna, tandis que les Français distinguent ces deux unités par les expressions : personnes physiques vs sociétés. Un autre exemple vient d’un traducteur débutant qui a rendu ręce do góry par levez vos bras, ce qui n’était aucunement approprié au contexte du film de guerre polonais (Stawka większa niż życie) dont il était question (haut les mains). 9. Approche applicative – dictionnaires et recherches De ce qui précède, il s’ensuit qu’une acquisition appropriée d’un lexique de spécialité quelconque nécessite une connaissance plus ou moins approfondie des concepts (savoir spécialisé, le mieux dans une perspective comparée), des degrés de conventionnalisation (en langue maternelle et étrangère ainsi que dans différents interprétants, p.ex. : texte scientifique vs texte de vulgarisation), ainsi que des profils considérés comme conventionnalisés dans un interprétant donné. Afin d’aboutir à cet objectif, nous proposons une double voie dans l’usage de ce qui précède. 10. Dictionnaires des pièges La description proposée peut être adoptée par les dictionnaires bilingues. Les dictionnaires traditionnels s’avèrent en fait fort insuffisants. Leurs entrées présentent toujours un item de la langue L1 suivi de quelques équivalents de L2. On n’y trouve donc ni explications des divergences conceptuelles (concurrentiel – konkurencyjny) ni explications des différences entre les conventions (osobowość prawna – personne juridique). Une certaine expérience est donc requise pour faire usage approprié de ce type de dictionnaires. Paradoxalement, ils servent davantage à des apprenants/traducteurs expérimentés qu’aux débutants. Si, en fait, les seconds ont tendance à les traiter comme une sorte d’autorité infaillible, les premiers considèrent les équivalents trouvés comme une suggestion de continuer leurs recherches autonomes au-delà (vérifiant ou falsifiant la solution proposée par le dictionnaire). Face à ces difficultés, on pourrait donc proposer des dictionnaires enrichis des aspects CCP décrits ci-dessus. Un bon exemple de dictionnaire de ce type est fourni par Wilczyńska et Rabiller 2010. Mais ce dictionnaire est conçu notamment pour la langue générale, et, à notre connaissance, il n’y pas de dictionnaires techniques de ce type. 11. Recherches individuelles D’ailleurs, aussi riche que soit le dictionnaire CCP, il sera toujours probable qu’on n’y trouve pas ce qu’on y cherche. Pour remédier à ces défauts, nous proposons donc de faire des analyses décrites cidessus une technique d’acquisition du lexique de spécialité. Le but de l’exercice proposé sera de faire acquérir des réflexes de recherches CCP autonomes et appropriées. La matrice de l’exercice est très simple. Dans le domaine à caractère plus ou moins spécialisé, il faut comparer les équivalents choisis du point de vue des CCP : DOMAINE expression française éléments étrangers à e.p. expression polonaise éléments communs aux deux expressions évent: exemples d’empois ratés éléments étrangers à e.p. Les domaines sont choisis individuellement ou suggérés par l’enseignant (sciences économiques, comptabilité, médecine, etc.). L’explicitation des emplois ratés a pour but de prévenir les calques éventuels des CCP de la langue maternelle. On sensibilise aux inconvénients des équivalents dictionnairiques qui suggèrent l’identité des étendues des termes comparés sans rendre compte des degrés de leur conventionnalisation. On conseille donc des recherches en bibliothèque, de travailler avec des manuels appropriés ou des recherches sur Internet, le degré de conventionnalisation du terme étant considéré comme proportionnel au nombre de ses occurrences. Restent encore les spécialistes dans le domaine ; certains étudiants travaillent déjà dans des entreprises bilingues. Le nombre des mots n’est pas déterminé, mais les lexiques comparés doivent créer un microsystème suffisamment cohérent. La réalisation de l’exercice n’est pas facile. On ne reçoit pas toujours, non plus, des résultats satisfaisants. L’exercice laisse cependant des empreintes durables dans l’esprit de l’élève et TRANSLATION AND MEANING, PART 10, 2013, 281 - 290 289 fait que dans son développement linguistique ultérieur, il sera beaucoup plus conscient des pièges qui l’attendent dans son travail autonome. Voici quelques exemples: Musique : éléments de notation nuta – une note (son ou son signe, à valeur déterminée ; terme courant) Cvfr = Cvpl, Cpfr = Cppl, Pfr = PPl ćwierćnuta – une noire (son ou son signe, val. déterminée ; courant ; valeur vs aspect) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl półnuta – une ronde (son ou son signe, val. déterminée ; courant ; valeur vs aspect) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl tonacja durowa/molowa – tonalité majeure/mineure (durs/mous vs majeurs/mineurs) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl c-d-e-f… – do-re-mi-fa… (système ang.-sax. vs solmisation ; b-moll=si/bémol mineur) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl nuty – partition (notation compos. musicale ; pol. quelques pages vs fr. toute longueur) Cvfr = Cvpl, Cpfr ≠ Cppl, Pfr ≠ PPl partytura – partition (notation comp. musicale ; pol. orchestrale vs fr. toute comp.) Cvfr ≠ Cvpl, Cpfr ≠ Cppl, Pfr = PPl próba – répétition (temps exercices d’une pièce ; courant ; essayer vs répéter) Cvfr = Cvpl, Cpfr ≠ Cppl, Pfr ≠ PPl mieć próbę – répéter (au moins 2 personnes doivent participer dans une próba pol.) Cvfr = Cvpl, Cpfr ≠ Cppl, Pfr ≠ PPl Exemples d’emplois non réussis (ratés) : Tu as oublié d’apporter les notes ! (Zapomniałeś przynieść nuty.) Tu as oublié d’apporter ta partition ! Comment tu veux qu’on fasse notre essai ? (Jak mamy teraz zrobić próbę?) Comment tu veux qu’on répète ? Comptabilité : éléments d’équation pour le résultat de l’entreprise : obrót – chiffre d’affaires (recettes brutes ; « tourner » de l’argent vs faire des affaires) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl próg – seuil (base imposable ; terme plus courant en pol. qu’en fr. ; profil de traverse) Cvfr ≠ Cvpl, Cpfr = Cppl, Pfr = PPl próg – assiette (base imposable ; courant dans les 2 langues ; traverse vs récipient) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl rachunek zysków i strat – compte de résultat (courant ; bénéfices et pertes vs résultat) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl dobra trwałe – immobilisations (écriture comptable ; courant dans les deux langues) Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl Football : termes choisis : aut – touche Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl spalony – hors-jeu Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl karny (rzut) – pénalité (cet équivalent français n’est employé que pour le rugby) Cvfr ≠ Cvpl, Cpfr = Cppl, Pfr = PPl karny (rzut) – penalty Cvfr = Cvpl, Cpfr = Cppl, Pfr = PPl pole karne – surface de réparation Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl faul – faute Cvfr = Cvpl, Cpfr = Cppl, Pfr ≈ PPl napastnik – avant 290 JACEK FLORCZAK Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl obrońca – arrière Cvfr = Cvpl, Cpfr = Cppl, Pfr ≠ PPl 12. Voies d’application en didactique de LE Les derniers chapitres font remarquer que les analyses proposées peuvent être exploitées d’une double façon : en lexicographie, afin de créer un nouveau type de dictionnaire bilingue et en didactique des langues étrangères, en tant qu’outil utile dans l’acquisition autonome de la compétence lexicale (pour le lexique de spécialité et pour le lexique général). Les deux manières d’appliquer ces analyses correspondent à la double présence du savoir sur la langue dans le processus d’enseignement/apprentissage de LE. Le lecteur du dictionnaire peut ne pas savoir comment le dictionnaire a été créé. Le destinataire des analyses susmentionnées crée lui même ce dictionnaire, tout en se servant du savoir-faire linguistique qui lui a été préalablement transmis. Si les deux peuvent travailler de manière autonome, c’est avant tout le second qui jouira de cette autonomie de manière appropriée. Averti des pièges qui le guettent et doté d’un savoir-faire lui permettant de les prévenir, il sera beaucoup plus prudent vis-à-vis des problèmes qui l’attendent et auxquels sera toujours plus exposé un apprenant non averti. Bibliographie Florczak, J. (2010), Językoznawcze aspekty modelu kształtowania kompetencji języka obcego. Editions de l’Université de Łódź, Łódź. Kacprzak, A. (2000), Terminologie médicale française et polonaise. Analyse formelle et sémantique. Editions de l’Université de Łódź, Łódź. Lyons, J. (1980), (trad. de l’anglais par J. Durand et D. Boulonnais), Sémantique linguistique. Larousse, Paris. Peirce, Ch.-S. (1978), (rass. et trad. de l’anglais par G. Deledalle), Ecrits sur le signe. Le Seuil, Paris. Rutkowski, P. (2001), “Małpa w Internecie”, in : Wiedza i człowiek 7(89)/2001 ; d’après Karen Steffen, http://linguistlist.org. Trier, J. (1934),“ Das sprachliche Feld. Eine Auseinandersetzung“, in : Neue Jahrbücher für Wissenschaft und Jugendbildung 10, pp.428-449. Wilczyńska, W. & Rabiller, B. (2010), Francuski. Słownik pułapek językowych. Wiedza powszechna, Varsovie. TRANSLATION AND MEANING, PART 10, 2013, 291 - 298 291 LE LATIN JURIDIQUE A L’EPREUVE DE LA TRADUCTION Ksenia Gałuskina Université de Silésie, Katowice, Poland Abstract: This paper focuses on an analysis of Latin expressions and maxims implemented in legal language. The frequency of use of Latin quotations varies from one legal culture to another. Moreover, each language has its own legal Latin expressions and maxims. However, independently of the language, the use of Latin quotations is one of the characteristics of legal discourse. The first part of the paper provides an overview of the characteristics of Latin quotations emerging in different types of legal texts. Then there is an attempt to discuss use of legal Latin in the context of comparative legal linguistics, illustrated by a series of examples from Polish and French legal texts. Finally, the issues of translation of Latin expressions and maxims in legal texts, which could be source of mistakes and misunderstanding, will be discussed in the second part of the paper. 1. Introduction L’usage des latinismes dans le langage juridique est un de ses traits caractéristiques, au moins pour tous les systèmes juridiques qui ont subi l’influence, directe ou indirecte, du droit romain. Quand même, leur usage dans les différentes langues reste inégal ce qui pose la question sur le comportement d’expressions et de locutions latines dans la traduction. Le langage juridique n’est pas seul qui se sert du latin. D’autres langages spécialisés recourent aussi au latin, notamment celui du domaine de la médecine qui en puise encore plus largement. En Pologne, les diagnoses médicales sont souvent écrites au latin ce qui les rende compréhensibles, par principe, pour tout médecin parlant n’importe quelle langue. Ainsi, fractura femoris dextri veut dire « la fracture du fémur droit » en français, « złamanie prawej kości udowej » en polonais, « right femur fracture » en anglais, etc. et fait référence à une diagnose et à un état de santé univoques. Dans le cas présent, nous nous concentrons sur le sens de ces expressions et nous laissons à part le problème de la fréquence (et les causes de sa diversité) de l’emploi du latin dans le discours médical qui se diffère d’une langue à l’autre. Cette illustration de l’usage du latin en médecine nous permet de constater qu’au niveau sémantique, il est une langue commune des médecins et son rôle dans la traduction est celui du tertium comparationis, soit de la base de comparaison. Il aide à trouver un équivalent univoque dans une autre langue, la relation sémantique entre le terme latin et le terme dans une autre langue est biunivoque. Le but du présent article est de montrer que le latin en droit ne joue pas le même rôle, ainsi que de présenter les conséquences de cet état de choses sur la traduction juridique. Il faut préciser que nous laissons à part d’autres phénomènes liés au rôle du latin et du droit romain (ainsi que du droit médieval, exprimé en latin) dans l’évolution d’institutions et de catégories du droit, notamment dans la formation du vocabulaire juridique. Dans le présent article, nous nous penchons sur les traces de cette influence sous la forme de mots et de locutions latins qui sont cités, en latin, dans les textes juridiques. Nous comprenons tous ces mots et expressions sous le terme de « latin juridique ». Il faut aussi remarquer que l’objet de la présente analyse s’inscrit dans la problématique plus vaste de l’emploi dans les textes de départ des expressions en troisième – par rapport à la paire des langues concernées – langue et de l’influence de leur présence sur le processus et les modes de traduction. Cependant, dans les textes juridiques (au moins dans le contexte contemporain), les expressions latines étant en troisième langue appartiennent en même temps, sur le niveau sémantique, au système de la langue de départ. Leur sens dépend essentiellement de cette langue et du système juridique concerné. Contrairement au latin en médecine, en biologie, etc., il n’existe pas le latin juridique universel. Chaque système juridique à son propre latin juridique, même si nous pouvons constater l’exis- 292 KSENIA GAŁUSKINA tence d’un tronc commun (cf. Mattila, 2002). En plus de cela, le latin juridique reste une langue « vivante » au sein de chaque système juridique indépendamment (cf. Cornu, 2005:405). La formation de nouvelles expressions et locutions latines ne s’est pas terminée avec la reine du latin dans le domaine du droit. Elle se produit toujours, séparément dans chaque système du droit national et international (cf. l’exemple de la maxime lex retro non agit formulée en latin au début du XXe siècle et employée en droit polonais, évoqué par Mattila 2006: 154). Pendant notre recherche concernant le latin en langage juridique, nous avons utilisé en tant que corpus spécialisés, les bases des données juridiques Lex (en polonais), Dalloz (en français), EURlex (pour les textes européens). Sauf mention contraire dans le texte, les exemples sont tirés des textes français et représentent le latin juridique français. 2. Le latin dans le droit Il est utile de commencer par la caractéristique et la classification des expressions latines apparues dans le discours juridique, notamment dans les textes juridiques de la doctrine et de la jurisprudence. Dans la tradition polonaise, l’emploi des latinisme est principalement restreint à ces types des textes juridiques. Cependant, les autres systèmes juridiques connaissent l’emploi de mots et d’expressions latins aussi dans les textes légistatifs (cf. Mattila, 2006 :137). Le latin dans le langage du droit a plusieurs facettes. Il sert à parler d’institutions et de catégories juridiques soit dans le contexte historique, soit dans le contexte contemporain. Dans le premier cas, on parle de l’histoire du droit romain ou médieval, par conséquent c’est le contexte d’un système juridique historique qui indique les modes d’interprétation de ces termes. Par exemple, le terme justum matrimonium (et ses variantes graphiques dans les différentes langues : matrimonium justum, iustum matrimonium et matrimonium iustum) renvoie d’une manière univoque au mariage légitime du droit romain (cf. Litewski, 1995 :159). Dans le deuxième cas, il faut interpréter le terme en question (et le concept du droit auquel il fait référence) par rapport à son évolution pendant des siècles dans un système juridique donné, c’est-à-dire dans le contexte du droit contemporain de l’État concerné. Il s’agit donc de mots et de locutions latins dont la connotation est juridique (qui possède un sens à l’égard du droit), comme la preuve prima faciae, l’action de in rem verso, l’assipiens, le solvens, brevitas causa, de lege ferenda, intuitus personae, la condition sine qua non et qui apparaissent dans le discours juridique contemporain. Il est important de remarquer que leur usage peut se varier, et il se varie très souvent, d’un système juridique à l’autre. Les exemples ci-dessus étant tirés des textes français, la personne parlant le polonais et connaissant son langage juridique remarque tout de suite qu’ils n’appartiennent pas tous (ou bien sous cette forme) à « son » latin juridique. Il semble que l’usage le plus caractéristique du latin dans le droit s’opère par la présence d’adages, maximes, principes, brocards latins. Tous ces termes sont employés parfois d’une manière interchangeable, mais ils se cachent derrière eux certaines nuances de la portée des locutions latines, ainsi nommées. Tous ses termes font référence aux formules juridiques courtes et résumant le fond d’un problème du droit. Quand même, le terme « adage », qui semble être aujourd’hui le plus répandu, a étymologiquement une forte réflexion morale, « brocard » représente une formule juridique sous une brève forme populaire, « maxime » est une formule qui serve de règle juridique ou principe général du droit (les définitions de tous ses termes sont loin d’être univoques pour différents auteurs ce qui explique leur interchangeabilité dans l’usage, cf. Vincent, Guinchard, 2009 :51; Cornu, 2004 :25, 121, 571, 707; Bissardon, 2009 :111, 147). Plusieurs de ces formules sont jusqu’aujourd’hui citées en latin, comme le principe contra non valentem, l’adage nemo plus juris ad alium transfere potest quam ipse habet, la maxime fraus omnia corrumpit, mais certains ont passé au français, comme « nul n’est censé ignorer la loi » qui est une variante française de l’adage latine ignorantia juris non excusat. TRANSLATION AND MEANING, PART 10, 2013, 291 - 298 293 Il existe encore deux catégories des latinismes qui dépassent la frontière de l’usage juridique. D’un côté, il s’agit des expressions comme op. cit., eod. loc., in fine, cf. supra, cf. infra, adde, ibid., verbo qui sont des mots de référence, employés souvent par abréviation. Et de l’autre côté, ce sont des locutions de type lato sensu, stricto sensu, sui generis, ad hoc qui, en leur acception juridique, servent souvent à déterminer l’extension d’un terme juridique et de type a contrario, a fortriori, a pari, a priori, mutatis mutandis, volens nolens qui, aussi en leur acception juridique, découle du raisonnement juridique et apparaissent comme les signes de l’argumentation juridique. Toutes les expressions et locutions latines, notamment les « petits mots » de cette dernière catégorie, possèdent une valeur rhétorique dans le discours juridique. À part de cela, le français a incorporé plusieurs mots juridiques du latin et les a francisés. Aujourd’hui, ils ne sont que les empruntes du latin, traité comme chaque autre mot français, p.ex. des mots et des locutions latins devenus français et ayant une connotation juridique comme le consensus, le quorum, le récépissé, le de cujus, le statu quo. Cependant, dans les textes juridiques, la plupart de ces termes peuvent être présents soit sous leurs formes latines (et en italiques), soit sous leurs formes francisées. 3. Le latin dans la jurilinguistique comparative La fréquence de l’emploi de mots et de locutions latins est très inégale dans les différentes langues, notamment dans leurs langages juridiques (cf. Mattila, 2006 :156). Au sein d’un même système juridique, elle est aussi inégale pour les différentes branches du droit. Ils semblent être beaucoup plus fréquents en polonais (et dans certaines autres langues p. ex. en anglais et en italien), qu’en français. Quand même, les textes doctrinaux français restent ornés par des latinismes, mais ils sont de moins en moins compréhensibles pour les jeunes juristes. Les raisons de cet état de choses sont multiples. D’un côté, la langue latine n’est plus une langue commune des juristes. Les étudiants français en droit n’apprennent plus la langue latine au cours de leurs études, ainsi que l’histoire du droit romain constitue seulement une partie du cours de l’histoire du droit. En Pologne, la présence des tels cours dans le programme d’études en droit explique la persistance plus forte de mots et de locutions latins dans le vocabulaire actif (pour certaines expressions) et passif des juristes, et par conséquent dans le discours juridique polonais. D’autre côté, il fait l’obstacle en communication avec des participants non-juristes de la vie juridique. Les gens qui viennent chez un avocat, un notaire ou un magistrat ne comprennent guère ces tournures de la rhétorique juridique. C’est pourquoi le mouvement international de plain language voit la réalisation de son but, à savoir la clarté des textes juridiques, entre autres, dans la suppression des latinismes dans ces textes comme appartenant au jargon des juristes (cf. Stępkowski, 2001). L’autre grand problème lié au postulat de la clarté des textes juridiques est la densité des locutions latines, la complexité de leurs sens (par les défenseurs des latinismes considéré comme avantage, cf. Roland, 2006 :VIII). Comme nous avons mentionné, en français, on observe que le recours aux latinismes est beaucoup plus fréquent dans les textes doctrinaux que dans les autres types des textes juridiques. Quand même, il faut souligner que la plupart d’eux restent obscurs et exigent la consultation d’un dictionnaire pour un juriste « moyen », au moins que l’expression mise en œuvre n’a pas été expliqué dans le texte, ce qui n’est pas rare et ce qui montre le problème de la compréhension des latinismes même par les juristes. En polonais et en anglais, les expressions latines d’usage courant dans le discours juridique sont d’un nombre plus élevé par rapport au français (cf. Houbert : 21) et n’exige pas (d’habitude et pour les expressions qui à cause de leur usage fréquent sont devenues familières pour un juriste « moyen ») des définitions supplémentaires ou la consultation des références externes à un texte donné. C’est qui est vrai pour les textes composés d’origine dans ces langues, ne l’est pas pour les traductions vers elles, notamment s’il s’agit des textes des documents européens comprenant un grand nombre des latinismes indifféremment de la langue. 294 KSENIA GAŁUSKINA Une recherche intéressante a été effectuée par H. E. S. Mattila (2002) qui a comparé le contenu des dictionnaires des expressions latines des différents pays, donc le vocabulaire latin potentiel d’un juriste formé dans une culture juridique donnée. Il en ressort un constat frappant que seulement une petite partie de ces expressions est commune pour les langages juridiques des pays représentant différentes cultures juridiques. Que cette constatation soit vraie montre notre propre expérience. Le moindre contact avec un autre système juridique permet de constater que l’intercompréhension des expressions latines dans le discours juridique est encore plus difficile. Les principaux obstacles sont l’absence de certaines expressions dans une langue (et présence d’autres méconnues par la première) et les diversités de la prononciation. Ainsi, nous avons passé plusieurs heures pour comprendre qu’un [tʁ ibynal adɔ ɡ ] (prononciation défectueuse au lieu de [tʁ ibynal ad’ɔ k], déjà suffisamment difficile à comprendre à cause de l’absence de « h » dans la prononciation) est un « tribunal ad hoc » bien connu. Et nous avons perdu toute la confiance en valeur du latin comme lingua franca des juristes après une incompréhension totale et décevante de la locution res judicata par les juristes français qui est forte présente dans le langage juridique polonais et complètement désuète en français (ou plus précisément, présente sous la forme de son équivalent français l’autorité de la chose jugée). En plus de cela, il existe de petits changements liés à la graphie de « i » semi-consonne [j] soit comme « i » en polonais (p. ex. res iudicata), soit comme « j » en plusieurs d’autres langues, notamment en français et en anglais (p. ex. res judicata) ce qui influence largement la prononciation dans ces deux langues. Ces expériences nous ont invités à nous poser la question concernant les stratégies d’équivalence des expressions latines dans la traduction juridique. Traduire ou ne pas traduire ? Et si traduire, comment le faire ? 4. Le latin dans la traduction Nous avons déjà mentionné le problème de la prononciation latine ce qui peut provoquer le malentendu et ce qui rend l’interprétation juridique (ici, dans le sens de la traduction effectuée à l’oral) du discours orné des expressions latines doublement difficile. D’un côté, il s’agit de tous les problèmes d’équivalence des latinismes et de l’autre côté, il s’agit de les prononcer selon les règles de la langue en question. Ensuite, les locutions latines fonctionnent dans chaque langue comme les expressions du plus haut degré de figement, la moindre modification de leur graphie, mais aussi de leur prononciation peut causer son incompréhension (cf. Mattila, 2002). En Pologne, on prononce le latin selon les règles de la prononciation évolués pendant des siècles du présence du latin dans l’enseignement, appélée la prononciation traditionnelle (cf. Jędraszko, 1988 : 6), qui est bien différente de la prononciation du latin classique, dite restituéee qui est enseigné en France (cf. Bissardon, 2009 :75). Quand même en France aussi, il existe la prononciation latine scolaire qui est la plus fréquente et qui néglige souvent les règles d’accentuation du latin, la prononciation du « h » et « e » dans toutes les positions et la prononciation correcte du « r » roulé latin. Cependant, le moins compréhensible semble être prononciation du latin en anglais qui constitue sur le niveau phonétique une mélange extraordinaire du « anglo-latin ». La principale stratégie envers les latinismes semble être celle de ne pas traduire, de les recopier dans le texte d’arrivée tels quels sont apparus dans le texte de départ. Nous trouvons cette solution inadaptée à la plupart des cas de l’usage de mots et de locutions latins dans les textes juridiques. Puisque les locutions latines sont une source inépuisable des pièges en traduction, nous voulons signaler quelques-uns dans la partie suivante du présent article. Notre analyse est non exhaustive et sert à illustrer le problème à partir de quelques exemples frappants. D’abord, il existe des mots et des locutions latins qui ont l’air d’être de la même forme dans deux langues, mais qui ne le sont pas. Par exemple, à la locution « polonaise » ab intestato correspond en TRANSLATION AND MEANING, PART 10, 2013, 291 - 298 295 français la locution ab intestat qui a perdu le « o » final dans sa graphie contemporaine. Les traducteurs des documents de l’UE ne savaient pas toujours échapper à ce piège. Voyons deux exemples : (1) PL : « dziedziczenie ab intestato » ; FR : « succession ab intestato » ; EN : « intestate succession » ( dont l’équivalent polonais est « dzidziczenie beztestamentowe/ustawowe » et français reste toujours « succession ab intestat » ; en même temps, l’anglais connaît le terme « succession ab intestato »)1. (2) PL : « dziedziczenie ab intestato » ; FR : « succession "ab intestat" » (graphie originale) ; EN : « intestate succession »2. Il faut mentionner que la locution adverbiale ab intestat forme simplement les groupes verbaux en français, comme « mourir ab intestat », « héritier ab intestat ». Pendant qu’en polonais une telle construction sera fautive (ces constructions sont marquées par l’astérisque) et exigera la traduction de la locution latine en polonais : umrzeć nie pozostawiając testamentu (*umrzeć ab intestato), dziedziczyć ustawowo/bez testamentu (*dziedziczyć ab intestato). L’autre exemple trompeur présente la locution du latin « polonais » status quo qui veut dire « l’état actuel des choses » : utrzymać status quo, zachować status quo, bronić status quo. En polonais, qui est une langue flexionnelle, il faut décliner les mots déclinables latins selon le cas approprié (dans notre exemple, exigé par le verbe). Cependant, on observe qu’aujourd’hui, il est plus en plus fréquent de les garder au nominatif comme dans les acceptions de la locution status quo citées ci-dessus, notamment là où on a affaire à l’accusatif dont la forme est souvent égale à la forme du nominatif en polonais. Mais la forme correcte reste celle qui est déclinée : utrzymać statum quo (accusatif), zachować statum quo (accusatif), bronić status quo (génitif). Le français juridique connaît l’expression latine statu quo qui provient de la locution in statu quo ante qui veut dire « dans l’état où les choses était auparavant » et, conformément à cette étymologie, peut signifier soit la situation actuelle (son maintien), soit la situation préexistante (son rétablissement) (cf. Lexique Dalloz :679). On observe donc que la locution contemporaine statu quo ignore (encore une fois) le fait que le latin est une langue flexionnelle et que le même nom a plusieurs formes grammaticales en déclinaison. Ainsi, le nom masculin de la 4e déclinaison status (au nominatif) est présente en français sous sa forme de l’ablatif statu, même s’il est utilisé d’une manière indépendante (notamment dans le sens de la « situation actuelle ») par rapport à sa source. En plus de cela, en français, les locutions latines nominales sont précédées du déterminant qui est d’habitude un article défini : maintenir le statu quo, conserver le statu quo, défendre le statu quo. Les cooccurrences de ces deux locutions en polonais et en français montre qu’elles sont correspondantes, quand même pour le sens de la « situation préexistante » il sera beaucoup plus correcte à utiliser en polonais la locution status quo ante (cf. Jędraszko, 1988: 277). Ainsi, au lieu d’écrire « przywrócić status quo », il sera donc mieux employer « przywrócić status quo ante ». La deuxième locution est aussi acceptable en français sous la forme de statu quo ante : « restituer le statu quo ante », mais le sens du verbe lève suffisamment l’ambiguïté de cette locution et il semble plus correcte à la laisser sous sa forme habituelle en français : « restituer le statu quo ». L’anglais, quant à lui, connaît les deux locutions et la différence de la forme marque la différence du sens. La locution du latin « anglais » in statu quo (qui constitue la forme raccourcie de in statu quo 1 Livre vert - Successions et testaments, Numéro CELEX: 52005DC0065. Avis du Comité économique et social européen sur le Livre Vert sur les successions et testaments, Numéro CELEX: 52005AE1242. 2 296 KSENIA GAŁUSKINA ante), qui coexiste en usage avec la locution status quo ante, signifie « dans l’état des choses précédent » et se distingue de status quo qui signifie « la situation actuelle ». Ainsi, l’expression anglaise « to maintain status quo » sera traduite en polonais « utrzymać status quo » et en français « maintenir le statu quo », pendant qu’une autre « to return in statu quo », sera traduite « przywrócić status quo ante » en polonais et « restituer le statu quo » en français. Le problème suivant réside dans la restriction des cooccurrants de certaines locutions que nous avons mentionées plus haut sur l’exemple des locutions ab intestat. Cette locution peut former les cooccurrences avec des verbes en français, pendant que ab intestato en polonais ne le peut pas. De bonnes illustrations constituent aussi d’autres locutions adverbiales, p. ex. ad solemnitatem et ad personam. En français, la locution adverbiale ad solemnitatem forme les cooccurrances suivantes : date ad solemnitatem, condition ad solemnitatem, écrit ad solemnitatem, titre exigé ad solemnitatem, formalité ad solemnitatem, forme (authentique) requise ad solemnitatem, écrit s’impose ad solemnitatem, formalisme ad solemnitatem, authenticité d’un contrat requise ad solemnitatem. En même temps en polonais, seules les constructions se référant à la forme sont acceptables : forma (pisemna, aktu notarialnego, czynności prawnej) ad solemnitatem, forma zastrzeżona ad solemnitatem. D’autres constructions françaises exigent la traduction par les termes polonais appropriés, p. ex. « data pewna » pour « date ad solemnitatem ». Dans le langage juridique français, on emploie aussi la locution ad validitatem qui est synonymique à la locution ad solemnitatem et apparaît dans les mêmes constructions se référant à la forme d’un acte juridique (cf. Roland, Boyer, 1998 :21). Cependant, cette locution est méconnue par le polonais et ne peut pas apparaître dans un texte en polonais. La locution adjectivale ad personam est polysémique en français et monosémique en polonais. Dans les constructions de type « argument, attaque ad personam », elle est dotée d’un sens extrajuridique et signifie une attaque dirigée contre une personne déterminée. Cet usage correspond à son sens en polonais : argument, atak ad personam. En français, elle possède encore un autre sens de « à titre personnel », comme dans les expressions titre/charge/privilège/droit ad personam. Alors l’expression (ou plutôt une néologie) « représentant ad personam » rencontrée dans les documents européens semble être acceptable en français, mais son équivalent polonais « przedstawiciel ad personam » doit être jugé incorrecte en tant que calque d’une autre langue (peut-être du français ou de l’anglais « ad personam replacement ») qui néglige le sens univoque de la locution ad personam en polonais3. Vu que les locutions latines sont présentes dans différents langages spécialisés, il peut arriver qu’une locution de l’usage commun dans une langue appartient à un domaine bien restreint dans une autre. C’est le cas de la locution adverbiale lato sensu, qui signifie « au sens large », dont l’équivalent polonais est une autre locution latine, à savoir sensu largo. Toutefois, la locution sensu lato (après changement d’ordre des mots) est présente en polonais, mais exclusivement pour la description des espèces végétales et animales en biologie et en zoologie (cf. Jędraszko, 1988 :265; Smoczyk). Ainsi, la construction « produkty rolne sensu lato » (« produits agricoles lato sensu » en français et « agricultural products sensu lato » en anglais) qui apparaît dans les documents européens est fautive en polonais4. Comme nous avons observé dans le paragraphe précédent, les expressions connues par deux langues en question peuvent avoir un ordre de mots renversé. Cette modification qui n’entraîne pas de changement du sens n’est pas rare. Elle concerne les locutions qui n’ont qu’une valeur rhétorique dans le 3 Cf. Comité des régions – Règlement intérieur, Numéro CELEX: 32010Q0109(01). Cf. Proposition de décision du Conseil relative à la position que l'Union doit adopter au sein du comité intérimaire UE Serbie concernant son règlement intérieur et notamment le mandat et la structure des sous comités UE-Serbie, Numéro CELEX: 52010PC0022. 4 TRANSLATION AND MEANING, PART 10, 2013, 291 - 298 297 texte, p. ex. l’équivalent en polonais de sensu stricto est stricto sensu, ou bien de la locution volens nolens est écrite habituellement avec un ordre inverse en polonais, à savoire nolens volens (même si l’ordre « français » correspond à l’ordre latin original et à la portée de cette expression). Les adages aussi subissent parfois certaines modifications. À titre d’exemple, nous pouvons citer un adage présent en polonais ignorantia iuris nocet qui est rendu en français par un autre adage latin ignorantia juris non excusat (dont la portée est moins rigoureuse) ou bien par un adage français nul n’est censé ignorer la loi dans l’usage courant. Ou encore, le principe général du droit exprimé en polonais par l’adage lex specialis derogat generali, est exprimé en français sous la forme modifiée, à savoir specialia generalibus derogant (cf. Roland, Boyer 1999 :843). L’adage latin nasciturus pro iam nato habetur quotiens de commodis eius agitur, qui est la source du terme nasciturus en polonais, est connu en français sous la forme infans conceptus pro jam nato habetur quotiens de commodis ejus agitur (cf. Roland, Boyer : 1999 :327). La modification de ce dernier adage provoque aussi l’absence du terme nasciturus en français qui peut être traduit soit par le terme français « enfant à naître », soit par la locution latine infans conceptus. Les constructions « partiellement » latines peuvent aussi causer les problèmes dans la traduction. Nous les nommons, d’après Sourioux et Lerat (1975 :26-27), les mixtes franco-latins. En français juridique, il existe plusieurs locutions qui sont composées sur la base d’une locution latine dont un des mots latins est remplacé par son dérivé français, comme dans les locutions « une condition sine qua non » et « une action de in rem verso ». Elles peuvent se traduire vers le polonais soit à l’aide de la locution latine complète, c’est-à-dire conditio (ou condicio) sine qua non et actio de in rem verso, soit en gardant cette mixité linguistique par « warunek sine qua non » et « skarga de in rem verso ». La dernière locution pourra exiger une explication qu’il s’agit d’un nom habituel d’une action du droit français permettant d’agir en cas de l’enrichissement sans cause et non plus d’une action du droit romain (ce qui fait le contexte habituel pour cette locution en polonais). Les exemples évoqués ci-dessus de nasciturus et de res iudicata montre le phénomène de l’existence dans le langage juridique d’un pays donné des latinismes qui sont méconnus par un autre système juridique. C’est aussi le cas des mots latins du français juridique, comme de cujus, accipiens, solvens qui ne sont pas employés en polonais juridique. Il s’agit ici d’une situation quand les catégories juridiques de deux systèmes en question existent et se correspondent, mais elles diffèrent par l’usage de leurs noms en latin et demandent une traduction par des termes appropriés dans la langue d’arrivée. La situation est plus complexe avec des noms latins des institutions propres à un système juridique en question. Un bon exemple fournit le terme culpa in contrahendo employé en droit polonais pour la responsabilté dans les relations précontractuelles. La conception de la période précontractuelle en droit polonais et français est totalement différente. En France, le fondement de l’indemnisation du préjudice causé par l’annulation du contrat est délictuel, pendant qu’en Pologne (depuis 2003), il est contractuel. Il s’agit donc du manque de correspondance entre deux systèmes juridiques concernés, la culpa in contrahendo proprement dite n’existe pas en droit français (même si l’on rencontre dans les textes doctrinaux les constructions du type « culpa in contrahendo à la française »). Par conséquent, la traduction de ce terme vers le français peut se produire soit par l’équivalent générique, c’est-à-dire « la responsabilité précontractuelle », soit par la reprise de l’expression latine avec une explication, si le contexte l’exige. 5. Conclusion En conclusion, l’emploi des latinismes dans les textes juridiques peut causer de nombreux problèmes pendant leur traduction. Nous pouvons adopter de différentes stratégies et techniques de traduction. Certes, dans de multiples cas, il est possible d’employer le « même » latinisme après l’établissement du lien d’équivalence de son emploi dans deux langues concernées. Dans tous les cas, l’apparition 298 KSENIA GAŁUSKINA des expressions latines dans le texte de départ ne permet pas au traducteur de les mettre entre parenthèses et considérer comme « hors traduction ». Au contraire, le traducteur est obligé de les inclure au processus de la traduction en adoptant les techniques de traduction appropriés à la stratégie choisie. Bibliographie Bissardon, S. (2009), Guide du langage juridique. Vocabulaire, pièges et difficultés, LexisNexis Litec, Paris. Cornu, G. (dir.) (2004), Vocabulaire juridique, Qudrige/PUF, Paris. Cornu, G. (2005), Lingustique juridique, Montchrestien, Paris. Houbert, F. (2005), Guide pratique de la traduction juridique. Anglais – Français, La Maison du Dictionnaire, Paris. Jędraszko, Cz. (1988), Łacina na co dzień, Nasza Księgarnia, Warszawa. Litewski, W. (1995), Rzymskie prawo prywatne, Wydawnictwa Prawnicze PWN, Warszawa. Mattila, H.E.S. (2002), “De Aequalitate latinitatis jurisperitorum” (Le latin juridique dans les grandes familles de droit contemporaines à la lumière des dictionnaires spécialisés), Revue internationale de droit comparé, vol. 54, no 3, pp. 717 – 758. Mattila, H.E.S. (2006), Comparative Legal Linguistics, Ashgate, Aldershot. Roland, H. (2006), Lexique juridique. Expressions latines, Litec, Paris. Roland, H., L. Boyer (1998), Locutions latines du droit français, Litec, Paris. Roland, H., L. Boyer (1999), Adages du droit frnaçais, Litec, Paris. Smoczyk, M., Skróty (n.d.), “wyrazy i wyrażenia łacińskie używane w biologii”, http://www.biolog.pl/content-55.html (consulté le 30 août 2010). Sourioux, J.-L., P. Lerat (1975), Le langage du droit, PUF, Paris. Stępkowski, A. (2001), “Maksymy prawne na wyspach brytyjskich”, [in:] W. Wołodkiewicz, J. Krzynówek, Łacińskie paremie w europejskiej kulturze prawnej i orzecznictwie sądów polskich, Liber, Warszawa. Vincent, J., S. Guinchard (éds.) (2009), Lexique des termes juridiques, Dalloz, Paris. Dictionnaires consultés Dictionnaire du logiciel Antidote RX. Garner, B.A. (2004), Black’s Law Dictionary, Thomson West, St. Paul. Le Trésor de la Langue Française informatisé (dictionnaire consultable sur le site www.cnrtl.fr). Sobol, E. (red) (1995), Słownik wyrazów obcych, PWN, Warszawa. TRANSLATION AND MEANING, PART 10, 2013, 299 - 306 299 LEGAL TERMS AND MEANINGFUL UNITS. A CORPUS-BASED STUDY OF THE TERM DISCOVERY IN THE UNITED STATES SUPREME COURT OPINIONS. Stanisław Goźdź-Roszkowski University of Łódź, Poland Abstract: This article demonstrates a different perspective on investigating legal terminology by focusing on various types of textual recurrence. It discusses the results of a corpus-based analysis of discovery, a term central to the US trial practice and criminal proceedings in a corpus of US Supreme Court opinions. The analysis starts from investigating collocational patterns, both nominal and verbal. It describes meanings resulting from the emerging cooccurrence patterns between the term and various word forms identified in its co-text. This article refers to the concepts of semantic preference and semantic sequence (Hunston, 2008) to show how discovery tends to be found with a range of different word forms albeit all belonging to the meaning group of’ limitation’ or ‘restriction’. The connotational value of the term is presented by providing ample textual evidence of discovery found in co-texts where writers express their unfavourable evaluation towards this legal concept. By drawing on Hunston’s (2008) recent approach to the study of textual recurrence, the analysis illustrates how corpus linguistics can complement more traditional approaches to terminology description. 1. Introduction This paper aims to discuss the results of a corpus-based analysis of a selected legal term, i.e. discovery, with a view to underscoring the importance of contextual and phraseological information in terminology description. The analysis is based on a corpus of US Supreme Court opinions compiled by the author of this paper and aided by the linguistic software, viz. Wordsmith Tools. The study attempts to rely on certain concepts proposed in corpus linguistics research to inform terminology practice. This study reflects the recent trend which attempts to bring the study of terminology back to the study of real language usage (cf. Temmerman, 2000). Thus, a more descriptive approach is advocated, whereby legal terminology is studied in texts, which implies the study of parole, i.e. real language usage. Further, it is argued that any descriptive terminological work should widen its scope by examining the co-texts in which a term is found with a view to uncovering its relevant patterns and phraseological items. This paper emphasizes the role of language patterns as one of the basic constructs employed by institutions to build up our linguistic, conceptual and ideological view of the world which may convey messages implicitly without the reader being intuitively or consciously aware. Different types of recurrence have been variously described as collocational frameworks (Renouf & Sinclair, 1991), lexical bundles (Biber et al., 1999), formulaic sequences (e.g. Wray, 2002), grammar patterns (e.g. Hunston and Francis, 2000), etc. Increasingly, such concepts have been employed in specialized discourse. Schulze and Römer (2008) in their introduction to the special issue of International Journal of Corpus Linguistics: Patterns, Meaningful Units and Specialised Discourses emphasize the inextricable link between the performativity in language and the way “speakers and writers do things by predominantly and unconsciously employing patterns or phraseological items, i.e. strings of words that are highly structured, well-organized and firmly entrenched in the human being’s mind“ (2008: 1). Studies of specialised language collected in that issue demonstrate how specialised meanings are expressed and encoded by means of phraseological patterns and how they are linked to the particular domains in which they were used. The present paper aims to contribute to this growing body of research by investigating recurrent patterns of a legal term discovery. To this end, the corpus linguistics methodology has been employed drawing upon such concepts as semantic sequence (Hunston, 2009), semantic preference and semantic prosody. The benefits of corpus-based methodology terminology research and practice can hardly be overestimated. Over a decade ago Bowker (1996) was one of the first scholars advocating this approach 300 STANISŁAW GOŹDŹ-ROSZKOWSKI and pointing out its three major advantages. First, machine-readable corpora combined with the requisite tools enable one to reduce the time needed to research candidate terms. Large quantities of data can be processed rapidly exposing terminologists to a large number of conceptual descriptions. Having recourse to well-annotated corpora allows terminologists to focus on those parts of the texts that are terminologically relevant (Bowker, 1996: 31-32). Thus, with the right corpus and tools, it is possible to gain access to knowledge-rich contexts (Bowker, 1996) useful for conceptual analysis. Second, corpus data facilitates the examination of syntactic and semantic information encoded in linguistic patterns which are difficult to intuit or observe when scanning texts manually. This type of information can be retrieved through the use of concordances, also known as key words in context (KWIC). Concordances can provide information on the combinatorics of terms. Third, unlike in the case of conventional term banks, corpora provide a wealth of examples. Bowker (1996: 32-33) rightly observes that corpora present a variety of contexts as well as more extensive contexts. The accumulation of contexts is essential if one aims to examine whether a particular term is found in recurrent regular patterns of co-occurring items. In what follows, I turn to present the data and methodology adopted in this study (Section 2) and then discuss the findings (Section 3). 2. Data and Methodology The data used in the analysis consists of 114 different opinions given by the Supreme Court of the United States of America. These were collected at random from the period between 2000 and 2007. The corpus totals 1,182,246 words. Under the largely judge-made law system, the Supreme Court opinions represent the primary vehicle through which American law develops (Lee, Hall and Hurley, 1999). The Supreme Court is regarded as the national symbol of justice. It is the highest court in the United States consisting of the Chief Justice and eight Associate Justices. The U.S. Supreme Court has both original and appellate jurisdiction, but it exercises the former only in rare instances. The primary task of the Court is thus appellate. In that capacity it serves as the final arbiter in the construction of the Constitution of the United States and it provides a uniform interpretation of the law. However, to a large extent, it attempts to adhere to precedent, the well-known doctrine of stare decisis, i.e. “let the decision stand”. In this way, the precedents are given the authority of established law. It is thus little wonder that learning to read and understand cases, especially those given by the US Supreme Court, is a fundamental task that is mastered in all law schools. Worth bearing in mind is that the term opinion can be used in at least two somewhat different senses. First, more generally, an opinion may refer to the official decision of a court of justice and it is then interchangeable with the term judgment, which is defined as “the official and authentic decision of a court of justice upon the respective rights and claims of the parties to an action or suit therein litigated and submitted to its determination. The final decision of the court resolving the dispute and determining the rights and obligations of the parties” (Black, 1990). Second, the term opinion is also used to denote the reason which the court gives for its decision. It then refers to the statement made by a judge or court of the decision reached in connection with a case heard before them. Such a statement explains the law as applied to the case and provides the reason on the basis of which the judgment is made. The analysis presented below is based on opinions in the second sense, i.e. strictly judicial discourse providing legal argumentation for the decision reached. This study focuses on the legal term discovery, which denotes a significant concept in American legal procedure. The authoritative Black’s Law Dictionary clarifies this term in the following ways: First, in the general sense, discovery means “the ascertainment of that which was previously unknown; the disclosure or coming to light of what was previously hidden”. More specifically, it is employed as a fundamental concept in the trial practice where it is attributed the sense of “Disclosure by defendant of facts, deeds, documents or other things which are in his exclusive knowledge or possession and which are necessary to party seeking discovery as a part of a cause of action pending”. In this sense, discovery is aided by the following tools: “depositions, written interrogatories, productions of documents or things, permission to enter upon land, etc.” (Black’s Law Dictionary). As can be seen, discovery is a highly technical term used to denote a fact- TRANSLATION AND MEANING, PART 10, 2013, 299 - 306 301 finding process based on the assumption that a free exchange of information regarding a particular case is beneficial to the parties in that it helps to uncover the truth. In what follows we examine the contexts in which discovery is found in the hope of identifying prevalent patterns of using this term in judicial practice. The data was handled using the popular suite of text-processing programmes known as the WordSmith Tools (published by Oxford University Press since 1996, now at version 5.0 http:// www. lexically.net/wordsmith/). The collocational analysis was carried out using the in-built collocate feature of the Concord. The collocate horizon was set at 5, which means that the programme searched for potential collocates within five words to the right and left of the node word, i.e. the term under investigation. 3. Results The Concord search of discovery has returned 205 instances of this word found in the entire corpus. The initial investigation focused on the collocates identified within the span of five words to the right and to the left of the node word. The following collocational patterns have emerged: • N + 1 collocates (lemmas): • discovery <requests (26), order (s) (29), plan (13), process (7)> The figures in ( ) brackets denote the frequency with which a given word collocates with the term discovery. Discovery is frequently found in two-word nominal phrases closely related to the procedure where it serves a modifier, as for example in (1) and (2): (1) (2) When petitioner, an attorney representing a plaintiff, failed to comply with certain discovery orders, the Magistrate Judge granted the respondent's motion for sanctions against petitioner under Federal Rule of Civil Procedure 37(a)(4). Some of the sanctions in this case were based on the fact that petitioner provided partial responses and objections to some of the defendants' discovery requests. There is a clear and strong co-occurrence between discovery and requests, order, plan and process. Apart from the noun + noun combinations, there are several important verbal collocates immediately (within three words to the left) preceding discovery as represented by the V + N pattern. • <allow, object, conduct, narrow, obtain, reject, order, seek, submit, vacate> In terms of frequency, the verbal collocations are weaker than the nominal ones and they display a wider range of lexical choice. (3) (4) Moreover, even if this Court affirms the decision below and allows discovery to proceed in the District Court, the issue that would ultimately present itself still would have no bearing upon the reputation and integrity of Richard Cheney. The District Court ordered discovery here, not to remedy known statutory violations, but to ascertain whether FACA's disclosure requirements even apply to the NEPDG in the first place. The sentence examples (3) and (4) illustrate how courts operate with respect to discovery. They are active agents overseeing the process of discovery ensuring that it complies with the requisite law. Obviously, parties involved in a litigation can also influence this process, as illustrated by the following examples: (5) Banks attached affidavits from Farr and Cook to a February 1999 motion seeking discovery and an evidentiary hearing 302 STANISŁAW GOŹDŹ-ROSZKOWSKI (6) The Government, however, did in fact object to the scope of discovery and asked the District Court to narrow it in some way. Excerpt (6) shows the co-occurrence of discovery and a word related to the degree to which discovery is expected to be conducted. Interestingly, it can be also seen that discovery can be a contentious issue between the parties to litigation or a litigant and the overseeing court. Undoubtedly, there are conflicting views as for the extent to which the process of discovery should be carried out, with one party often insisting on some degree of its limitation. The issue of discovery scope tends to play a role upon the examination of adjectives which precede this term. • discovery <broad (8)> There are eight instances when the term is immediately preceded by the word broad, as for example, in excerpts (7) and (8): (7) (8) Thus, granting broad discovery in this case effectively prejudged the merits of respondents' claim for mandamus relief--an outcome entirely inconsistent with the extraordinary nature of the writ. Broad discovery should be encouraged when it serves the salutary purpose of facilitating the prompt and fair resolution of concrete disputes. These excerpts contain a strong evaluative element in the judicial argumentation pointing either towards the advantages or disadvantages of broad discovery in pretrial proceedings. Interestingly, broad is not restricted solely to the one sentential position. It can be more freely distributed as in the following example: (9) Although acknowledging that the scope of respondents' discovery requests was overly broad, the appeals court nonetheless agreed with the District Court that petitioners should bear the burden of invoking executive privilege and of objecting to the discovery orders with detailed precision. Worth noting is the presence of the adverb overly which clearly carries negative evaluation of the scope of such discovery. The data shows that discovery also collocates with the lemma narrow. There are several attested instances of limiting the scope of discovery by using this lemma. • discovery <narrow + N (12)> (10) (11) (12) (13) Mindful of "the judiciary's responsibility to police the separation of powers in litigation involving the executive," the Court of Appeals expressed confidence that the District Court would "respond to petitioners' concern and narrow discovery to ensure that [respondents] obtain no more than they need to prove their case." Id., at 1106. The District Court, moreover, did not err in failing to narrow discovery on its own initiative; The Government, however, did in fact object to the scope of discovery and asked the District Court to narrow it in some way; Although the Court cites United States v. Poindexter, 727 F. Supp. 1501 (DC 1989), as "sound precedent" for district-court narrowing of discovery, see ante, at 19-20, the target of the subpoena in that case, former President Reagan, unlike petitioners in this case, affirmatively requested such narrowing, 727 F. Supp., at 1503. Examples (10) through (13) show that the co-occurrence between discovery and narrow can be expressed through different word forms found in various textual positions. It appears that, perhaps, the broad v. narrow opposition might be to some extent indicative of the way in which discovery is conceptualized in the judicial discourse. In other words, there might exist some semantic consistency TRANSLATION AND MEANING, PART 10, 2013, 299 - 306 303 between discovery and certain co-occurring items. In order to validate this claim, more evidence of such items would be needed. It should be stressed at this point that the traditional collocational analysis which looks at the strength of collocational relations between the node word and an individual word form might offer only limited information on the collocational behavior of the term in question. However, it did reveal a certain semantic regularity represented by items such as scope, narrow, broad. The next step was therefore to study each and every concordance line in the hope of identifying more lexical evidence. Indeed, the search has proved to be extremely rewarding showing that there are several words which could be subsumed under the ‘restricting the scope of discovery’ category. (14) First, as the Court of Appeals recognized, see supra, at 8-9; infra, at 11, should the Government so move, the District Court could contain discovery so that it would not be "wideranging." (15) Indeed, the appeals court has already suggested tailored discovery that would avoid "effectively prejudg[ing] the merits of respondents' claim; (16) Without taking up the District Court's suggestion of that court's readiness to rein in discovery, see 219 F. Supp. 2d, at 54, the Government, on behalf of the Vice President, moved, unsuccessfully, for a protective order and for certification of an interlocutory appeal pursuant to 28 U. S. C. §1292(b). (17) A district court is not subject to criticism if it awaits a party's motion before tightening the scope of discovery; (18) On limiting discovery to the issue of membership, the Court of Appeals indicated its agreement; Excerpts (14) through (18) provide further evidence of different words expressing the idea of restricting the scope of discovery. These are verbs, such as contain, rein in, tighten or limit or adjectival forms like tailored and wide-ranging. Note that even wide-ranging is used in the negative context, where it is indicated that it is not a desirable situation to have ‘a wide-ranging discovery’. The opposite seems to prevail, that is, discovery should be clearly kept in check. Quantitatively, individual word forms expressing the restrictive attitude toward discovery are not numerous (sometime merely single or a few instances have been found) but they may appear in other related word forms as in a tightly-reined discovery, reined-in discovery, discovery-tailoring measures, limited discovery, keep discovery within appropriate limits, reduction of the District Court’s discovery, object to the scope of discovery, etc. The findings provided above can be viewed in terms of the concept of semantic preference. This concept is understood here as “the relation between the node word and individual word-forms which co-occur frequently with it”. (Stubbs, 2009: 22). Thus, it seems that there is a semantic preference for the term discovery and the lexical field of restrictive vocabulary. Strictly, there are also words indicating the expansion of the scope of discovery like wide-ranging or extensive but they are not as numerous and frequent as those expressing the restrictive scope. (19) The District Court approved respondents' extensive discovery plan, which included detailed and far-ranging interrogatories and sweeping requests for production of documents. There is yet another concept useful in the description of recurrence or regularity in texts. In 2008, in her article “Starting with the small words. Patterns, lexis and semantic sequences”. Susan Hunston proposed the concept of semantic sequence. Semantic sequence is defined as “recurring sequences of words and phrases that may be very diverse in form and which are therefore more usefully characterized as sequences of meaning elements rather than as formal sequences” (2008: 271). In addition, the semantic sequence consists of “the core word, item’, the complementation pattern or patterns associated with that word (such as a that-clause, wh-clause, or a prepositional phrase with a specific preposition) and a number of phrase types occurring before the core word which are, in spite of being 304 STANISŁAW GOŹDŹ-ROSZKOWSKI diverse in form, consistent in terms of meaning” (2008: 272). The linguistic material analyzed in this study seems to fit in with the first characteristic of a semantic sequence, i.e. the co-occurrence between the word or term and a range of various word forms referring to the same semantic field. As can be seen in the examples provided above, discovery tends to be found with different word forms albeit linked to the semantic value of limitation or restriction. There is definitely an emerging pattern which could be generalized in the following manner: • legal or natural person (court/government/defendants/plaintiffs) + ‘delimited scope’ vocabulary [e.g. narrow, broad, tailored, contain, wide-ranging, etc.] + discovery Quantitatively, this pattern accounts for approximately 30% of all instances. In subject positions, there is usually a legal or natural person, such as court or plaintiff, co-occurring with the lexical set from the ‘delimited scope’ field and the term discovery. It should be stressed that the sequences are by no means fixed in that they do not follow any specific formal structure. If we look at the other characteristic of a semantic sequence, then, clearly the pattern proposed above does not contain any prevalent complementation pattern. Thus, it is not entirely clear whether the discovery pattern could qualify as a semantic sequence as defined by Hunston (2008). Still, this concept looks promising because it might help to account for the inherent semantic quality in the term discovery. The description of meaning in terminology tends to focus on the denotational value admitting of little if any connotational value. In fact, the connotational significance of lexis has not always been widely recognized. Partington (1998: 65) provides a brief but useful overview of how connotation has been defined and perceived in traditional semantics. The term connotation has been defined, for example, as “the vaguer associations of a word for a group or individual” (Cook, 1992: 8) or in terms of “secondary implication” of an item (Lyons, 1977). Overall, connotation was perceived as “incidental to language rather than an essential part of it (Leech, 1974: 15). The potential connotations of terms appear to be beyond the realm of traditional terminology, which tends to focus on terms alone without paying much attention to the context of their use (cf. Wüster, 1991). In the case of the term discovery, connotation, especially expressive connotation, seems to play an important role in understanding its underlying concept. In what follows, the term discovery has been examined in order to find out whether its use involves favourable or unfavourable evaluation by the writer towards what they describe. The analysis has revealed that in the overwhelming majority of cases, discovery is found in negative contexts. First and foremost, discovery is often mentioned in the context of abusing its original purpose, i.e. ascertaining the facts of the case (excerpts 20 and 21). (20) (21) Allowing a plaintiff to forgo giving any indication of the economic loss and proximate cause would bring about the very sort of harm the securities statutes seek to avoid, namely the abusive practice of filing lawsuits with only a faint hope that discovery might lead to some plausible cause of action. In 1970, the prerequisites for imposing sanctions were redesigned "to encourage judges to be more alert to abuses occurring in the discovery process." Discovery tends to be perceived as a potential threat to the smooth, speedy and effective trial practice. The judiciary appears to be concerned that discovery procedure could manipulated by unscrupulous litigants and their council. Protecting the interests of the parties is a recurring theme in such contexts (Excerpts 22 and 23) as can be seen particularly well in Excerpt (23) which provides the justification for the disposition handed down by a judge. (22) (23) To permit an immediate appeal from such a sanctions order would undermine the very purposes of Rule 37(a), which was designed to protect courts and opposing parties from delaying or harassing tactics during the discovery process. In accord with the Court of Appeals, I am "confident that [were it moved to do so] the district court here [would] protect petitioners' legitimate interests and keep discovery within TRANSLATION AND MEANING, PART 10, 2013, 299 - 306 305 appropriate limits." 334 F. 3d, at 1107.9 I would therefore affirm the judgment of the Court of Appeals. There are many more examples of negative discourse surrounding the term discovery. A selection of such abbreviated examples is provided below: • • • • • the contention that discovery would be unduly burdensome; reined-in discovery--is surely a doubtful proposition the Government mentioned "excessive discovery" in support of its plea to be shielded from any discovery; There are no checks in civil discovery analogous to the constraints imposed in the criminal justice system to filter out insubstantial legal claims; Pretrial motions and discovery consumed three years; The negativity of the contexts in which discovery is found is not related to a single expression or pattern. Instead, we note the presence of a plethora of diverse lexical items which signal a negative stance on the part of the writers. In some cases, the negative evaluation should be inferred, as in the last two examples, where the period of three months suggests a protracted procedure or the absence of checks in the civil law discovery is viewed as a serious drawback. 4. Conclusions This brief study has attempted to demonstrate a different perspective on investigating legal terminology by focusing on a different type of textual recurrence. First, I documented the most common nominal, adjectival and verbal combinations in which the term discovery is found. This basic collocational analysis has made it possible to notice a clear and consistent meaning group co-occurring with the term in question. Further investigation has revealed that a significant proportion of the cooccurring items belong to the lexical field which refers to the idea of limiting or restricting. In this particular case, the lexical items refer to the imposition of constraints on the scope of discovery. These lexical items cannot be subsumed under any specific formal and fixed structure. Instead, they represent diverse grammatical categories used in various syntactic positions. Apart from the shared lexical field, the only other common feature is related to the subject or agent, which is a legal or natural person involved in the legal proceedings. In order to account for this regularity, I found it useful to invoke certain concepts proposed within the realm of corpus linguistics, such as semantic preference and semantic sequence. The connotational significance of the analysed term bears also some resemblance to the concept of semantic prosody (Sinclair, 1991) in the sense of spreading of connotational colouring beyond single word boundaries. It seems clear that a legal term should not be considered in isolation, without taking into account its immediate co-texts as well the wider context of the type of legal discourse in which it is found and the discourse community which uses it. This type of knowledge could be extremely useful for legal translators or any other LSP user and it should be integrated into modern terminographical resources. References Biber, D., S. Johansson, G. Leech, S. Conrad and E. Finegan (1999), The Longman Grammar of Spoken and Written English. London: Longman. Black, H.L. & Black, H.C. (1990), Black’s Law Dictionary with Pronunciations. New York: West Publishing Company. Bowker, L. (1996), “Towards a Corpus-Based Approach to Terminography”. Terminology 3.1: 27-52. Cook, G. (1992), The Discourse of Advertising. London: Routledge. Hunston, S. (2008), “Starting with the small words. Patterns, lexis and semantic sequences”, International Journal of Corpus Linguistics 13:3, 271-295. Hunston, S. and Francis, G. (2000), Pattern Grammar: a Corpus-driven Approach to the Lexical Grammar of English. Amsterdam, Benjamins. Leech, G. (1974), Semantics. Harmondsworth: Penguin. 306 STANISŁAW GOŹDŹ-ROSZKOWSKI Lee, D.S., C. Hall, and M. Hurley (1999), American Legal English. Using Language in Legal Contexts. Ann Arbor: The University of Michigan Press. Lyons, J. (1977), Semantics. Cambridge: Cambridge University Press. Partington, A. (1998), Patterns and Meanings: using corpora for English language research and teaching. Amsterdam: Benjamins. Renouf, A. and J. Sinclair (1991),“Collocational frameworks in English“. In: Aijmer K, Altenberg B. (eds.) English corpus linguistics: studies in honour of Jan Svartvik. London: Longman. Schulze, R. & U. Römer (2008), “Patterns, meaningful units and specialized discourses”. In Intenational Journal of Corpus Linguistics 13:3, 265 – 270. Sinclair, J. (1991), Corpus, concordance, collocation. Oxford, Oxford University Press. Stubbs, M. (2009), “Technology and phraseology. With notes on the history of corpus linguistics”. In: Ute Römer and Rainer Schulze (eds.) Exploring the Lexis-Grammar Interface. John Benjamins Publishing Company: Amsterdam/Philadelphia. Temmerman, R. (2000), Towards New Ways of Terminology Description. The Sociocognitive Approach. John Benjamins Publishing Company: Amsterdam/Philadelphia. Wray, A. (2002), Formulaic Language and the Lexicon. Cambridge: Cambridge University Press. Wüster, E. (1991), Einfürung in die allgemeine Terminologielehre und terminologische Lexikographie. 3. Bonn: Romanistischer Verlag. TRANSLATION AND MEANING, PART 10, 2013, 307 - 315 307 THE CHALLENGE OF MEANING CONSTRUCTION: EXPERT VS. NON-EXPERT PROCESSING Ewa Kościałkowska-Okońska Nicholaus Copernicus University, Toruń, Poland Abstract: The key function and task of translation in general is the transfer of meaning. It shall be analysed with a view to differences in processing characteristics that occur between expert and non-expert translators. Despite the fact that for both experts and non-experts the transfer of meaning in translation tasks is the highest priority, the way they perceive it obviously substantially varies. Those differences result in the emergence of two distinct types of competence, with each of them based on different criteria and different domains. Translation competence as such shall not be the focus of our considerations, yet for the purpose of this article a generally accepted definition of translation competence is postulated which assumes the ability to express the source language message in the target language on all language levels, also including cultural elements present in the text or utterance (and extralinguistic items). This transfer is expected to be accurate and reflect the complexity of the original. Therefore, competence may be approached as a system of knowledge (both declarative for storing patterns and rules, and procedural for storing rules, procedures and strategies; cf Sternberg 1999), mechanisms of information processing, decision-taking, problem-solving and other cognitive factors such as memory or intelligence; moreover, culture and communication are equally important for translation performance. A plethora of approaches towards competence in translation is distinguished. All of them can be somewhat arbitrarily divided into categories that rely on concepts essential for the operation of competence.The concepts of expert and non-expert competence are briefly addressed, with the focus being laid on the role of motivation, experience and creativity as factors central to meaning construction and transfer. 1. Meaning construction Comprehension in translation generally entails the formulation of a propositional meaning by the translator. The meaning is represented in a source language text item being mentally processed within the working memory (and with the simultaneous access to the long-term memory; for memoryrelated concept see e.g. Baddeley 1990, Sternberg, 1999). One of major theories applied to processing meaning is the propositional hypothesis (see Anderson and Bower, 1973) that mostly refers to implied or referential meaning. It might be therefore inferred that the text comprehension depends on such factors as the (quick) access to word meaning, with the help of either the memory store or context-related information, employing propositions, abstracting relevant information from the text or constructing mental models. The success of translation performance lies in the transfer of meaning between the realities of the source and target languages. This (initial) relation between translation and the transfer of meaning perceived as moving (semiotic, linguistic) material from one place and time to another place and time seems to be derived from the claim saying that “if there were no material transfer, it texts were not moved across time and space, there would be no translation” (Pym, 1992: 17). The production of the target text is thus a search for words rendering the meaning “activated by the interplay of potential word meaning and context determination” (Kussmaul, 1995: 105). Identifying meaning of the source language text involves identifying linguistic items and their context-relation. In order to be able to do it, the translator should be capable of assessing the source language items and structures under scrutiny with his or her own assumptions or expectations concerning the potentiality of cultural and situational functions being performed by these items or structures. This capability is manifested owing to the occurrence and automatic activation of extralinguistic knowledge. Understanding meaning is based on, as it was said above, the construction of (tentative) mental models (see Johnson-Laird, 1983, 1989) that are established following the textual input in the mind of the translator. In translation “meaning is constructed through the componentiality of the different linguistic 308 EWA KOŚCIAŁKOWSKA-OKOŃSKA structures whether or not they are seen as a continuum (lexicon, morphology, syntax, semantics) inasmuch as each contributes to meaning” (Dancette, 1997: 79). The text consisting of a variety of elements is first conceptualised and then processed on the basis of its cognitive representation and in a given social and cultural situation. Not understanding the text results from the inability to understand its meaning in an appropriate way, including its cultural and social implications that lie beyond the scope of language and syntax. Communicative actions occurring among participants of a given translation situation integrate the meaning derived from (literal) representation of the text (in the form of lingu